Print

Print


Thank you...
Re question 2, the shell script tbss_2_reg can take the command line flag, -f target.hdr (user chosen target) to do N comparisons instead of NxN. But I think you end up with a skeleton with N-1 timepoints.
-Marc

On 5/24/07, Steve Smith <[log in to unmask]> wrote:
On 24 May 2007, at 11:01, Marc Dubin wrote:

> Two other quick questions...
>
> 1. I would look at tbss_tstat2 in this case and tbss_tstat1 would
> be unimportant?

It depends on your contrasts. If contrast 2 is asking about
correlation with EV2 then yes that's the one of interest.

> 2. If I passed a -f flag for pre-chosen target to tbss_2_reg then
> the target will not be included in all_FA_skeleton (reducing my n
> by 1), right?

Sorry - I can't tell what you're asking - tbss_2_reg doesn't take any
arguments?

Steve.


>
> Best,
> Marc
>
> On 5/23/07, Steve Smith < [log in to unmask]> wrote:Hi - yes,
> setup the model with Glm (or Glm_gui) and make the first EV
> constant and the second the demeaned subject scores (see the FEAT
> manual, higher-level example setups for exactly this example).
>
> Cheers.
>
>
> On 23 May 2007, at 21:53, Marc Dubin wrote:
>
> > Hello all,
> >
> > I am new to TBSS.
> >
> > I have a 4D all_FA_skeletonised file
> > containing 10 subjects (time points) and want to see
> > while voxels are linearly correlated with performnce on
> > a cognitive test (with integer score). Does anyone know
> > which function to use?
> >
> > Thanks!
> > Marc
>
>
> ----------------------------------------------------------------------
> --
> ---
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director,  Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
> +44 (0) 1865 222726  (fax 222717)
> [log in to unmask]    http://www.fmrib.ox.ac.uk/ ~steve
> ----------------------------------------------------------------------
> --
> ---
>
>
>
> --
> 646 831 8886
> [log in to unmask]


------------------------------------------------------------------------

---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---



--
646 831 8886
[log in to unmask]
------=_Part_42246_12958662.1180028403279-- ========================================================================= Date: Thu, 24 May 2007 10:43:21 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Alphabetical listing in skeleton file In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_42342_15552776.1180028601663" ------=_Part_42342_15552776.1180028601663 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks, Ged. This is very helpful. It sounds like as my files are currently in the form "filex" where x is a precisely 5 digit number, that I should not have to do any conversion at the moment, but your script may come in handy at some point.Thanks again, Marc On 5/24/07, Ged Ridgway <[log in to unmask]> wrote: > > Hi Marc, > > You might find that you need to zero-pad filenames. E.g. if you have > sub1, sub2, ..., sub10, sub11, ... you'll find that sub1 sorts after > sub19, etc. Whereas if the files are like sub001, sub002, ..., sub010, > ... then things should be fine. > > You can script something up to rename files using printf, based on the > following example. > > Hope this helps, > Ged. > > #!/bin/sh > # Zero-pad files, e.g. sub1.nii -> sub001.nii > pre=sub > post='.nii' > N=1 > M=20 > for (( i=$N; i <= $M; i=i+1 )) > do > out=`printf "%s%03d%s" $pre $i $post` > mv ${pre}${i}$post $out > done > > > Marc Dubin wrote: > > Does anyone know how the alphabetised listing in the 4D all_FA_skeleton > file extends to numbers > > in the filenames. For instance, if three subjects went into the analysis > as sub11.nii.gz, sub12.nii.gz > > and sub13.nii.gz would these be the 1st, 2nd and 3rd in the skeleton > file? > > > -- 646 831 8886 [log in to unmask] ------=_Part_42342_15552776.1180028601663 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks, Ged. This is very helpful. It sounds like as my files are currently in the form "filex" where x is a precisely 5 digit number, that I should not have to do any conversion at the moment, but your script may come in handy at some point.
Thanks again,
Marc

On 5/24/07, Ged Ridgway <[log in to unmask]> wrote:
Hi Marc,

You might find that you need to zero-pad filenames. E.g. if you have
sub1, sub2, ..., sub10, sub11, ... you'll find that sub1 sorts after
sub19, etc. Whereas if the files are like sub001, sub002, ..., sub010,
... then things should be fine.

You can script something up to rename files using printf, based on the
following example.

Hope this helps,
Ged.

#!/bin/sh
# Zero-pad files, e.g. sub1.nii -> sub001.nii
pre=sub
post='.nii'
N=1
M=20
for (( i=$N; i <= $M; i=i+1 ))
do
    out=`printf "%s%03d%s" $pre $i $post`
    mv ${pre}${i}$post $out
done


Marc Dubin wrote:
> Does anyone know how the alphabetised listing in the 4D all_FA_skeleton file extends to numbers
> in the filenames. For instance, if three subjects went into the analysis as sub11.nii.gz, sub12.nii.gz
> and sub13.nii.gz would these be the 1st, 2nd and 3rd in the skeleton file?
>



--
646 831 8886
[log in to unmask]
------=_Part_42342_15552776.1180028601663-- ========================================================================= Date: Thu, 24 May 2007 10:46:01 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Alphabetical listing in skeleton file In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_42406_23356072.1180028761390" ------=_Part_42406_23356072.1180028761390 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks Steve, this helps! On 5/24/07, Steve Smith <[log in to unmask]> wrote: > > If you type something like > > cd FAi > echo *FAi.hdr > > you will see what order the list gets expanded in. > > > > On 24 May 2007, at 11:44, Marc Dubin wrote: > > > Does anyone know how the alphabetised listing in the 4D > > all_FA_skeleton file extends to numbers > > in the filenames. For instance, if three subjects went into the > > analysis as sub11.nii.gz, sub12.nii.gz > > and sub13.nii.gz would these be the 1st, 2nd and 3rd in the > > skeleton file? > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > -- 646 831 8886 [log in to unmask] ------=_Part_42406_23356072.1180028761390 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks Steve, this helps!

On 5/24/07, Steve Smith <[log in to unmask]> wrote:
If you type something like

cd FAi
echo *FAi.hdr

you will see what order the list gets expanded in.



On 24 May 2007, at 11:44, Marc Dubin wrote:

> Does anyone know how the alphabetised listing in the 4D
> all_FA_skeleton file extends to numbers
> in the filenames. For instance, if three subjects went into the
> analysis as sub11.nii.gz, sub12.nii.gz
> and sub13.nii.gz would these be the 1st, 2nd and 3rd in the
> skeleton file?


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/ ~steve
------------------------------------------------------------------------
---



--
646 831 8886
[log in to unmask] ------=_Part_42406_23356072.1180028761390-- ========================================================================= Date: Thu, 24 May 2007 19:56:04 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: TBSS xgrid port? Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain I just got started on FSL/TBSS for analyzing DTI data and I'm finding tha= t the spatial registration=20 step (tbss_2_reg) can take 2-3 days to run on my intel-64 mac if I do the= NxN analysis. Does=20 anyone knwo of an FSL/TBSS port to run on an xgrid? Would this be difficu= lt to set up? Best, Marc ========================================================================= Date: Thu, 24 May 2007 20:30:27 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Daniel Peterson <[log in to unmask]> Subject: ROI Drawing Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="ISO-8859-1" Just curious- what program do you guys use for drawing ROIs and defining binary masks, like for probtrack for example.=20 Thanks -Dan Peterson ========================================================================= Date: Thu, 24 May 2007 15:46:37 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Matt <[log in to unmask]> Organization: ma-tea.com Subject: Re: ROI Drawing In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit FSLView is what most people use. Peace, Matt. -----Original Message----- From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Daniel Peterson Sent: Thursday, May 24, 2007 3:30 PM To: [log in to unmask] Subject: [FSL] ROI Drawing Just curious- what program do you guys use for drawing ROIs and defining binary masks, like for probtrack for example. Thanks -Dan Peterson ========================================================================= Date: Thu, 24 May 2007 21:33:07 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Peter Smale <[log in to unmask]> Subject: Principal diffusion directions Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="ISO-8859-1" Hi all, Our Siemens scanner has recently undergone a software upgrade. Now the images we get raw from the scanner are unaveraged mosaics: 52 "super-slic= es" for a 12 direction + 1 b0 scan. They used to be averaged non-mosaic volum= es. With each of the new mosaic datasets processed with FSL, I am having the same problem: The V1 vector field looks very strange. I tried to attach screen grabs from fslview, but the email became too big for the list. Anyway, in axial slices, the eigenvector field is asymmetric, so the cor= pus callosum is half red and half green. The CSTs in the coronal slice are a more washed-out blue than they should be.= Clearly this is not a matter of flipping eigenvector components, and ther= e is no possibility of any reasonable tractography in this data. I am at a loss as to why it should be like this. The strange thing is tha= t it has only happened since the software upgrade, with the unaveraged mosa= ic format scans - I got good results from the old format. I am reasonably confident about my mosaic averaging method. Any suggestions will be much appreciated. Regards, Peter ========================================================================= Date: Fri, 25 May 2007 07:46:05 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: TBSS xgrid port? In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi, for the next release we are hoping to supply script-level parallelisation for some programs such as tbss_2_reg using SGE. We're aiming for lateish summer for the release. Cheers. On 24 May 2007, at 20:56, Marc Dubin wrote: > I just got started on FSL/TBSS for analyzing DTI data and I'm > finding that the spatial registration > step (tbss_2_reg) can take 2-3 days to run on my intel-64 mac if I > do the NxN analysis. Does > anyone knwo of an FSL/TBSS port to run on an xgrid? Would this be > difficult to set up? > Best, > Marc ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Fri, 25 May 2007 07:48:12 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Voxel correlations with cognitive score In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi Marc, On 24 May 2007, at 19:40, Marc Dubin wrote: > Thank you... > Re question 2, the shell script tbss_2_reg can take the command > line flag, -f target.hdr (user chosen target) No it can't! :) I assume you're referring to the line: if [ -f target.hdr ] ; then This isn't a command-line option, it is an automatic test to see if you have put target.hdr in place in FAi and if so use that instead of the NxN registrations. Hope that makes sense? Cheers, Steve. > to do N comparisons instead of NxN. But I think you end up with a > skeleton with N-1 timepoints. > -Marc > > On 5/24/07, Steve Smith <[log in to unmask]> wrote: On 24 May > 2007, at 11:01, Marc Dubin wrote: > > > Two other quick questions... > > > > 1. I would look at tbss_tstat2 in this case and tbss_tstat1 would > > be unimportant? > > It depends on your contrasts. If contrast 2 is asking about > correlation with EV2 then yes that's the one of interest. > > > 2. If I passed a -f flag for pre-chosen target to tbss_2_reg then > > the target will not be included in all_FA_skeleton (reducing my n > > by 1), right? > > Sorry - I can't tell what you're asking - tbss_2_reg doesn't take any > arguments? > > Steve. > > > > > > Best, > > Marc > > > > On 5/23/07, Steve Smith < [log in to unmask]> wrote:Hi - yes, > > setup the model with Glm (or Glm_gui) and make the first EV > > constant and the second the demeaned subject scores (see the FEAT > > manual, higher-level example setups for exactly this example). > > > > Cheers. > > > > > > On 23 May 2007, at 21:53, Marc Dubin wrote: > > > > > Hello all, > > > > > > I am new to TBSS. > > > > > > I have a 4D all_FA_skeletonised file > > > containing 10 subjects (time points) and want to see > > > while voxels are linearly correlated with performnce on > > > a cognitive test (with integer score). Does anyone know > > > which function to use? > > > > > > Thanks! > > > Marc > > > > > > > ---------------------------------------------------------------------- > > -- > > --- > > Stephen M. Smith, Professor of Biomedical Engineering > > Associate Director, Oxford University FMRIB Centre > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > +44 (0) 1865 222726 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/ ~steve > > > ---------------------------------------------------------------------- > > -- > > --- > > > > > > > > -- > > 646 831 8886 > > [log in to unmask] > > > ---------------------------------------------------------------------- > -- > > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ---------------------------------------------------------------------- > -- > --- > > > > -- > 646 831 8886 > [log in to unmask] ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Fri, 25 May 2007 01:55:30 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Voxel correlations with cognitive score In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_46750_1662271.1180072530009" ------=_Part_46750_1662271.1180072530009 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Yes, that makes perfect sense.I have been including the -f, but I guess it has been redundant. Parallelisation at the end of the summer would be very nice. BTW, I'm thinking of signing up for the 9/07 course. Best, Marc On 5/25/07, Steve Smith <[log in to unmask]> wrote: > > Hi Marc, > > On 24 May 2007, at 19:40, Marc Dubin wrote: > > > Thank you... > > Re question 2, the shell script tbss_2_reg can take the command > > line flag, -f target.hdr (user chosen target) > > No it can't! :) > > I assume you're referring to the line: > if [ -f target.hdr ] ; then > > This isn't a command-line option, it is an automatic test to see if > you have put target.hdr in place in FAi and if so use that instead of > the NxN registrations. > > Hope that makes sense? > > Cheers, Steve. > > > > > to do N comparisons instead of NxN. But I think you end up with a > > skeleton with N-1 timepoints. > > -Marc > > > > On 5/24/07, Steve Smith <[log in to unmask]> wrote: On 24 May > > 2007, at 11:01, Marc Dubin wrote: > > > > > Two other quick questions... > > > > > > 1. I would look at tbss_tstat2 in this case and tbss_tstat1 would > > > be unimportant? > > > > It depends on your contrasts. If contrast 2 is asking about > > correlation with EV2 then yes that's the one of interest. > > > > > 2. If I passed a -f flag for pre-chosen target to tbss_2_reg then > > > the target will not be included in all_FA_skeleton (reducing my n > > > by 1), right? > > > > Sorry - I can't tell what you're asking - tbss_2_reg doesn't take any > > arguments? > > > > Steve. > > > > > > > > > > Best, > > > Marc > > > > > > On 5/23/07, Steve Smith < [log in to unmask] > wrote:Hi - yes, > > > setup the model with Glm (or Glm_gui) and make the first EV > > > constant and the second the demeaned subject scores (see the FEAT > > > manual, higher-level example setups for exactly this example). > > > > > > Cheers. > > > > > > > > > On 23 May 2007, at 21:53, Marc Dubin wrote: > > > > > > > Hello all, > > > > > > > > I am new to TBSS. > > > > > > > > I have a 4D all_FA_skeletonised file > > > > containing 10 subjects (time points) and want to see > > > > while voxels are linearly correlated with performnce on > > > > a cognitive test (with integer score). Does anyone know > > > > which function to use? > > > > > > > > Thanks! > > > > Marc > > > > > > > > > > > ---------------------------------------------------------------------- > > > -- > > > --- > > > Stephen M. Smith, Professor of Biomedical Engineering > > > Associate Director, Oxford University FMRIB Centre > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > +44 (0) 1865 222726 (fax 222717) > > > [log in to unmask] http://www.fmrib.ox.ac.uk/ ~steve > > > > > ------------------------------ ---------------------------------------- > > > -- > > > --- > > > > > > > > > > > > -- > > > 646 831 8886 > > > [log in to unmask] > > > > > > ---------------------------------------------------------------------- > > -- > > > > --- > > Stephen M. Smith, Professor of Biomedical Engineering > > Associate Director, Oxford University FMRIB Centre > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > +44 (0) 1865 222726 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > ---------------------------------------------------------------------- > > -- > > --- > > > > > > > > -- > > 646 831 8886 > > [log in to unmask] > > > ------------------------------------------------------------------------ > > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/ ~steve > ------------------------------------------------------------------------ > --- > -- 646 831 8886 [log in to unmask] ------=_Part_46750_1662271.1180072530009 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Yes, that makes perfect sense.
I have been including the -f, but I guess
it has been redundant.
Parallelisation at the end of the summer
would be very nice. 
BTW, I'm thinking of signing up for the 
9/07 course.
Best,
Marc 

On 5/25/07, Steve Smith <[log in to unmask]> wrote:
Hi Marc,

On 24 May 2007, at 19:40, Marc Dubin wrote:

> Thank you...
> Re question 2, the shell script tbss_2_reg can take the command
> line flag, -f target.hdr (user chosen target)

No it can't!   :)

I assume you're referring to the line:
if [ -f target.hdr ] ; then

This isn't a command-line option, it is an automatic test to see if
you have put target.hdr in place in FAi and if so use that instead of
the NxN registrations.

Hope that makes sense?

Cheers, Steve.



> to do N comparisons instead of NxN. But I think you end up with a
> skeleton with N-1 timepoints.
> -Marc
>
> On 5/24/07, Steve Smith <[log in to unmask]> wrote: On 24 May
> 2007, at 11:01, Marc Dubin wrote:
>
> > Two other quick questions...
> >
> > 1. I would look at tbss_tstat2 in this case and tbss_tstat1 would
> > be unimportant?
>
> It depends on your contrasts. If contrast 2 is asking about
> correlation with EV2 then yes that's the one of interest.
>
> > 2. If I passed a -f flag for pre-chosen target to tbss_2_reg then
> > the target will not be included in all_FA_skeleton (reducing my n
> > by 1), right?
>
> Sorry - I can't tell what you're asking - tbss_2_reg doesn't take any
> arguments?
>
> Steve.
>
>
> >
> > Best,
> > Marc
> >
> > On 5/23/07, Steve Smith < [log in to unmask] > wrote:Hi - yes,
> > setup the model with Glm (or Glm_gui) and make the first EV
> > constant and the second the demeaned subject scores (see the FEAT
> > manual, higher-level example setups for exactly this example).
> >
> > Cheers.
> >
> >
> > On 23 May 2007, at 21:53, Marc Dubin wrote:
> >
> > > Hello all,
> > >
> > > I am new to TBSS.
> > >
> > > I have a 4D all_FA_skeletonised file
> > > containing 10 subjects (time points) and want to see
> > > while voxels are linearly correlated with performnce on
> > > a cognitive test (with integer score). Does anyone know
> > > which function to use?
> > >
> > > Thanks!
> > > Marc
> >
> >
> >
> ----------------------------------------------------------------------
> > --
> > ---
> > Stephen M. Smith, Professor of Biomedical Engineering
> > Associate Director,  Oxford University FMRIB Centre
> >
> > FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
> > +44 (0) 1865 222726  (fax 222717)
> > [log in to unmask]     http://www.fmrib.ox.ac.uk/ ~steve
> >
> ------------------------------ ----------------------------------------
> > --
> > ---
> >
> >
> >
> > --
> > 646 831 8886
> > [log in to unmask]
>
>
> ----------------------------------------------------------------------
> --
>
> ---
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director,  Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
> +44 (0) 1865 222726  (fax 222717)
> [log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
> ----------------------------------------------------------------------
> --
> ---
>
>
>
> --
> 646 831 8886
> [log in to unmask]


------------------------------------------------------------------------

---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/ ~steve
------------------------------------------------------------------------
---



--
646 831 8886
[log in to unmask]
------=_Part_46750_1662271.1180072530009-- ========================================================================= Date: Fri, 25 May 2007 07:22:25 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Tim Behrens <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.2) Content-Type: multipart/alternative; boundary=Apple-Mail-12--96517370 --Apple-Mail-12--96517370 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Hi - if you run bedpost_datacheck on your directory it might tell you what's wrong - for example make sure the number of entriess in bvals and bvecs is the same as the number of volumes in your data. T On 24 May 2007, at 00:17, David Kideckel wrote: > hi. thank you for your support. once i run bet from the command > line (using cygwin), it does not seem to be working properly. but > when i use the gui, it does... > anyway, i am trying to rum dtifit from the command line and get the > following message: > 0 256 0 256 0 50 > 0 slices processed > Aborted > > Any idea why this is happening? Did I enter something wrong? > > Thanks. > David > > On 5/23/07, Tim Behrens <[log in to unmask]> wrote: > If you select dtifit from the drop down menu on the same page, all` > the information about file specifications is there. Don't forget to > end your bvecs and bvals files with carriage returns. > > Cheers > > T > > On 23 May 2007, at 18:53, David Kideckel wrote: > >> Hi. That was one of the firs things I did do. I will try to be >> more specific with my question. I want to do TBSS, and no >> tractography. So, would I do the following in this order: >> Eddy Correct >> BET >> DTI Fit >> >> If so, where is it in the manual specified how to create and input >> the bvecs and bvals files? I follow the link it gives in the FAQ >> section, but I can't seem to find this info. >> >> Thanks again for your help. >> David >> >> On 5/23/07, Tim Behrens < [log in to unmask]> wrote: >> Hi - if you go to >> http://www.fmrib.ox.ac.uk/fsl/fdt/index.html >> >> and select "processing pipeline", all should be clear. >> >> T >> >> On 23 May 2007, at 17:22, David Kideckel wrote: >> >>> Hi. Thank you for your response. Can you please send me this >>> information. I can't seem to find it on the FSL website. >>> Thanks you. >>> David >>> >>> On 5/23/07, Steve Smith < [log in to unmask]> wrote: >>> Hi, >>> >>> Indeed - you should run eddy_correct and then bet and then dtifit to >>> produce the FA image - see the FDT manual pages for more details. >>> Sorry the TBSS page isn't clearer about that - we'll fix that >>> shortly. >>> >>> Cheers. >>> >>> >>> >>> On 22 May 2007, at 18:51, David Kideckel wrote: >>> >>> > Dear FSLers, >>> > >>> > I am new to TBSS and am wondering what the preprocessing steps are >>> > to do prior to TBSS. This information seems to be different on the >>> > FAQ section of the FSL website and the technical report on TBSS. >>> > >>> > Is it correct to assume that to compare 2 groups with diffusion >>> > data, I would need to run: >>> > Eddy correct >>> > Registration >>> > DTIFit >>> > TBSS >>> > >>> > I am unclear as to whether I need to run BET (brain extraction >>> > tool) and/or BEDPOST. >>> > >>> > Any guidance would be much appreciated. >>> > >>> > David Kideckel, PhD Candidate >>> >>> >>> -------------------------------------------------------------------- >>> ---- >>> --- >>> Stephen M. Smith, Professor of Biomedical Engineering >>> Associate Director, Oxford University FMRIB Centre >>> >>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >>> +44 (0) 1865 222726 (fax 222717) >>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >>> -------------------------------------------------------------------- >>> ---- >>> --- >>> >> >> > > --Apple-Mail-12--96517370 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=ISO-8859-1 Hi - if you run = bedpost_datacheck on your directory it might tell you what's wrong = -=A0
for example make sure the number of entriess in bvals and bvecs = is the same as the number of volumes in your data.

T
On 24 May = 2007, at 00:17, David Kideckel wrote:

hi. thank = you for your support. once i run bet from the command line (using = cygwin), it does not seem to be working properly. but when i use the = gui, it does...
anyway, i am trying to rum dtifit from the command = line and get the following message:
0 256 0 256 0 50
0 slices = processed
Aborted <core dumped>

Any idea why this is = happening? Did I enter something = wrong?

Thanks.
David

On = 5/23/07, Tim Behrens <[log in to unmask]> = wrote:
If you select dtifit from the drop down menu on = the same page, all` the information about file specifications is there. = Don't forget to end your bvecs and bvals files with carriage = returns.

Cheers

T

On 23 May 2007, at = 18:53, David Kideckel wrote:

Hi. That = was one of the firs things I did do. I will try to be more specific with = my question. I want to do TBSS, and no tractography. So, would I do the = following in this order:
Eddy Correct
BET
DTI Fit

If = so, where is it in the manual specified how to create and input the = bvecs and bvals files? I follow the link it gives in the FAQ section, = but I can't seem to find this info.

Thanks again for your = help.
David

On 5/23/07, Tim Behrens < [log in to unmask]> = wrote:
Hi - if you go to=A0
= http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be = clear.

T
=

On 23 May 2007, at 17:22, David Kideckel wrote: =

Hi. Thank you for your response. Can = you please send me this information. I can't seem to find it on the FSL = website.
Thanks you.
David

On 5/23/07, Steve = Smith < = [log in to unmask]> wrote:
Hi,

Indeed = - you should run eddy_correct and then bet and then dtifit to =
produce the FA image - see the FDT manual pages for more = details.
Sorry the TBSS page isn't clearer about that - we'll fix = that shortly.

Cheers.



On 22 May 2007, at 18:51, = David Kideckel wrote:

> Dear FSLers,
>
> I am new = to TBSS and am wondering what the preprocessing steps are
> to do = prior to TBSS. This information seems to be different on the
> = FAQ section of the FSL website and the technical report on TBSS. =
>
> Is it correct to assume that to compare 2 groups with = diffusion
> data, I would need to run:
> Eddy = correct
> Registration
> DTIFit
> = TBSS
>
> I am unclear as to whether I need to run BET (brain = extraction
> tool) and/or BEDPOST.
>
> Any guidance = would be much appreciated.
>
> David Kideckel, PhD Candidate =


--------------------------------------------------------------= ----------
---
Stephen M. Smith, Professor of Biomedical = Engineering
Associate Director,=A0=A0Oxford University FMRIB = Centre

FMRIB, JR Hospital, Headington, Oxford=A0=A0OX3 9DU, UK =
+44 (0) 1865 222726=A0=A0(fax 222717)
[log in to unmask]=A0=A0=A0=A0= = http://www.fmrib.ox.ac.uk/~steve
----------------------------------= -------------------------------------- =
---



=



= --Apple-Mail-12--96517370-- ========================================================================= Date: Fri, 25 May 2007 09:46:46 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: FEEDS data -paradigm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Dear all, I'm quite curious which paradigm was used for the FEEDS example data set, ie. the visual and auditory stimuli. Are they described anywhere? Thanks Alle Meije ========================================================================= Date: Fri, 25 May 2007 09:52:41 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: FEEDS data -paradigm In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi Alle Meije it's 30s on/off blocks for visual and 45s on/off for auditory - oth starting with on I think. cheers christian On 25 May 2007, at 09:46, Alle Meije Wink wrote: > Dear all, > > I'm quite curious which paradigm was used for the FEEDS example > data set, ie. the visual and auditory stimuli. Are they described > anywhere? > > Thanks > Alle Meije ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Fri, 25 May 2007 09:56:49 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: Re: FEEDS data -paradigm In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Hi Christian True, but that information is also in the descripition of the FEAT part of http://fsl.fmrib.ox.ac.uk/fsl/feeds/doc/#examples Is there a page/paper that describes the stimuli though (eg. checkered screen for visual, 100 Hz beep for auditory)? Thanks Alle Meije Christian Beckmann wrote: > Hi Alle Meije > > it's 30s on/off blocks for visual and 45s on/off for auditory - oth > starting with on I think. > cheers > christian ========================================================================= Date: Fri, 25 May 2007 10:29:48 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: FEEDS data -paradigm In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi it's a blue/yellow checkerboard reversing at 8H and listening to a boring radio show (about mortgages if I remember correctly) cheers Christian On 25 May 2007, at 09:56, Alle Meije Wink wrote: > Hi Christian > > True, but that information is also in the descripition of the FEAT > part of http://fsl.fmrib.ox.ac.uk/fsl/feeds/doc/#examples > > Is there a page/paper that describes the stimuli though (eg. > checkered screen for visual, 100 Hz beep for auditory)? > > Thanks > Alle Meije > > Christian Beckmann wrote: >> Hi Alle Meije >> it's 30s on/off blocks for visual and 45s on/off for auditory - >> oth starting with on I think. >> cheers >> christian ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Fri, 25 May 2007 12:23:12 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christina Hugenschmidt <[log in to unmask]> Subject: melodic ICA-denoising, hpf, task activity Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" Hi! We are enjoying exploring ICA with melodic, and as you all, especially=20= Christian, have been most helpful in answering our questions, I was hopin= g=20 you might be kind enough to answer a few more. First, where can you find information about what high pass filter was use= d=20 in an analysis? I (shamefully) did not write down the high pass filter us= ed=20 in a couple of analyses, and did not see it in the melodic log. Also, we have a subject whose data did not converge. In reading posts abo= ut=20 this, I saw the suggestion that the data be denoised using melodic. The=20= post mentioned the following link=20 (http://www.fmrib.ox.ac.uk/fslcourse/lectures/melodic/=20 _3200_denoise.fpd/index.html) for more information about denoising. I cou= ld=20 not follow the link, so I looked in the melodic lecture and found mention= =20 of denoising, but I am still not sure exactly how to accomplish this. Is=20= there a paper that has more detailed methodology? Lastly, we have a small debate going on in our lab and I was wondering if= =20 you could weigh in. Is it appropriate to use ICA to find components when = a=20 task paradigm is used? Will there by any bias introduced by the fact that= =20 much of the signal and variance are driven by the task, such that non-tas= k=20 related activity might be more difficult to identify? Thank you for any insights you can offer:~) Christina ========================================================================= Date: Fri, 25 May 2007 12:56:59 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: =?UTF-8?Q?=E5=9B=9E=E5=A4=8D=EF=BC=9A?= Re: [FSL] BET in Melodic In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=GB2312; delsp=yes; format=flowed Content-Transfer-Encoding: quoted-printable Hi, when you use the GUI instead of the command line then betfunc is =20 already used to find the mask which is then being passed to the =20 melodic binary hope this helps christian On 24 May 2007, at 12:59, pengxu wei wrote: > Dear Christian, > > Thank you. > > It is introduced that betfunc uses a -f 0.3 for 4D fmri data while =20 > I found bet in Melodic uses a -f 0.4. Bet in melodic cut off the =20 > most top slice of my data but betfunc didn't. I think it is better =20 > for my data to use betfunc first then pass the mask made to melodic =20= > explicitly, as you said. > > Regards, > > wei > > Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA > Hi > > melodic uses a different approach to robustifying bet for 4D data. In > betfund, the inital mask is generated from the first image and then > dilated by 1 voxel in x,y and z before masking all volumes by this > new dilated mask. In melodic the bet outline is found by running bet > on the mean image across time. I have not evaluated the differences > explicitly but guess that they are small. Note that you can always > pass a mask to melodic explicitly by using the -m option > cheers > christian > > > > On 23 May 2007, at 15:49, pengxu wei wrote: > > > Dear list, > > > > I have read some message in the archive but no answer available. > > betfunc is better than routine bet for 4D fmri data, so whether bet > > in Melodic is betfunc(when using GUI or command line), or just > > routine bet? > > > > Regards, > > > > wei > > > > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC= 20M=B8=BD=BC=FE=A3=A1 > > ____ > Christian F. Beckmann > University Research Lecturer > Oxford University Centre for Functional MRI of the Brain (FMRIB) > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > tel: +44 1865 222551 fax: +44 1865 222717 > > > =D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE= =A3=A1 ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Fri, 25 May 2007 13:15:50 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: melodic ICA-denoising, hpf, task activity In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi Christina, melodic.log is ceated by the melodic binary, so all the pre- processing which is initiated by the melodic GUI is not recorded in there but in report.log. There should be a command line call to $ {FSLDIR}/bin/ip - the first number after the -t option relates to the highpass filter sigma (in volumes), i.e. if TR is 3 sec and you're using a 100s hp filter then this number is 100/2*TR = 16.667. Denoising is straight forward. First, run melodic on the original data (filterd_func_data.nii.gz, say). This gives an .ica directory (blah.ica, say). Go throught the components in there and note down the numbers of the components which you are certain (!) are structured noise. Then simply run melodic -i filtered_func_data --ICs=blah.ica/melodic_IC -- mix=blah.ica/melodic_mix.txt -o blah2.ica In blah2.ica you should then find a file melodic_ICAfiltered - which is the original data (filtered_func_data.nii.gz) where the structured noise effects have been removed using simple linear regression. Note that this results in a small loss in degrees of freedom which - if the original time series is reasonably long and if you have not removed too many components - is pretty irrelevant for the final GLM stats. We're working on integrating melodic denoising into feat, which will then take care of these things automagically, watch this space ;) Wrt your final question: task related signal changes are treated like any other form of signal change in the data. If these changes are structured variations then ICA will try to represent these components as components. If a task-paradigm is used then any induced variation has no more (but also no less) impact on the remaining components than any other component being estimated. In theory, the existence of an additional source will simply increase the estimate for the number of components so that the additional signal will be represented as an additional source estimate. hope this helps christian On 25 May 2007, at 12:23, Christina Hugenschmidt wrote: > Hi! > > We are enjoying exploring ICA with melodic, and as you all, especially > Christian, have been most helpful in answering our questions, I was > hoping > you might be kind enough to answer a few more. > > First, where can you find information about what high pass filter > was used > in an analysis? I (shamefully) did not write down the high pass > filter used > in a couple of analyses, and did not see it in the melodic log. > > Also, we have a subject whose data did not converge. In reading > posts about > this, I saw the suggestion that the data be denoised using melodic. > The > post mentioned the following link > (http://www.fmrib.ox.ac.uk/fslcourse/lectures/melodic/ > _3200_denoise.fpd/index.html) for more information about denoising. > I could > not follow the link, so I looked in the melodic lecture and found > mention > of denoising, but I am still not sure exactly how to accomplish > this. Is > there a paper that has more detailed methodology? > > Lastly, we have a small debate going on in our lab and I was > wondering if > you could weigh in. Is it appropriate to use ICA to find components > when a > task paradigm is used? Will there by any bias introduced by the > fact that > much of the signal and variance are driven by the task, such that > non-task > related activity might be more difficult to identify? > > Thank you for any insights you can offer:~) > > Christina ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Fri, 25 May 2007 14:03:11 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: avwmaths ODT MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Hi all, I converted a segmented image (with region labels 0 for background, 1 for cluster 1, etc) that was 16-bit int (short) to 8-bit int (char) with avwmaths++ segmented.nii segmentedBYTE.nii -odt char The segmented.nii has on enormous cluster in the parietal cortex, which has totally disappeared in the conversion. am I right in thinkning that in the conversion from 16-bit to 8-bit, all numbers are divided by 256? If not, what could be happening here? Cheers Alle Meije ========================================================================= Date: Fri, 25 May 2007 21:09:07 +0800 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: pengxu wei <[log in to unmask]> Subject: =?gb2312?Q?=BB=D8=B8=B4=A3=BA_Re:_[FSL]_=E5=9B=9E=E5=A4=8D=EF=BC=3F?= Re: [FSL] BET in Melodic In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-1149882233-1180098547=:29858" --0-1149882233-1180098547=:29858 Content-Type: text/plain; charset=gb2312 Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by bofur.jiscmail.ac.uk id l4PD9JWx021753 Dear Christian, =20 Thank you. but some further doubts induced from your reply... =20 (1) =20 Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA when you use the GUI instead of the command line then betfunc is=20 already used to find the mask which is then being passed to the=20 melodic binary but from http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=3Dind03&L=3DF= SL&P=3DR12990&D=3D0&I=3D-3&m=3D418=20 : ...thus melodic is always called with the --nobet option when using = the GUI. =20 --So I thought if using the GUI, melodic could not include bet. Is it=20 right? And in fact, when I use melodic GUI, the melodic.log always sho= w me --nobet, every time, whether or not I choose bet. =20 (2) =20 from http://www.fmrib.ox.ac.uk/fsl/bet/index.html:=20 betfunc COMMAND-LINE SCRIPT...the -f threshold is reduced to 0.3... =20 --After melodic analysis with command line function, I found in=20 melodic.log that: Calling BET: /usr/local/fsl/bin/bet /tmp/tcac.0=20 /tmp/tcac.0_brain -m -f 0.4 > /dev/null=20 =20 Since -f 0.4 is not 0.3, so I thought that in melodic, bet instead of b= etfunc is used. Is it right? =20 hope your clarification. =20 Regards, wei =20 Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA Hi, when you use the GUI instead of the command line then betfunc is=20 already used to find the mask which is then being passed to the=20 melodic binary hope this helps christian On 24 May 2007, at 12:59, pengxu wei wrote: > Dear Christian, > > Thank you. > > It is introduced that betfunc uses a -f 0.3 for 4D fmri data while=20 > I found bet in Melodic uses a -f 0.4. Bet in melodic cut off the=20 > most top slice of my data but betfunc didn't. I think it is better=20 > for my data to use betfunc first then pass the mask made to melodic=20 > explicitly, as you said. > > Regards, > > wei > > Christian Beckmann =D0=B4=B5=C0=A3=BA > Hi > > melodic uses a different approach to robustifying bet for 4D data. In > betfund, the inital mask is generated from the first image and then > dilated by 1 voxel in x,y and z before masking all volumes by this > new dilated mask. In melodic the bet outline is found by running bet > on the mean image across time. I have not evaluated the differences > explicitly but guess that they are small. Note that you can always > pass a mask to melodic explicitly by using the -m option > cheers > christian > > > > On 23 May 2007, at 15:49, pengxu wei wrote: > > > Dear list, > > > > I have read some message in the archive but no answer available. > > betfunc is better than routine bet for 4D fmri data, so whether bet > > in Melodic is betfunc(when using GUI or command line), or just > > routine bet? > > > > Regards, > > > > wei > > > > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC= 20M=B8=BD=BC=FE=A3=A1 > > ____ > Christian F. Beckmann > University Research Lecturer > Oxford University Centre for Functional MRI of the Brain (FMRIB) > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > tel: +44 1865 222551 fax: +44 1865 222717 > > > =D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE= =A3=A1 ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 =20 --------------------------------- =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M= =B8=BD=BC=FE=A3=A1=20 --0-1149882233-1180098547=:29858 Content-Type: text/html; charset=gb2312 Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by bofur.jiscmail.ac.uk id l4PD9JWx021753
Dear Christian,
 
Thank you. but some f= urther doubts induced from your reply...
 
(1)
 
Christian Beckman= n <[log in to unmask]> =D0=B4=B5=C0=A3= =BA
http://www.jiscmail.ac.uk/= cgi-bin/webadmin?A2=3Dind03&L=3DFSL&P=3DR12990&D=3D0&I=3D= -3&m=3D418
:  ...thus melodic is always= called with the --nobet option when using the GUI.
=20
 
--So I thought if using the GUI, melodic could n= ot include bet. Is it
right?  And in fact, when I use m= elodic GUI, the melodic.log always show me --nobet, every time, whether o= r not I choose bet.
 
(2)
 =
from http://www.fmrib.ox.ac.uk/fsl/bet/index.html:
betfunc COMMAND-L= INE SCRIPT...the -f threshold is reduced to 0.3...
=
 
--After melodic analysis with command line func= tion, I found in
melodic.log that: Calling BET: /usr/local/f= sl/bin/bet /tmp/tcac.0
/tmp/tcac.0_brain -m -f 0.4 > /dev/null
 
Since -f 0.4 is no= t 0.3, so I thought that in melodic, bet instead of betfunc is used.= Is it right?
 
hope your clarification.
 
=20
Regards,
wei

Christian Beckma= nn <[log in to unmask]> =D0=B4=B5=C0=A3=BA
Hi,

when you use the GUI instead of th= e command line then betfunc is
already used to find the mask which is= then being passed to the
melodic binary
hope this helps
christ= ian

On 24 May 2007, at 12:59, pengxu wei wrote:

> Dear C= hristian,
>
> Thank you.
>
> It is introduced tha= t betfunc uses a -f 0.3 for 4D fmri data while
> I found bet in Me= lodic uses a -f 0.4. Bet in melodic cut off the
> most top slice o= f my data but betfunc didn't. I think it is better
> for my data t= o use betfunc first then pass the mask made to melodic
> explicitl= y, as you said.
>
> Regards,
>
> wei
>
&= gt; Christian Beckmann <[log in to unmask]>=D0=B4=B5=C0=A3=BA
> Hi
>
> m= elodic uses a different approach to robustifying bet for 4D data. In
&= gt; betfund, the inital mask is generated from the first image and then> dilated by 1 voxel in x,y and z before masking all volumes by this=
> new dilated mask. In melodic the bet outline is found by running= bet
> on the mean image across time. I have not evaluated the diff= erences
> explicitly but guess that they are small. Note that you c= an always
> pass a mask to melodic explicitly by using the -m optio= n
> cheers
> christian
>
>
>
> On 23 = May 2007, at 15:49, pengxu wei wrote:
>
> > Dear list,
= > >
> > I have read some message in the archive but no ans= wer available.
> > betfunc is better than routine bet for 4D fmr= i data, so whether bet
> > in Melodic is betfunc(when using GUI = or command line), or just
> > routine bet?
> >
> > Regards,
> >
> > wei> >
> > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43= .5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE=A3=A1
>
> ____
> C= hristian F. Beckmann
> University Research Lecturer
> Oxford = University Centre for Functional MRI of the Brain (FMRIB)
> John Ra= dcliffe Hospital, Headington, Oxford OX3 9DU, UK.
> [log in to unmask] ox.ac.uk http://www.fmrib.ox.ac.uk/~beckmann
> tel: +44 1865 222551= fax: +44 1865 222717
>
>
> =D1=C5=BB=A2=C3=E2=B7=D1=D3= =CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE=A3=A1

____
Chris= tian F. Beckmann
University Research Lecturer
Oxford University Cen= tre for Functional MRI of the Brain (FMRIB)
John Radcliffe Hospital, H= eadington, Oxford OX3 9DU, UK.
[log in to unmask] http://www.fmri= b.ox.ac.uk/~beckmann
tel: +44 1865 222551 fax: +44 1865 222717


=C7= =C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8= =BD=BC=FE=A3=A1=20 --0-1149882233-1180098547=:29858-- ========================================================================= Date: Fri, 25 May 2007 21:29:51 +0800 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: pengxu wei <[log in to unmask]> Subject: =?gb2312?Q?=BB=D8=B8=B4=A3=BA_[FSL]_=BB=D8=B8=B4=A3=BA_Re:_[FSL]__=E5=9B=9E=E5=A4=8D=EF=BC=3F?= Re: [FSL] BET in Mel odic In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-720768613-1180099791=:4834" --0-720768613-1180099791=:4834 Content-Type: text/plain; charset=gb2312 Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by bifur.jiscmail.ac.uk id l4PDTrcK003832 Dear Christian, =20 I think I got it. Melodic GUI never use bet, instead, it uses betfunc i= n pre-processing step. Or, I can choose command line function to do some = 'liberal' analysis. =20 Thank you. =20 Regards, =20 wei pengxu wei <[log in to unmask]> =D0=B4=B5=C0=A3=BA Dear Christian, =20 Thank you. but some further doubts induced from your reply... =20 (1) =20 Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA when you use the GUI instead of the command line then betfunc is=20 already used to find the mask which is then being passed to the=20 melodic binary but from http://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=3Dind03&L=3DF= SL&P=3DR12990&D=3D0&I=3D-3&m=3D418=20 : ...thus melodic is always called with the --nobet option when using = the GUI. =20 --So I thought if using the GUI, melodic could not include bet. Is it=20 right? And in fact, when I use melodic GUI, the melodic.log always sho= w me --nobet, every time, whether or not I choose bet. =20 (2) =20 from http://www.fmrib.ox.ac.uk/fsl/bet/index.html:=20 betfunc COMMAND-LINE SCRIPT...the -f threshold is reduced to 0.3... =20 --After melodic analysis with command line function, I found in=20 melodic.log that: Calling BET: /usr/local/fsl/bin/bet /tmp/tcac.0=20 /tmp/tcac.0_brain -m -f 0.4 > /dev/null=20 =20 Since -f 0.4 is not 0.3, so I thought that in melodic, bet instead of b= etfunc is used. Is it right? =20 hope your clarification. =20 Regards, wei =20 Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA Hi, when you use the GUI instead of the command line then betfunc is=20 already used to find the mask which is then being passed to the=20 melodic binary hope this helps christian On 24 May 2007, at 12:59, pengxu wei wrote: > Dear Christian, > > Thank you. > > It is introduced that betfunc uses a -f 0.3 for 4D fmri data while=20 > I found bet in Melodic uses a -f 0.4. Bet in melodic cut off the=20 > most top slice of my data but betfunc didn't. I think it is better=20 > for my data to use betfunc first then pass the mask made to melodic=20 > explicitly, as you said. > > Regards, > > wei > > Christian Beckmann =D0=B4=B5=C0=A3=BA > Hi > > melodic uses a different approach to robustifying bet for 4D data. In > betfund, the inital mask is generated from the first image and then > dilated by 1 voxel in x,y and z before masking all volumes by this > new dilated mask. In melodic the bet outline is found by running bet > on the mean image across time. I have not evaluated the differences > explicitly but guess that they are small. Note that you can always > pass a mask to melodic explicitly by using the -m option > cheers > christian > > > > On 23 May 2007, at 15:49, pengxu wei wrote: > > > Dear list, > > > > I have read some message in the archive but no answer available. > > betfunc is better than routine bet for 4D fmri data, so whether bet > > in Melodic is betfunc(when using GUI or command line), or just > > routine bet? > > > > Regards, > > > > wei > > > > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC= 20M=B8=BD=BC=FE=A3=A1 > > ____ > Christian F. Beckmann > University Research Lecturer > Oxford University Centre for Functional MRI of the Brain (FMRIB) > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > tel: +44 1865 222551 fax: +44 1865 222717 > > > =D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE= =A3=A1 ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 =20 --------------------------------- =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC2= 0M=B8=BD=BC=FE=A3=A1=20 =20 --------------------------------- =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M= =B8=BD=BC=FE=A3=A1=20 --0-720768613-1180099791=:4834 Content-Type: text/html; charset=gb2312 Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by bifur.jiscmail.ac.uk id l4PDTrcK003832
Dear Christian,
 
I think I got it. Mel= odic GUI never use bet, instead, it uses betfunc in pre-processing step. = Or, I can choose command line function to do some 'liberal' analysis.
 
Thank you.
 
Reg= ards,
 
wei

pengxu wei <peng= [log in to unmask]> =D0=B4=B5=C0=A3=BA
Dear Christian,
 
Th= ank you. but some further doubts induced from your reply...
 
(1)
 
Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA
when you use the GUI instead of the command line then betfunc is
already used to find the mask which is then bei= ng passed to the
melodic binary
:  ...th= us melodic is always called with the --nobet option when using the GUI.
 
--So I thought if using the GUI, m= elodic could not include bet. Is it
right?  And in fact= , when I use melodic GUI, the melodic.log always show me --nobet, every t= ime, whether or not I choose bet.
 
(2)
 
from http://www.fmrib.ox.ac.uk/fsl/bet/index.html:
bet= func COMMAND-LINE SCRIPT...the -f threshold is reduced to 0.3...
 
--After melodic a= nalysis with command line function, I found in
melodic.log t= hat: Calling BET: /usr/local/fsl/bin/bet /tmp/tcac.0
/tmp/tc= ac.0_brain -m -f 0.4 > /dev/null
 <= /DIV>
Since -f 0.4 is not 0.3, so I thought that in melodic, b= et instead of betfunc is used. Is it right?
 
hope your clarification.
 
Regards,
wei

Christian Beckmann <beckmann@FMRIB= .OX.AC.UK> =D0=B4=B5=C0=A3=BA
Hi,

when you use the GUI instead of the command line then b= etfunc is
already used to find the mask which is then being passed to= the
melodic binary
hope this helps
christian

On 24 May = 2007, at 12:59, pengxu wei wrote:

> Dear Christian,
>
> Thank you.
>
> It is introduced t= hat betfunc uses a -f 0.3 for 4D fmri data while
> I found bet in = Melodic uses a -f 0.4. Bet in melodic cut off the
> most top slice= of my data but betfunc didn't. I think it is better
> for my data= to use betfunc first then pass the mask made to melodic
> explici= tly, as you said.
>
> Regards,
>
> wei
>> Christian Beckmann <[log in to unmask]>=D0=B4=B5=C0=A3=BA
&= gt; Hi
>
> melodic uses a different approach to robustifying = bet for 4D data. In
> betfund, the inital mask is generated from th= e first image and then
> dilated by 1 voxel in x,y and z before mas= king all volumes by this
> new dilated mask. In melodic the bet out= line is found by running bet
> on the mean image across time. I hav= e not evaluated the differences
> explicitly but guess that they ar= e small. Note that you can always
> pass a mask to melodic explicitly by using the -m option
> cheers
>= christian
>
>
>
> On 23 May 2007, at 15:49, peng= xu wei wrote:
>
> > Dear list,
> >
> > I= have read some message in the archive but no answer available.
> &= gt; betfunc is better than routine bet for 4D fmri data, so whether bet> > in Melodic is betfunc(when using GUI or command line), or jus= t
> > routine bet?
> >
> > Regards,
> &g= t;
> > wei
> >
> > =C7=C0=D7=A2=D1=C5=BB=A2=C3= =E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE=A3=A1
><= BR>> ____
> Christian F. Beckmann
> University Research Le= cturer
> Oxford University Centre for Functional MRI of the Brain (= FMRIB)
> John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.> [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann
>= ; tel: +44 1865 222551 fax: +44 1865 222717
>
>
> =D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE= =A3=A1

____
Christian F. Beckmann
University Research Lectur= er
Oxford University Centre for Functional MRI of the Brain (FMRIB)John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.
beckmann@fmr= ib.ox.ac.uk http://www.fmrib.ox.ac.uk/~beckmann
tel: +44 1865 222551 f= ax: +44 1865 222717


=C7=C0=D7=A2=D1=C5=BB=A2=C3=E2= =B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC=FE=A3=A1

=09


=C7= =C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8= =BD=BC=FE=A3=A1=20 --0-720768613-1180099791=:4834-- ========================================================================= Date: Fri, 25 May 2007 14:57:40 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit I tried this on a segmented brain mask image (value 0/1): avwmaths++ segmented.nii segmentedBYTE.nii -odt char after which segmentedBYTE.nii was zero everywhere, and avwmaths++ segmented.nii -mul 256 segmentedBYTE.nii -odt char after which the values that were 1 in segmentedBYTE.nii were 0 outside the mask and 1 inside teh mask. In other words: this is a way to make it work, but I'm not sure that this is correct! How many users expect the values in their images to change when they do a type conversion? Best Alle Meije ========================================================================= Date: Fri, 25 May 2007 16:13:03 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Tamar Gold <[log in to unmask]> Subject: logical operation Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" I have generated two masks that I would like to perform a Logical Operati= on=20 on, in order to see areas of overlap (regions in common to both of them).= =20=20 What command can I use to do this? ========================================================================= Date: Fri, 25 May 2007 16:46:02 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Michael Browning <[log in to unmask]> Subject: effective connectivity Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hi, looking for some information. Have been reading around about effective connectivity, specifically about = the pros and cons of DCM vs SEM. I get the impression that there's a fair = amount of reservation about the use of (both of) these modelling techniques= with fmri data. However most of the published stuff i could find was = relatively uncritical, particularly of DCM. Does anyone know of any = references about the limitations of DCM in fmri data? thanks mike browning ========================================================================= Date: Fri, 25 May 2007 17:29:46 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Tim Behrens <[log in to unmask]> Subject: Re: logical operation In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.2) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit The OR command is this: avmaths A -add B -bin AorB But it sounds like you need the AND command, which can be done like this avwmaths A -mas B AandB or avwmaths A -mul B -bin AandB Cheers T On 25 May 2007, at 16:13, Tamar Gold wrote: > I have generated two masks that I would like to perform a Logical > Operation > on, in order to see areas of overlap (regions in common to both of > them). > What command can I use to do this? ========================================================================= Date: Fri, 25 May 2007 17:44:05 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi, I can't replicate this on my laptop (mac os x). Instead, it works as expected - if I create a binary mask and then do the type conversion I still have the mask intact. AFAIK avwmaths does not use division but simply casts the values to the required type in the ususal c++ way. Here is what I did: i) load in avg152 in fslview and create a binary mask, save this to [Josi:~] avwhd avg152T1_brain-mask.nii.gz filename avg152T1_brain-mask.nii.gz sizeof_hdr 348 data_type INT16 ... This is a binary file: [Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R 0.000000 1.000000 0.000000 1.000000 When using avwmaths++ to convert the file [Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char it remains a binary mask [Josi:~] avwstats test -r -R 0.000000 1.000000 0.000000 1.000000 and has the desired type [Josi:~] avwhd test filename test.nii.gz sizeof_hdr 348 data_type UINT8 ... I guess we need to investigate - what platform are you running this on and did you compile yourself? cheers christian On 25 May 2007, at 14:57, Alle Meije Wink wrote: > I tried this on a segmented brain mask image (value 0/1): > > avwmaths++ segmented.nii segmentedBYTE.nii -odt char > > after which segmentedBYTE.nii was zero everywhere, and > > avwmaths++ segmented.nii -mul 256 segmentedBYTE.nii -odt char > > after which the values that were 1 in segmentedBYTE.nii were 0 > outside the mask and 1 inside teh mask. > > In other words: this is a way to make it work, but I'm not sure > that this is correct! How many users expect the values in their > images to change when they do a type conversion? > > Best > Alle Meije ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Fri, 25 May 2007 17:48:27 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: =BB=D8=B8=B4=A3=BA_[FSL]_=BB=D8=B8=B4=A3=BA_Re:_[FSL]__=E5=9B=9E=E5=A4=8D=EF=BC=3F Re: [FSL] BET in Mel odic In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=GB2312; delsp=yes; format=flowed Content-Transfer-Encoding: quoted-printable Hi That's right, the GUI does call betfunc and let's the command-line =20 version know that no betting is required by setting the -nobet =20 option. Other than switching bet off inside melodic (either by =20 passing a mask in explicitly via -m or using --nobet) there is no way =20= to change the behaviour, i.e. you can't pass in command line options. =20= You could change the setting nside the code and compile for yourself, =20= though. cheers christian On 25 May 2007, at 14:29, pengxu wei wrote: > Dear Christian, > > I think I got it. Melodic GUI never use bet, instead, it uses =20 > betfunc in pre-processing step. Or, I can choose command line =20 > function to do some 'liberal' analysis. > > Thank you. > > Regards, > > wei > > pengxu wei <[log in to unmask]> =D0=B4=B5=C0=A3=BA > Dear Christian, > > Thank you. but some further doubts induced from your reply... > > (1) > > Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA > when you use the GUI instead of the command line then betfunc is > already used to find the mask which is then being passed to the > melodic binary > but from http://www.jiscmail.ac.uk/cgi-bin/webadmin?=20 > A2=3Dind03&L=3DFSL&P=3DR12990&D=3D0&I=3D-3&m=3D418 > : ...thus melodic is always called with the --nobet option when =20 > using the GUI. > > --So I thought if using the GUI, melodic could not include bet. Is it > right? And in fact, when I use melodic GUI, the melodic.log always =20= > show me --nobet, every time, whether or not I choose bet. > > (2) > > from http://www.fmrib.ox.ac.uk/fsl/bet/index.html: > betfunc COMMAND-LINE SCRIPT...the -f threshold is reduced to 0.3... > > --After melodic analysis with command line function, I found in > melodic.log that: Calling BET: /usr/local/fsl/bin/bet /tmp/tcac.0 > /tmp/tcac.0_brain -m -f 0.4 > /dev/null > > Since -f 0.4 is not 0.3, so I thought that in melodic, bet instead =20 > of betfunc is used. Is it right? > > hope your clarification. > > Regards, > wei > > Christian Beckmann <[log in to unmask]> =D0=B4=B5=C0=A3=BA > Hi, > > when you use the GUI instead of the command line then betfunc is > already used to find the mask which is then being passed to the > melodic binary > hope this helps > christian > > On 24 May 2007, at 12:59, pengxu wei wrote: > > > Dear Christian, > > > > Thank you. > > > > It is introduced that betfunc uses a -f 0.3 for 4D fmri data while > > I found bet in Melodic uses a -f 0.4. Bet in melodic cut off the > > most top slice of my data but betfunc didn't. I think it is better > > for my data to use betfunc first then pass the mask made to melodic > > explicitly, as you said. > > > > Regards, > > > > wei > > > > Christian Beckmann =D0=B4=B5=C0=A3=BA > > Hi > > > > melodic uses a different approach to robustifying bet for 4D =20 > data. In > > betfund, the inital mask is generated from the first image and then > > dilated by 1 voxel in x,y and z before masking all volumes by this > > new dilated mask. In melodic the bet outline is found by running bet > > on the mean image across time. I have not evaluated the differences > > explicitly but guess that they are small. Note that you can always > > pass a mask to melodic explicitly by using the -m option > > cheers > > christian > > > > > > > > On 23 May 2007, at 15:49, pengxu wei wrote: > > > > > Dear list, > > > > > > I have read some message in the archive but no answer available. > > > betfunc is better than routine bet for 4D fmri data, so whether =20= > bet > > > in Melodic is betfunc(when using GUI or command line), or just > > > routine bet? > > > > > > Regards, > > > > > > wei > > > > > > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC= 20M=B8=BD=BC=FE=A3=A1 > > > > ____ > > Christian F. Beckmann > > University Research Lecturer > > Oxford University Centre for Functional MRI of the Brain (FMRIB) > > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > > tel: +44 1865 222551 fax: +44 1865 222717 > > > > > > =D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20M=B8=BD=BC= =FE=A3=A1 > > ____ > Christian F. Beckmann > University Research Lecturer > Oxford University Centre for Functional MRI of the Brain (FMRIB) > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > tel: +44 1865 222551 fax: +44 1865 222717 > > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20= M=B8=BD=BC=FE=A3=A1 > > > =C7=C0=D7=A2=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E43.5G=C8=DD=C1=BF=A3=AC20= M=B8=BD=BC=FE=A3=A1 ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Fri, 25 May 2007 23:16:54 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-997547621-1180131414=:35858" Content-Transfer-Encoding: 8bit --0-997547621-1180131414=:35858 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi Christian, Hmm, well I did the same now with the diagnostics on our EPI template (which is basically the MNI template, but thresholded for air and with the eyes removed). That is a 16-bit image. ================================================ $ export FSLOUTPUTTYPE=NIFTI $ avwhd++ EPI.nii | head -4 filename EPI.nii sizeof_hdr 348 data_type INT16 $ avwstats++ EPI.nii -r -R 0.000000 0.547875 0.000000 0.831373 $ segmentIm EPI.nii segmented image segmented.nii written (2 regions [1 clusters]) $ avwhd++ segmented.nii | head -4 filename segmented.nii sizeof_hdr 348 data_type INT16 $ avwstats++ segmented.nii -r -R 0.000000 0.003260 0.000000 0.003260 $ avwmaths++ segmented.nii segmentedBYTE.nii -odt char $ avwhd++ segmentedBYTE.nii.gz |head -4 filename segmentedBYTE.nii sizeof_hdr 348 data_type UINT8 $ avwstats++ segmentedBYTE.nii -r -R 0.000000 0.000000 0.000000 0.000000 ================================================ So first the diagnostics on the template. It's a 16-bit image. Its voxel values are just the same as the 8-bit MNI template, so not sure what the output of avwstats means. After that, segmenting the EPI template. 2 regions: the background and the brain. The diagnostics: still 16-bit data type, voxel values are 0 and 1 (I wrote the program that wrote the voxels), again not sure what avwstats output means. After that, cast the segmented image to char. The header shows that the data type has been changed. The stats show zero all... There's more happening there than just a type cast!!!! How can a integer-valued image have these float-valued stats? Would it help if I send you the images involved? I'm running Linux 3 2.6.10-1.771_FC2smp #1 SMP Mon Mar 28 01:10:51 EST 2005 i686 i686 i386 GNU/Linux And the FSL I'm using for this is a compiled version of FSL-3.3.8 Thanks, Alle Meije Christian Beckmann <[log in to unmask]> wrote: Hi, I can't replicate this on my laptop (mac os x). Instead, it works as expected - if I create a binary mask and then do the type conversion I still have the mask intact. AFAIK avwmaths does not use division but simply casts the values to the required type in the ususal c++ way. Here is what I did: i) load in avg152 in fslview and create a binary mask, save this to [Josi:~] avwhd avg152T1_brain-mask.nii.gz filename avg152T1_brain-mask.nii.gz sizeof_hdr 348 data_type INT16 ... This is a binary file: [Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R 0.000000 1.000000 0.000000 1.000000 When using avwmaths++ to convert the file [Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char it remains a binary mask [Josi:~] avwstats test -r -R 0.000000 1.000000 0.000000 1.000000 and has the desired type [Josi:~] avwhd test filename test.nii.gz sizeof_hdr 348 data_type UINT8 ... I guess we need to investigate - what platform are you running this on and did you compile yourself? cheers christian --------------------------------- Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your freeaccount today. --0-997547621-1180131414=:35858 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi Christian,

Hmm, well I did the same now with the diagnostics on our EPI template (which is basically the MNI template, but thresholded for air and with the eyes removed). That is a 16-bit image.

================================================
$ export FSLOUTPUTTYPE=NIFTI
$ avwhd++ EPI.nii | head -4
filename       EPI.nii

sizeof_hdr     348
data_type      INT16
$ avwstats++ EPI.nii -r -R
0.000000 0.547875 0.000000 0.831373
$ segmentIm EPI.nii
segmented image segmented.nii written (2 regions [1 clusters])
$ avwhd++ segmented.nii | head -4
filename       segmented.nii

sizeof_hdr     348
data_type      INT16
$ avwstats++ segmented.nii -r -R
0.000000 0.003260 0.000000 0.003260
$ avwmaths++ segmented.nii segmentedBYTE.nii -odt char
$ avwhd++ segmentedBYTE.nii.gz |head -4
filename       segmentedBYTE.nii

sizeof_hdr     348
data_type      UINT8
$ avwstats++ segmentedBYTE.nii -r -R
0.000000 0.000000 0.000000 0.000000
================================================

So first the diagnostics on the template. It's a 16-bit image. Its voxel values are just the same as the 8-bit MNI template, so not sure what the output of avwstats means.
After that, segmenting the EPI template. 2 regions: the background and the brain. The diagnostics: still 16-bit data type, voxel values are 0 and 1 (I wrote the program that wrote the voxels), again not sure what avwstats output means.
After that, cast the segmented image to char. The header shows that the data type has been changed. The stats show zero all...

There's more happening there than just a type cast!!!!

How can a integer-valued image have these float-valued stats?

Would it help if I send you the images involved?

I'm running
Linux 3 2.6.10-1.771_FC2smp #1 SMP Mon Mar 28 01:10:51 EST 2005 i686 i686 i386 GNU/Linux

And the FSL I'm using for this is a compiled version of FSL-3.3.8

Thanks,
Alle Meije

Christian Beckmann <[log in to unmask]> wrote:
Hi,

I can't replicate this on my laptop (mac os x). Instead, it works as
expected - if I create a binary mask and then do the type conversion
I still have the mask intact. AFAIK avwmaths does not use division
but simply casts the values to the required type in the ususal c++ way.

Here is what I did:
i) load in avg152 in fslview and create a binary mask, save this to

[Josi:~] avwhd avg152T1_brain-mask.nii.gz
filename avg152T1_brain-mask.nii.gz
sizeof_hdr 348
data_type INT16
...

This is a binary file:

[Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R
0.000000 1.000000 0.000000 1.000000

When using avwmaths++ to convert the file

[Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char

it remains a binary mask
[Josi:~] avwstats test -r -R
0.000000 1.000000 0.000000 1.000000

and has the desired type

[Josi:~] avwhd test
filename test.nii.gz
sizeof_hdr 348
data_type UINT8
...

I guess we need to investigate - what platform are you running this
on and did you compile yourself?
cheers
christian


Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your free account today. --0-997547621-1180131414=:35858-- ========================================================================= Date: Fri, 25 May 2007 23:52:52 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Paul Borghesani <[log in to unmask]> Subject: F tests after using nudge_gui Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: multipart/mixed; boundary="xG7ndtSSXF5I9zX+sNBe'r'Q(xbGwiIyFP(wQ1jISs2eL'uSHtl:MgaRj(l8-u?4d8IW(M" --xG7ndtSSXF5I9zX+sNBe'r'Q(xbGwiIyFP(wQ1jISs2eL'uSHtl:MgaRj(l8-u?4d8IW(M Content-Transfer-Encoding: 7bit Content-Type: multipart/mixed; boundary="9Yrs9HxhB/kWz8ekjH:e7S?rFzq9DXP?VooM6U8v'NEh,rh2TX)REbp)OmFKFTWUilB?pT" --9Yrs9HxhB/kWz8ekjH:e7S?rFzq9DXP?VooM6U8v'NEh,rh2TX)REbp)OmFKFTWUilB?pT Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Hi -=20 In order to correct minor problems with example_func2standard transformat= ion I have used the=20 nudge_gui to make a transformation matrix which I then applied to my reg_= standard/stats folder.=20 Thus I modified my reg_standard/stats copes and varcopes in my individual= .feat folders. When I=20 redid my higher level analysis with these modified files I get exactly wh= at I expect; i.e. slightly=20 different zstats in slightly different locations (frankly, quite minor ch= anges after considerable=20 work!). However, when I try to run an F test on these updated .feat fold= ers I get the following=20 error message (also see attached report.log): __________ An exception has been thrown Runtime error:- detected by Newmat: matrix is singular MatrixType =3D Crout # Rows =3D 2; # Cols =3D 2 Trace: Crout(lubksb); GeneralSolvI; InvertedMatrix::Evaluate; Gsmanager::= f_ols_contrast;=20 Gsmanager::ols_contrasts; Gsmanager::ols; Gsmanager::run. echo 4 > stats/dof /bin/rm -f stats/zem* stats/zols* stats/mask* Rendering using zmin=3D2.3 zmax=3D8 _________ Note: I am only using 6 .feat files inorder to de-bug this process. Runn= ing the same F test with=20 the "pre-nudged" data works exactly as expected. Is there a reason why a= fter modifying the=20 reg_standard/stats folder that higher leval analyses in general work, but= F tests do not? I suspect=20 that it has to do with the other .feat files (copes, varcopes, dof, reg f= older) that I have not updated=20 with my nudge transform, but from reviewing the log files I can't figure = out which of these might=20 be a problem. Thanks for you help=20=20 Paul Borghesani University of Washington --9Yrs9HxhB/kWz8ekjH:e7S?rFzq9DXP?VooM6U8v'NEh,rh2TX)REbp)OmFKFTWUilB?pT-- --xG7ndtSSXF5I9zX+sNBe'r'Q(xbGwiIyFP(wQ1jISs2eL'uSHtl:MgaRj(l8-u?4d8IW(M Content-Transfer-Encoding: 7bit Content-Type: APPLICATION/OCTET-STREAM; name="report.log" Started FEAT at Thu May 24 16:49:27 PDT 2007 on d-128-95-25-97.dhcp4.washington.edu FEAT output is /Users/paulborghesani/test_two++.gfeat/cope1.feat /bin/cp /Users/paulborghesani/test_two++.gfeat/design.fsf design.fsf /Applications/fsl/bin/feat_model design cat ../design.lcon | awk '{ print }' > design.lcon /Applications/fsl/bin/flame --cope=filtered_func_data --vc=var_filtered_func_data --mask=mask --ld=stats --dm=design.mat --cs=design.grp --tc=design.con --fc=design.fts --ols --nj=10000 --bi=500 --se=1 --fm --zlt=100000 --zut=100000 Log directory is: stats Setting up: nevs=2 ntpts=6 ngs=1 nvoxels=301661 Running: An exception has been thrown Runtime error:- detected by Newmat: matrix is singular MatrixType = Crout # Rows = 2; # Cols = 2 Trace: Crout(lubksb); GeneralSolvI; InvertedMatrix::Evaluate; Gsmanager::f_ols_contrast; Gsmanager::ols_contrasts; Gsmanager::ols; Gsmanager::run. echo 4 > stats/dof /bin/rm -f stats/zem* stats/zols* stats/mask* Rendering using zmin=2.3 zmax=8 --xG7ndtSSXF5I9zX+sNBe'r'Q(xbGwiIyFP(wQ1jISs2eL'uSHtl:MgaRj(l8-u?4d8IW(M-- ========================================================================= Date: Sat, 26 May 2007 10:26:28 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi, Your segmented.nii does not appear to be a binary image at all, the robust range is 0 - 0.003260. How did you deal with the header information in the new image then? Did you simply clone the header from the epi template? This also already appears to be inconsistent wrt to it's range and the data type. ? Maybe it's indeed worthwhile to upload the data. What happens when you load the original epp.nii and segmented.nii into fslview - what values are reporte? The behaviour of avwstats++ is as I expect it to be - all values are cast to 0. cheers christian On 25 May 2007, at 23:16, Alle Meije Wink wrote: > Hi Christian, > > Hmm, well I did the same now with the diagnostics on our EPI > template (which is basically the MNI template, but thresholded for > air and with the eyes removed). That is a 16-bit image. > > ================================================ > $ export FSLOUTPUTTYPE=NIFTI > $ avwhd++ EPI.nii | head -4 > filename EPI.nii > > sizeof_hdr 348 > data_type INT16 > $ avwstats++ EPI.nii -r -R > 0.000000 0.547875 0.000000 0.831373 > $ segmentIm EPI.nii > segmented image segmented.nii written (2 regions [1 clusters]) > $ avwhd++ segmented.nii | head -4 > filename segmented.nii > > sizeof_hdr 348 > data_type INT16 > $ avwstats++ segmented.nii -r -R > 0.000000 0.003260 0.000000 0.003260 > $ avwmaths++ segmented.nii segmentedBYTE.nii -odt char > $ avwhd++ segmentedBYTE.nii.gz |head -4 > filename segmentedBYTE.nii > > sizeof_hdr 348 > data_type UINT8 > $ avwstats++ segmentedBYTE.nii -r -R > 0.000000 0.000000 0.000000 0.000000 > ================================================ > > So first the diagnostics on the template. It's a 16-bit image. Its > voxel values are just the same as the 8-bit MNI template, so not > sure what the output of avwstats means. > After that, segmenting the EPI template. 2 regions: the background > and the brain. The diagnostics: still 16-bit data type, voxel > values are 0 and 1 (I wrote the program that wrote the voxels), > again not sure what avwstats output means. > After that, cast the segmented image to char. The header shows that > the data type has been changed. The stats show zero all... > > There's more happening there than just a type cast!!!! > > How can a integer-valued image have these float-valued stats? > > Would it help if I send you the images involved? > > I'm running > Linux 3 2.6.10-1.771_FC2smp #1 SMP Mon Mar 28 01:10:51 EST 2005 > i686 i686 i386 GNU/Linux > > And the FSL I'm using for this is a compiled version of FSL-3.3.8 > > Thanks, > Alle Meije > > Christian Beckmann <[log in to unmask]> wrote: Hi, > > I can't replicate this on my laptop (mac os x). Instead, it works as > expected - if I create a binary mask and then do the type conversion > I still have the mask intact. AFAIK avwmaths does not use division > but simply casts the values to the required type in the ususal c++ > way. > > Here is what I did: > i) load in avg152 in fslview and create a binary mask, save this to > > [Josi:~] avwhd avg152T1_brain-mask.nii.gz > filename avg152T1_brain-mask.nii.gz > sizeof_hdr 348 > data_type INT16 > ... > > This is a binary file: > > [Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R > 0.000000 1.000000 0.000000 1.000000 > > When using avwmaths++ to convert the file > > [Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char > > it remains a binary mask > [Josi:~] avwstats test -r -R > 0.000000 1.000000 0.000000 1.000000 > > and has the desired type > > [Josi:~] avwhd test > filename test.nii.gz > sizeof_hdr 348 > data_type UINT8 > ... > > I guess we need to investigate - what platform are you running this > on and did you compile yourself? > cheers > christian > > > Yahoo! Mail is the world's favourite email. Don't settle for less, > sign up for your free account today. ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Sat, 26 May 2007 15:35:42 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit That is what I found strange as well! As I said earlier, I know how segmented.nii is written, I know that there are only 0 and 1 voxels in that image. I opened it in MriCro though, and all the in-mask voxels show value 1 / .003260 (I guess that means the vvalue stored in the voxel array is 1, but in some way this represents 0.003260). The header is cloned from the template. I know that is not the `cleanest' way to go about. That's probably why the range etc are not set to 0..1. Fact is that before avwmaths++ I see the shape of the mask, and voxels that are stored as 1. After avwmaths they are gone, so more happens than just a type conversion. Probably header information is used? Where should I send the data, if you want to have a look? Thanks AM Christian Beckmann wrote: > Hi, > > Your segmented.nii does not appear to be a binary image at all, the > robust range is 0 - 0.003260. How did you deal with the header > information in the new image then? Did you simply clone the header from > the epi template? This also already appears to be inconsistent wrt to > it's range and the data type. ? Maybe it's indeed worthwhile to upload > the data. What happens when you load the original epp.nii and > segmented.nii into fslview - what values are reporte? The behaviour of > avwstats++ is as I expect it to be - all values are cast to 0. > cheers > christian > > > > On 25 May 2007, at 23:16, Alle Meije Wink wrote: > >> Hi Christian, >> >> Hmm, well I did the same now with the diagnostics on our EPI template >> (which is basically the MNI template, but thresholded for air and with >> the eyes removed). That is a 16-bit image. >> >> ================================================ >> $ export FSLOUTPUTTYPE=NIFTI >> $ avwhd++ EPI.nii | head -4 >> filename EPI.nii >> >> sizeof_hdr 348 >> data_type INT16 >> $ avwstats++ EPI.nii -r -R >> 0.000000 0.547875 0.000000 0.831373 >> $ segmentIm EPI.nii >> segmented image segmented.nii written (2 regions [1 clusters]) >> $ avwhd++ segmented.nii | head -4 >> filename segmented.nii >> >> sizeof_hdr 348 >> data_type INT16 >> $ avwstats++ segmented.nii -r -R >> 0.000000 0.003260 0.000000 0.003260 >> $ avwmaths++ segmented.nii segmentedBYTE.nii -odt char >> $ avwhd++ segmentedBYTE.nii.gz |head -4 >> filename segmentedBYTE.nii >> >> sizeof_hdr 348 >> data_type UINT8 >> $ avwstats++ segmentedBYTE.nii -r -R >> 0.000000 0.000000 0.000000 0.000000 >> ================================================ >> >> So first the diagnostics on the template. It's a 16-bit image. Its >> voxel values are just the same as the 8-bit MNI template, so not sure >> what the output of avwstats means. >> After that, segmenting the EPI template. 2 regions: the background and >> the brain. The diagnostics: still 16-bit data type, voxel values are 0 >> and 1 (I wrote the program that wrote the voxels), again not sure what >> avwstats output means. >> After that, cast the segmented image to char. The header shows that >> the data type has been changed. The stats show zero all... >> >> There's more happening there than just a type cast!!!! >> >> How can a integer-valued image have these float-valued stats? >> >> Would it help if I send you the images involved? >> >> I'm running >> Linux 3 2.6.10-1.771_FC2smp #1 SMP Mon Mar 28 01:10:51 EST 2005 i686 >> i686 i386 GNU/Linux >> >> And the FSL I'm using for this is a compiled version of FSL-3.3.8 >> >> Thanks, >> Alle Meije >> >> Christian Beckmann <[log in to unmask]> wrote: Hi, >> >> I can't replicate this on my laptop (mac os x). Instead, it works as >> expected - if I create a binary mask and then do the type conversion >> I still have the mask intact. AFAIK avwmaths does not use division >> but simply casts the values to the required type in the ususal c++ way. >> >> Here is what I did: >> i) load in avg152 in fslview and create a binary mask, save this to >> >> [Josi:~] avwhd avg152T1_brain-mask.nii.gz >> filename avg152T1_brain-mask.nii.gz >> sizeof_hdr 348 >> data_type INT16 >> ... >> >> This is a binary file: >> >> [Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R >> 0.000000 1.000000 0.000000 1.000000 >> >> When using avwmaths++ to convert the file >> >> [Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char >> >> it remains a binary mask >> [Josi:~] avwstats test -r -R >> 0.000000 1.000000 0.000000 1.000000 >> >> and has the desired type >> >> [Josi:~] avwhd test >> filename test.nii.gz >> sizeof_hdr 348 >> data_type UINT8 >> ... >> >> I guess we need to investigate - what platform are you running this >> on and did you compile yourself? >> cheers >> christian >> >> >> Yahoo! Mail is the world's favourite email. Don't settle for less, >> sign up for your free account today. > > ____ > Christian F. Beckmann > University Research Lecturer > Oxford University Centre for Functional MRI of the Brain (FMRIB) > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > tel: +44 1865 222551 fax: +44 1865 222717 > ========================================================================= Date: Sat, 26 May 2007 16:37:32 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Ged Ridgway <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Hi, Isn't this just the standard NIfTI intensity mapping using "scl_slope" of 0.00326 (and scl_inter, which I assume here is 0)? avwhd spm5/templates/EPI.nii | grep -E 'data_type|scl' data_type UINT8 scl_slope 0.003260 scl_inter 0.000000 E.g. using Matlab/SPM5: V = spm_vol(fullfile(spm('Dir'),'templates','EPI.nii')) img = spm_read_vols(V); max(img(:)) fid = fopen(V.fname); dat = fread(fid, inf, 'uint8'); max(dat) max(dat)*V.pinfo(1) shows the behaviour I'd expect. max(dat) is 255, but the image intensities read in via spm_read_vols (and presumably FSL?) are scaled by scl_slope. This is helpful with e.g. spm5/tpm/grey.nii where the scaling allows meaningful 0-to-1 probabilities to be stored as UINT8. By copying all the header information, including scl_slope, your binary 1 maps to 1*scl_slope or 0.003260. avwmaths is then casting this to zero, instead of setting an appropriate scl_slope on the output volume. I'm not sure if this is a bug or a feature... Though I believe SPM5 usually recomputes scl_slope for the output volume so that the raw data can be efficiently stored in the chosen data-type, while the intensities remain the same. In any case, feature or bug, if you make scl_slope 1 in your binary mask header, all should be well. E.g. use avwedithd. HTH, Ged Alle Meije Wink wrote: > That is what I found strange as well! > > As I said earlier, I know how segmented.nii is written, I know that > there are only 0 and 1 voxels in that image. I opened it in MriCro > though, and all the in-mask voxels show value 1 / .003260 (I guess that > means the vvalue stored in the voxel array is 1, but in some way this > represents 0.003260). > > The header is cloned from the template. I know that is not the > `cleanest' way to go about. That's probably why the range etc are not > set to 0..1. > > Fact is that before avwmaths++ I see the shape of the mask, and voxels > that are stored as 1. After avwmaths they are gone, so more happens than > just a type conversion. Probably header information is used? > > Where should I send the data, if you want to have a look? > > Thanks > > AM > > Christian Beckmann wrote: >> Hi, >> >> Your segmented.nii does not appear to be a binary image at all, the >> robust range is 0 - 0.003260. How did you deal with the header >> information in the new image then? Did you simply clone the header >> from the epi template? This also already appears to be inconsistent >> wrt to it's range and the data type. ? Maybe it's indeed worthwhile to >> upload the data. What happens when you load the original epp.nii and >> segmented.nii into fslview - what values are reporte? The behaviour of >> avwstats++ is as I expect it to be - all values are cast to 0. >> cheers >> christian >> >> >> >> On 25 May 2007, at 23:16, Alle Meije Wink wrote: >> >>> Hi Christian, >>> >>> Hmm, well I did the same now with the diagnostics on our EPI template >>> (which is basically the MNI template, but thresholded for air and >>> with the eyes removed). That is a 16-bit image. >>> >>> ================================================ >>> $ export FSLOUTPUTTYPE=NIFTI >>> $ avwhd++ EPI.nii | head -4 >>> filename EPI.nii >>> >>> sizeof_hdr 348 >>> data_type INT16 >>> $ avwstats++ EPI.nii -r -R >>> 0.000000 0.547875 0.000000 0.831373 >>> $ segmentIm EPI.nii >>> segmented image segmented.nii written (2 regions [1 clusters]) >>> $ avwhd++ segmented.nii | head -4 >>> filename segmented.nii >>> >>> sizeof_hdr 348 >>> data_type INT16 >>> $ avwstats++ segmented.nii -r -R >>> 0.000000 0.003260 0.000000 0.003260 >>> $ avwmaths++ segmented.nii segmentedBYTE.nii -odt char >>> $ avwhd++ segmentedBYTE.nii.gz |head -4 >>> filename segmentedBYTE.nii >>> >>> sizeof_hdr 348 >>> data_type UINT8 >>> $ avwstats++ segmentedBYTE.nii -r -R >>> 0.000000 0.000000 0.000000 0.000000 >>> ================================================ >>> >>> So first the diagnostics on the template. It's a 16-bit image. Its >>> voxel values are just the same as the 8-bit MNI template, so not sure >>> what the output of avwstats means. >>> After that, segmenting the EPI template. 2 regions: the background >>> and the brain. The diagnostics: still 16-bit data type, voxel values >>> are 0 and 1 (I wrote the program that wrote the voxels), again not >>> sure what avwstats output means. >>> After that, cast the segmented image to char. The header shows that >>> the data type has been changed. The stats show zero all... >>> >>> There's more happening there than just a type cast!!!! >>> >>> How can a integer-valued image have these float-valued stats? >>> >>> Would it help if I send you the images involved? >>> >>> I'm running >>> Linux 3 2.6.10-1.771_FC2smp #1 SMP Mon Mar 28 01:10:51 EST 2005 i686 >>> i686 i386 GNU/Linux >>> >>> And the FSL I'm using for this is a compiled version of FSL-3.3.8 >>> >>> Thanks, >>> Alle Meije >>> >>> Christian Beckmann <[log in to unmask]> wrote: Hi, >>> >>> I can't replicate this on my laptop (mac os x). Instead, it works as >>> expected - if I create a binary mask and then do the type conversion >>> I still have the mask intact. AFAIK avwmaths does not use division >>> but simply casts the values to the required type in the ususal c++ way. >>> >>> Here is what I did: >>> i) load in avg152 in fslview and create a binary mask, save this to >>> >>> [Josi:~] avwhd avg152T1_brain-mask.nii.gz >>> filename avg152T1_brain-mask.nii.gz >>> sizeof_hdr 348 >>> data_type INT16 >>> ... >>> >>> This is a binary file: >>> >>> [Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R >>> 0.000000 1.000000 0.000000 1.000000 >>> >>> When using avwmaths++ to convert the file >>> >>> [Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char >>> >>> it remains a binary mask >>> [Josi:~] avwstats test -r -R >>> 0.000000 1.000000 0.000000 1.000000 >>> >>> and has the desired type >>> >>> [Josi:~] avwhd test >>> filename test.nii.gz >>> sizeof_hdr 348 >>> data_type UINT8 >>> ... >>> >>> I guess we need to investigate - what platform are you running this >>> on and did you compile yourself? >>> cheers >>> christian >>> >>> >>> Yahoo! Mail is the world's favourite email. Don't settle for less, >>> sign up for your free account today. >> >> ____ >> Christian F. Beckmann >> University Research Lecturer >> Oxford University Centre for Functional MRI of the Brain (FMRIB) >> John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. >> [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann >> tel: +44 1865 222551 fax: +44 1865 222717 >> > ========================================================================= Date: Sat, 26 May 2007 22:22:59 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: TBSS Skeleton Atlas? Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Does anyone know if there is an atlas for the typical TBSS mean skeleton? Best, Marc ========================================================================= Date: Sun, 27 May 2007 07:22:31 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: TBSS Skeleton Atlas? In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi - you mean a labelling of different regions? No, I'm afraid not, but we are working on that. Cheers. On 26 May 2007, at 22:22, Marc Dubin wrote: > Does anyone know if there is an atlas > for the typical TBSS mean skeleton? > > Best, > Marc ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Sun, 27 May 2007 07:26:45 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: F tests after using nudge_gui In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi, It seems most likely that the f-test problem really isn't related to the different resampling of the first-level images. Are you _sure_ that the second-level model and contrasts (and f contrasts) are all correct and valid? You might want to send them in. In any case, an easy way to check that the resampling is ok is: - make sure you have overwritten the example_func2standard.mat transform in the .feat/reg directory with your Nudge output - rm -rf reg_standard - featregapply Cheers, Steve. On 25 May 2007, at 23:52, Paul Borghesani wrote: > Hi - > > In order to correct minor problems with example_func2standard > transformation I have used the > nudge_gui to make a transformation matrix which I then applied to > my reg_standard/stats folder. > Thus I modified my reg_standard/stats copes and varcopes in my > individual .feat folders. When I > redid my higher level analysis with these modified files I get > exactly what I expect; i.e. slightly > different zstats in slightly different locations (frankly, quite > minor changes after considerable > work!). However, when I try to run an F test on these > updated .feat folders I get the following > error message (also see attached report.log): > > __________ > > An exception has been thrown > Runtime error:- detected by Newmat: matrix is singular > > MatrixType = Crout # Rows = 2; # Cols = 2 > Trace: Crout(lubksb); GeneralSolvI; InvertedMatrix::Evaluate; > Gsmanager::f_ols_contrast; > Gsmanager::ols_contrasts; Gsmanager::ols; Gsmanager::run. > > > echo 4 > stats/dof > > /bin/rm -f stats/zem* stats/zols* stats/mask* > Rendering using zmin=2.3 zmax=8 > > _________ > > Note: I am only using 6 .feat files inorder to de-bug this > process. Running the same F test with > the "pre-nudged" data works exactly as expected. Is there a reason > why after modifying the > reg_standard/stats folder that higher leval analyses in general > work, but F tests do not? I suspect > that it has to do with the other .feat files (copes, varcopes, dof, > reg folder) that I have not updated > with my nudge transform, but from reviewing the log files I can't > figure out which of these might > be a problem. > > > Thanks for you help > > Paul Borghesani > University of Washington > > > ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Sun, 27 May 2007 07:28:13 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Indeed - also, there may be an issue with your version of FSLView not starting with a correct intensity display range, you might try changing that by hand - the next version will solve that. Cheers. On 26 May 2007, at 16:37, Ged Ridgway wrote: > Hi, > > Isn't this just the standard NIfTI intensity mapping using > "scl_slope" of 0.00326 (and scl_inter, which I assume here is 0)? > > avwhd spm5/templates/EPI.nii | grep -E 'data_type|scl' > data_type UINT8 > scl_slope 0.003260 > scl_inter 0.000000 > > E.g. using Matlab/SPM5: > > V = spm_vol(fullfile(spm('Dir'),'templates','EPI.nii')) > img = spm_read_vols(V); > max(img(:)) > fid = fopen(V.fname); > dat = fread(fid, inf, 'uint8'); > max(dat) > max(dat)*V.pinfo(1) > > shows the behaviour I'd expect. max(dat) is 255, but the image > intensities read in via spm_read_vols (and presumably FSL?) are > scaled by scl_slope. This is helpful with e.g. spm5/tpm/grey.nii > where the scaling allows meaningful 0-to-1 probabilities to be > stored as UINT8. > > By copying all the header information, including scl_slope, your > binary 1 maps to 1*scl_slope or 0.003260. > > avwmaths is then casting this to zero, instead of setting an > appropriate scl_slope on the output volume. I'm not sure if this is > a bug or a feature... Though I believe SPM5 usually recomputes > scl_slope for the output volume so that the raw data can be > efficiently stored in the chosen data-type, while the intensities > remain the same. > > In any case, feature or bug, if you make scl_slope 1 in your binary > mask header, all should be well. E.g. use avwedithd. > > HTH, > Ged > > > Alle Meije Wink wrote: >> That is what I found strange as well! >> As I said earlier, I know how segmented.nii is written, I know >> that there are only 0 and 1 voxels in that image. I opened it in >> MriCro though, and all the in-mask voxels show value 1 / .003260 >> (I guess that means the vvalue stored in the voxel array is 1, but >> in some way this represents 0.003260). >> The header is cloned from the template. I know that is not the >> `cleanest' way to go about. That's probably why the range etc are >> not set to 0..1. >> Fact is that before avwmaths++ I see the shape of the mask, and >> voxels that are stored as 1. After avwmaths they are gone, so more >> happens than just a type conversion. Probably header information >> is used? >> Where should I send the data, if you want to have a look? >> Thanks >> AM >> Christian Beckmann wrote: >>> Hi, >>> >>> Your segmented.nii does not appear to be a binary image at all, >>> the robust range is 0 - 0.003260. How did you deal with the >>> header information in the new image then? Did you simply clone >>> the header from the epi template? This also already appears to be >>> inconsistent wrt to it's range and the data type. ? Maybe it's >>> indeed worthwhile to upload the data. What happens when you load >>> the original epp.nii and segmented.nii into fslview - what values >>> are reporte? The behaviour of avwstats++ is as I expect it to be >>> - all values are cast to 0. >>> cheers >>> christian >>> >>> >>> >>> On 25 May 2007, at 23:16, Alle Meije Wink wrote: >>> >>>> Hi Christian, >>>> >>>> Hmm, well I did the same now with the diagnostics on our EPI >>>> template (which is basically the MNI template, but thresholded >>>> for air and with the eyes removed). That is a 16-bit image. >>>> >>>> ================================================ >>>> $ export FSLOUTPUTTYPE=NIFTI >>>> $ avwhd++ EPI.nii | head -4 >>>> filename EPI.nii >>>> >>>> sizeof_hdr 348 >>>> data_type INT16 >>>> $ avwstats++ EPI.nii -r -R >>>> 0.000000 0.547875 0.000000 0.831373 >>>> $ segmentIm EPI.nii >>>> segmented image segmented.nii written (2 regions [1 clusters]) >>>> $ avwhd++ segmented.nii | head -4 >>>> filename segmented.nii >>>> >>>> sizeof_hdr 348 >>>> data_type INT16 >>>> $ avwstats++ segmented.nii -r -R >>>> 0.000000 0.003260 0.000000 0.003260 >>>> $ avwmaths++ segmented.nii segmentedBYTE.nii -odt char >>>> $ avwhd++ segmentedBYTE.nii.gz |head -4 >>>> filename segmentedBYTE.nii >>>> >>>> sizeof_hdr 348 >>>> data_type UINT8 >>>> $ avwstats++ segmentedBYTE.nii -r -R >>>> 0.000000 0.000000 0.000000 0.000000 >>>> ================================================ >>>> >>>> So first the diagnostics on the template. It's a 16-bit image. >>>> Its voxel values are just the same as the 8-bit MNI template, so >>>> not sure what the output of avwstats means. >>>> After that, segmenting the EPI template. 2 regions: the >>>> background and the brain. The diagnostics: still 16-bit data >>>> type, voxel values are 0 and 1 (I wrote the program that wrote >>>> the voxels), again not sure what avwstats output means. >>>> After that, cast the segmented image to char. The header shows >>>> that the data type has been changed. The stats show zero all... >>>> >>>> There's more happening there than just a type cast!!!! >>>> >>>> How can a integer-valued image have these float-valued stats? >>>> >>>> Would it help if I send you the images involved? >>>> >>>> I'm running >>>> Linux 3 2.6.10-1.771_FC2smp #1 SMP Mon Mar 28 01:10:51 EST 2005 >>>> i686 i686 i386 GNU/Linux >>>> >>>> And the FSL I'm using for this is a compiled version of FSL-3.3.8 >>>> >>>> Thanks, >>>> Alle Meije >>>> >>>> Christian Beckmann <[log in to unmask]> wrote: Hi, >>>> >>>> I can't replicate this on my laptop (mac os x). Instead, it >>>> works as >>>> expected - if I create a binary mask and then do the type >>>> conversion >>>> I still have the mask intact. AFAIK avwmaths does not use division >>>> but simply casts the values to the required type in the ususal c+ >>>> + way. >>>> >>>> Here is what I did: >>>> i) load in avg152 in fslview and create a binary mask, save this to >>>> >>>> [Josi:~] avwhd avg152T1_brain-mask.nii.gz >>>> filename avg152T1_brain-mask.nii.gz >>>> sizeof_hdr 348 >>>> data_type INT16 >>>> ... >>>> >>>> This is a binary file: >>>> >>>> [Josi:~] avwstats avg152T1_brain-mask.nii.gz -r -R >>>> 0.000000 1.000000 0.000000 1.000000 >>>> >>>> When using avwmaths++ to convert the file >>>> >>>> [Josi:~] avwmaths++ avg152T1_brain-mask.nii.gz test -odt char >>>> >>>> it remains a binary mask >>>> [Josi:~] avwstats test -r -R >>>> 0.000000 1.000000 0.000000 1.000000 >>>> >>>> and has the desired type >>>> >>>> [Josi:~] avwhd test >>>> filename test.nii.gz >>>> sizeof_hdr 348 >>>> data_type UINT8 >>>> ... >>>> >>>> I guess we need to investigate - what platform are you running this >>>> on and did you compile yourself? >>>> cheers >>>> christian >>>> >>>> >>>> Yahoo! Mail is the world's favourite email. Don't settle for >>>> less, sign up for your free account today. >>> >>> ____ >>> Christian F. Beckmann >>> University Research Lecturer >>> Oxford University Centre for Functional MRI of the Brain (FMRIB) >>> John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. >>> [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann >>> tel: +44 1865 222551 fax: +44 1865 222717 >>> ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Sun, 27 May 2007 05:21:31 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: TBSS Skeleton Atlas? In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_61656_11388677.1180257691899" ------=_Part_61656_11388677.1180257691899 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline That would be very helpful as existing atlases could be used, but things look slightly different on the skeleton.Best, Marc On 5/27/07, Steve Smith <[log in to unmask]> wrote: > > Hi - you mean a labelling of different regions? No, I'm afraid not, > but we are working on that. > Cheers. > > On 26 May 2007, at 22:22, Marc Dubin wrote: > > > Does anyone know if there is an atlas > > for the typical TBSS mean skeleton? > > > > Best, > > Marc > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > -- 646 831 8886 [log in to unmask] ------=_Part_61656_11388677.1180257691899 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline That would be very helpful as existing atlases could be used, but things look slightly different on the skeleton.
Best,
Marc

On 5/27/07, Steve Smith <[log in to unmask]> wrote:
Hi - you mean a labelling of different regions? No, I'm afraid not,
but we are working on that.
Cheers.

On 26 May 2007, at 22:22, Marc Dubin wrote:

> Does anyone know if there is an atlas
> for the typical TBSS mean skeleton?
>
> Best,
> Marc


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------ ------------
---



--
646 831 8886
[log in to unmask]
------=_Part_61656_11388677.1180257691899-- ========================================================================= Date: Sun, 27 May 2007 12:25:47 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Alle Meije Wink <[log in to unmask]> Subject: Re: avwmaths ODT In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Ah, OK. So indeed avwmaths does not just read the array, typecast all the numbers in the array, and write it out. My software (which completely ignores scale_factor, as well as some other things) was more or less relying on that :). Segmented images (where the voxel values are cluster indices) should have scale_factor 1, I guess. I'll just add this into my segmentIm program. Cheers guys, Alle Meije Steve Smith wrote: > Indeed - also, there may be an issue with your version of FSLView not > starting with a correct intensity display range, you might try changing > that by hand - the next version will solve that. > > Cheers. > > > On 26 May 2007, at 16:37, Ged Ridgway wrote: > >> Hi, >> >> Isn't this just the standard NIfTI intensity mapping using "scl_slope" >> of 0.00326 (and scl_inter, which I assume here is 0)? >> >> avwhd spm5/templates/EPI.nii | grep -E 'data_type|scl' >> data_type UINT8 >> scl_slope 0.003260 >> scl_inter 0.000000 >> >> E.g. using Matlab/SPM5: >> >> V = spm_vol(fullfile(spm('Dir'),'templates','EPI.nii')) >> img = spm_read_vols(V); >> max(img(:)) >> fid = fopen(V.fname); >> dat = fread(fid, inf, 'uint8'); >> max(dat) >> max(dat)*V.pinfo(1) >> >> shows the behaviour I'd expect. max(dat) is 255, but the image >> intensities read in via spm_read_vols (and presumably FSL?) are scaled >> by scl_slope. This is helpful with e.g. spm5/tpm/grey.nii where the >> scaling allows meaningful 0-to-1 probabilities to be stored as UINT8. >> >> By copying all the header information, including scl_slope, your >> binary 1 maps to 1*scl_slope or 0.003260. >> >> avwmaths is then casting this to zero, instead of setting an >> appropriate scl_slope on the output volume. I'm not sure if this is a >> bug or a feature... Though I believe SPM5 usually recomputes scl_slope >> for the output volume so that the raw data can be efficiently stored >> in the chosen data-type, while the intensities remain the same. >> >> In any case, feature or bug, if you make scl_slope 1 in your binary >> mask header, all should be well. E.g. use avwedithd. >> >> HTH, >> Ged ========================================================================= Date: Sun, 27 May 2007 23:04:09 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Parametric stats in TBSS? Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Does anyone know if it is possible to use parametric stats in TBSS as well as the bootstrap stats in "randomise"? Best, Marc ========================================================================= Date: Mon, 28 May 2007 07:59:54 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Parametric stats in TBSS? In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi - please see the original paper on TBSS. Although the data is most likely Gaussian distributed, correcting for multiple comparison corrections across voxels would be difficult to get right with parametric statistics. Cheers. On 27 May 2007, at 23:04, Marc Dubin wrote: > Does anyone know if it is possible to use parametric stats > in TBSS as well as the bootstrap stats in "randomise"? > > Best, > Marc ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Mon, 28 May 2007 09:27:21 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Benny Liberg <[log in to unmask]> Subject: Peak voxel within mask in avwstats and dilation of ROI-mask Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" Dear FSLers, I have two questions, and I would really appreciate if anyone could help = me=20 out on this rather simple matter. I compare 5 individuals that have done two slightly different experiments= .=20 So far, I've done a FFX higher level, regional analysis of both=20 experiments. I rerun the higher level analysis with each of the eight=20 masks. I then extract the peak vxl from rendered_zstats (avwstats) and th= en=20 create a mask containing peak vxl and the 26 surrounding ones (avwmaths).= I=20 then run featquery with that new mask to get mean BOLD % change. I can't help thinking that this could be solved by running avwstats -x wi= th=20 some kind of mask option and then create the 27 vxl ROI with avwmaths. An= y=20 suggestions? Also, is there any way of dilating a mask? Best regards, Benny Liberg=20=20 Instead of rerunning the higher analysis with new pre-thr masks applied=20= masks (8) for each anatomical strucuture, I would like to:=20 1. extract the peak vxl within a mask (avwstats) and=20 2. create a new mask containing 26 vxls around the peak vxl (avwmaths) an= d=20 extract mean %BOLD change from these 27 vxls. I know how to do 1 and 2 but can't figure out how to avoid ========================================================================= Date: Mon, 28 May 2007 09:21:06 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Parametric stats in TBSS? In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_66961_16373490.1180358466065" ------=_Part_66961_16373490.1180358466065 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Is this a fundamental difference between FSL and SPM? On 5/28/07, Steve Smith <[log in to unmask]> wrote: > > Hi - please see the original paper on TBSS. > Although the data is most likely Gaussian distributed, correcting for > multiple comparison corrections across voxels would be difficult to > get right with parametric statistics. > Cheers. > > > On 27 May 2007, at 23:04, Marc Dubin wrote: > > > Does anyone know if it is possible to use parametric stats > > in TBSS as well as the bootstrap stats in "randomise"? > > > > Best, > > Marc > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > -- 646 831 8886 [log in to unmask] ------=_Part_66961_16373490.1180358466065 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Is this a fundamental difference between FSL and SPM?

On 5/28/07, Steve Smith <[log in to unmask]> wrote:
Hi - please see the original paper on TBSS.
Although the data is most likely Gaussian distributed, correcting for
multiple comparison corrections across voxels would be difficult to
get right with parametric statistics.
Cheers.


On 27 May 2007, at 23:04, Marc Dubin wrote:

> Does anyone know if it is possible to use parametric stats
> in TBSS as well as the bootstrap stats in "randomise"?
>
> Best,
> Marc


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---



--
646 831 8886
[log in to unmask]
------=_Part_66961_16373490.1180358466065-- ========================================================================= Date: Mon, 28 May 2007 09:22:23 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Kideckel <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_239206_31672536.1180358543945" ------=_Part_239206_31672536.1180358543945 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hello. i have tried your previous suggestions, and think that the problem may be in the bvecs file, but i do not know for sure. how would i check for this? thanks again. David On 5/25/07, Tim Behrens <[log in to unmask]> wrote: > > Hi - if you run bedpost_datacheck on your directory it might tell you > what's wrong - for example make sure the number of entriess in bvals and > bvecs is the same as the number of volumes in your data. > > T > On 24 May 2007, at 00:17, David Kideckel wrote: > > hi. thank you for your support. once i run bet from the command line > (using cygwin), it does not seem to be working properly. but when i use the > gui, it does... > anyway, i am trying to rum dtifit from the command line and get the > following message: > 0 256 0 256 0 50 > 0 slices processed > Aborted > > Any idea why this is happening? Did I enter something wrong? > > Thanks. > David > > On 5/23/07, Tim Behrens <[log in to unmask]> wrote: > > > > If you select dtifit from the drop down menu on the same page, all` the > > information about file specifications is there. Don't forget to end your > > bvecs and bvals files with carriage returns. > > Cheers > > > > T > > On 23 May 2007, at 18:53, David Kideckel wrote: > > > > Hi. That was one of the firs things I did do. I will try to be more > > specific with my question. I want to do TBSS, and no tractography. So, would > > I do the following in this order: > > Eddy Correct > > BET > > DTI Fit > > > > If so, where is it in the manual specified how to create and input the > > bvecs and bvals files? I follow the link it gives in the FAQ section, but I > > can't seem to find this info. > > > > Thanks again for your help. > > David > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > Hi - if you go to http://www.fmrib.ox.ac.uk/fsl/fdt/index.html > > > > > > and select "processing pipeline", all should be clear. > > > > > > T > > > > > > On 23 May 2007, at 17:22, David Kideckel wrote: > > > > > > Hi. Thank you for your response. Can you please send me this > > > information. I can't seem to find it on the FSL website. > > > Thanks you. > > > David > > > > > > On 5/23/07, Steve Smith < [log in to unmask]> wrote: > > > > > > > > Hi, > > > > > > > > Indeed - you should run eddy_correct and then bet and then dtifit to > > > > > > > > produce the FA image - see the FDT manual pages for more details. > > > > Sorry the TBSS page isn't clearer about that - we'll fix that > > > > shortly. > > > > > > > > Cheers. > > > > > > > > > > > > > > > > On 22 May 2007, at 18:51, David Kideckel wrote: > > > > > > > > > Dear FSLers, > > > > > > > > > > I am new to TBSS and am wondering what the preprocessing steps are > > > > > to do prior to TBSS. This information seems to be different on the > > > > > > > > > FAQ section of the FSL website and the technical report on TBSS. > > > > > > > > > > Is it correct to assume that to compare 2 groups with diffusion > > > > > data, I would need to run: > > > > > Eddy correct > > > > > Registration > > > > > DTIFit > > > > > TBSS > > > > > > > > > > I am unclear as to whether I need to run BET (brain extraction > > > > > tool) and/or BEDPOST. > > > > > > > > > > Any guidance would be much appreciated. > > > > > > > > > > David Kideckel, PhD Candidate > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > --- > > > > Stephen M. Smith, Professor of Biomedical Engineering > > > > Associate Director, Oxford University FMRIB Centre > > > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > > +44 (0) 1865 222726 (fax 222717) > > > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > > ------------------------------------------------------------------------ > > > > > > > > --- > > > > > > > > > > > > > > > > > > > ------=_Part_239206_31672536.1180358543945 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline hello. i have tried your previous suggestions, and think that the problem may be in the bvecs file, but i do not know for sure. how would i check for this?
thanks again.
David

On 5/25/07, Tim Behrens <[log in to unmask]> wrote:
Hi - if you run bedpost_datacheck on your directory it might tell you what's wrong - 
for example make sure the number of entriess in bvals and bvecs is the same as the number of volumes in your data.

T

On 24 May 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the command line (using cygwin), it does not seem to be working properly. but when i use the gui, it does...
anyway, i am trying to rum dtifit from the command line and get the following message:
0 256 0 256 0 50
0 slices processed
Aborted <core dumped>

Any idea why this is happening? Did I enter something wrong?

Thanks.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
If you select dtifit from the drop down menu on the same page, all` the information about file specifications is there. Don't forget to end your bvecs and bvals files with carriage returns.

Cheers

T

On 23 May 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try to be more specific with my question. I want to do TBSS, and no tractography. So, would I do the following in this order:
Eddy Correct
BET
DTI Fit

If so, where is it in the manual specified how to create and input the bvecs and bvals files? I follow the link it gives in the FAQ section, but I can't seem to find this info.

Thanks again for your help.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you go to 
http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be clear.

T

On 23 May 2007, at 17:22, David Kideckel wrote:

Hi. Thank you for your response. Can you please send me this information. I can't seem to find it on the FSL website.
Thanks you.
David

On 5/23/07, Steve Smith < [log in to unmask]> wrote:
Hi,

Indeed - you should run eddy_correct and then bet and then dtifit to
produce the FA image - see the FDT manual pages for more details.
Sorry the TBSS page isn't clearer about that - we'll fix that shortly.

Cheers.



On 22 May 2007, at 18:51, David Kideckel wrote:

> Dear FSLers,
>
> I am new to TBSS and am wondering what the preprocessing steps are
> to do prior to TBSS. This information seems to be different on the
> FAQ section of the FSL website and the technical report on TBSS.
>
> Is it correct to assume that to compare 2 groups with diffusion
> data, I would need to run:
> Eddy correct
> Registration
> DTIFit
> TBSS
>
> I am unclear as to whether I need to run BET (brain extraction
> tool) and/or BEDPOST.
>
> Any guidance would be much appreciated.
>
> David Kideckel, PhD Candidate


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---







------=_Part_239206_31672536.1180358543945-- ========================================================================= Date: Mon, 28 May 2007 14:27:59 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Parametric stats in TBSS? In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit No. The parametric statistics of relevance here are generic and suffer from the same limitations and dangers, regardless of the implementation. Steve. On 28 May 2007, at 14:21, Marc Dubin wrote: > Is this a fundamental difference between FSL and SPM? > > On 5/28/07, Steve Smith <[log in to unmask]> wrote: Hi - please > see the original paper on TBSS. > Although the data is most likely Gaussian distributed, correcting for > multiple comparison corrections across voxels would be difficult to > get right with parametric statistics. > Cheers. > > > On 27 May 2007, at 23:04, Marc Dubin wrote: > > > Does anyone know if it is possible to use parametric stats > > in TBSS as well as the bootstrap stats in "randomise"? > > > > Best, > > Marc > > > ---------------------------------------------------------------------- > -- > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ---------------------------------------------------------------------- > -- > --- > > > > -- > 646 831 8886 > [log in to unmask] ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Mon, 28 May 2007 09:32:42 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Parametric stats in TBSS? In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_67097_14298735.1180359162628" ------=_Part_67097_14298735.1180359162628 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Okay, but does SPM implement bootstrap statistics? On 5/28/07, Steve Smith <[log in to unmask]> wrote: > > No. The parametric statistics of relevance here are generic and > suffer from the same limitations and dangers, regardless of the > implementation. > Steve. > > > > On 28 May 2007, at 14:21, Marc Dubin wrote: > > > Is this a fundamental difference between FSL and SPM? > > > > On 5/28/07, Steve Smith <[log in to unmask]> wrote: Hi - please > > see the original paper on TBSS. > > Although the data is most likely Gaussian distributed, correcting for > > multiple comparison corrections across voxels would be difficult to > > get right with parametric statistics. > > Cheers. > > > > > > On 27 May 2007, at 23:04, Marc Dubin wrote: > > > > > Does anyone know if it is possible to use parametric stats > > > in TBSS as well as the bootstrap stats in "randomise"? > > > > > > Best, > > > Marc > > > > > > ---------------------------------------------------------------------- > > -- > > --- > > Stephen M. Smith, Professor of Biomedical Engineering > > Associate Director, Oxford University FMRIB Centre > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > +44 (0) 1865 222726 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > ---------------------------------------------------------------------- > > -- > > --- > > > > > > > > -- > > 646 831 8886 > > [log in to unmask] > > > ------------------------------------------------------------------------ > > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > -- 646 831 8886 [log in to unmask] ------=_Part_67097_14298735.1180359162628 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Okay, but does SPM implement bootstrap statistics?

On 5/28/07, Steve Smith <[log in to unmask]> wrote:
No. The parametric statistics of relevance here are generic and
suffer from the same limitations and dangers, regardless of the
implementation.
Steve.



On 28 May 2007, at 14:21, Marc Dubin wrote:

> Is this a fundamental difference between FSL and SPM?
>
> On 5/28/07, Steve Smith < [log in to unmask]> wrote: Hi - please
> see the original paper on TBSS.
> Although the data is most likely Gaussian distributed, correcting for
> multiple comparison corrections across voxels would be difficult to
> get right with parametric statistics.
> Cheers.
>
>
> On 27 May 2007, at 23:04, Marc Dubin wrote:
>
> > Does anyone know if it is possible to use parametric stats
> > in TBSS as well as the bootstrap stats in "randomise"?
> >
> > Best,
> > Marc
>
>
> ----------------------------------------------------------------------
> --
> ---
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director,  Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
> +44 (0) 1865 222726  (fax 222717)
> [log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
> ----------------------------------------------------------------------
> --
> ---
>
>
>
> --
> 646 831 8886
> [log in to unmask]


------------------------------------------------------------------------

---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---



--
646 831 8886
[log in to unmask] ------=_Part_67097_14298735.1180359162628-- ========================================================================= Date: Mon, 28 May 2007 14:35:36 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Parametric stats in TBSS? In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Yes; they have a permutation-based toolbox which can be used instead of parametric statistics, in the same way that "randomise" in FSL is a permutation-based alternative to parametric statistics. In SPM it is called SnPM and is a very similar implementation (mathematically) to randomise. On 28 May 2007, at 14:32, Marc Dubin wrote: > Okay, but does SPM implement bootstrap statistics? > > On 5/28/07, Steve Smith <[log in to unmask]> wrote: No. The > parametric statistics of relevance here are generic and > suffer from the same limitations and dangers, regardless of the > implementation. > Steve. > > > > On 28 May 2007, at 14:21, Marc Dubin wrote: > > > Is this a fundamental difference between FSL and SPM? > > > > On 5/28/07, Steve Smith < [log in to unmask]> wrote: Hi - please > > see the original paper on TBSS. > > Although the data is most likely Gaussian distributed, correcting > for > > multiple comparison corrections across voxels would be difficult to > > get right with parametric statistics. > > Cheers. > > > > > > On 27 May 2007, at 23:04, Marc Dubin wrote: > > > > > Does anyone know if it is possible to use parametric stats > > > in TBSS as well as the bootstrap stats in "randomise"? > > > > > > Best, > > > Marc > > > > > > > ---------------------------------------------------------------------- > > -- > > --- > > Stephen M. Smith, Professor of Biomedical Engineering > > Associate Director, Oxford University FMRIB Centre > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > +44 (0) 1865 222726 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > ---------------------------------------------------------------------- > > -- > > --- > > > > > > > > -- > > 646 831 8886 > > [log in to unmask] > > > ---------------------------------------------------------------------- > -- > > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ---------------------------------------------------------------------- > -- > --- > > > > -- > 646 831 8886 > [log in to unmask] ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Mon, 28 May 2007 14:38:59 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Peak voxel within mask in avwstats and dilation of ROI-mask In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi, On 28 May 2007, at 09:27, Benny Liberg wrote: > Dear FSLers, > > I have two questions, and I would really appreciate if anyone could > help me > out on this rather simple matter. > > I compare 5 individuals that have done two slightly different > experiments. > So far, I've done a FFX higher level, regional analysis of both > experiments. I rerun the higher level analysis with each of the eight > masks. I then extract the peak vxl from rendered_zstats (avwstats) > and then > create a mask containing peak vxl and the 26 surrounding ones > (avwmaths). I > then run featquery with that new mask to get mean BOLD % change. Fine - though it is dangerous to use rended_zstats images - they are combined greyscale/colourmap images showing activation combined with a background image. You should use thresh_zstat images > I can't help thinking that this could be solved by running avwstats > -x with > some kind of mask option and then create the 27 vxl ROI with > avwmaths. Any > suggestions? Indeed - see below > Also, is there any way of dilating a mask? Sure - use the -dil option in avwmaths. Steve. > > Best regards, > Benny Liberg > > > > > Instead of rerunning the higher analysis with new pre-thr masks > applied > masks (8) for each anatomical strucuture, I would like to: > > 1. extract the peak vxl within a mask (avwstats) and > 2. create a new mask containing 26 vxls around the peak vxl > (avwmaths) and > extract mean %BOLD change from these 27 vxls. > > I know how to do 1 and 2 but can't figure out how to avoid ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Mon, 28 May 2007 18:31:27 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Export jpeg from FSLVIEW Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Does anyone know how to export one of the images from fslview in jpeg or gif or tiff format? Thanks, Marc ========================================================================= Date: Mon, 28 May 2007 13:44:22 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Matt <[log in to unmask]> Organization: ma-tea.com Subject: Re: Export jpeg from FSLVIEW In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Marc, There isn't a way to do that; you need to take a screen capture. Peace, Matt. -----Original Message----- From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Marc Dubin Sent: Monday, May 28, 2007 1:31 PM To: [log in to unmask] Subject: [FSL] Export jpeg from FSLVIEW Does anyone know how to export one of the images from fslview in jpeg or gif or tiff format? Thanks, Marc ========================================================================= Date: Mon, 28 May 2007 13:47:43 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Export jpeg from FSLVIEW In-Reply-To: [log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_69045_305997.1180374463784" ------=_Part_69045_305997.1180374463784 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks, Matt! On 5/28/07, Matt <[log in to unmask]> wrote: > > Marc, > > There isn't a way to do that; you need to take a screen capture. > > Peace, > > Matt. > > -----Original Message----- > From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf > Of Marc Dubin > Sent: Monday, May 28, 2007 1:31 PM > To: [log in to unmask] > Subject: [FSL] Export jpeg from FSLVIEW > > Does anyone know how to export one of the images from fslview > in jpeg or gif or tiff format? > > Thanks, > Marc > -- 646 831 8886 [log in to unmask] ------=_Part_69045_305997.1180374463784 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks, Matt!

On 5/28/07, Matt <[log in to unmask]> wrote:
Marc,

There isn't a way to do that; you need to take a screen capture.

Peace,

Matt.

-----Original Message-----
From: FSL - FMRIB's Software Library [mailto: [log in to unmask]] On Behalf
Of Marc Dubin
Sent: Monday, May 28, 2007 1:31 PM
To: [log in to unmask]
Subject: [FSL] Export jpeg from FSLVIEW

Does anyone know how to export one of the images from fslview
in jpeg or gif or tiff format?

Thanks,
Marc



--
646 831 8886
[log in to unmask] ------=_Part_69045_305997.1180374463784-- ========================================================================= Date: Mon, 28 May 2007 11:51:55 -0600 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Beau Sapach <[log in to unmask]> Subject: FSLView Error MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0022_01C7A11E.96C45AD0" This is a multi-part message in MIME format. ------=_NextPart_000_0022_01C7A11E.96C45AD0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hello everyone, I've installed FSL from the Fedora Core 4 package and am getting errors with FSLView. I'm not sure if I've completed installation properly, I honestly was expecting an RPM package but nonetheless. I've simply unzipped and un-tar-ed the FSL package and then linked it to /usr/local/fsl. Lastly I followed the "Running" instructions (for bash shell in .bashrc) from the Downloading and Installing page. Is this all there is to it? Or have I missed something important? When I run FSLView the program crashes (and disappears from the screen) when I try to do File->Open. The output in the terminal window is included below. Does this error mean anything to anyone? My .scim folder doesn't even contain a global file. Any help would be much appreciated. Beau /usr/local/fsl/bin/fslview: line 6: 1119 Aborted ${FSLDIR}/lib/fslview/bin/fslview $@ ScimInputContextPlugin() WARNING: please edit ~/.scim/global and change /DefaultConfigModule to kconfig *** glibc detected *** /usr/local/fsl/lib/fslview/bin/fslview: free(): invalid pointer: 0x08d512e0 *** ======= Backtrace: ========= /lib/libc.so.6[0xc0f09d] /lib/libc.so.6(cfree+0x90)[0xc126f0] /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0x444c871] /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so(_ZN4scim23QScimInputContext Global10initializeEv+0xfd2)[0xb65fd3a2] /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so(_ZN4scim17QScimInputContext C1Ev+0x31e)[0xb65fe3fe] /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so(_ZN22ScimInputContextPlugin 6createERK7QString+0x7e)[0xb65f75de] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN26QInputContextPluginPrivate6createERK7 QString+0x1b)[0x4ba7c5b] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN20QInputContextFactory6createERK7QStrin gP7QWidget+0x5d)[0x4ba78ed] /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so(_ZN18QMultiInputConte xt17changeInputMethodE7QString+0xc3)[0xb668c51b] /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so(_ZN18QMultiInputConte xt5slaveEv+0x43)[0xb668c6cf] /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so(_ZN18QMultiInputConte xt15setHolderWidgetEP7QWidget+0x26)[0xb668c93e] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN20QInputContextFactory6createERK7QStrin gP7QWidget+0x75)[0x4ba7905] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget18createInputContextEv+0x72)[0x 48fdb22] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget17resetInputContextEv+0x22)[0x4 8fddc2] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit7setTextERK7QString+0x22)[0x4 a7d102] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit12setMaxLengthEi+0x36)[0x4a75 fd6] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog4initEv+0x36c)[0x4b5618c] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialogC1ERK7QStringS2_P7QWidgetP Kcb+0xa2)[0x4b68732] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog15getOpenFileNameERK7QStri ngS2_P7QWidgetPKcS2_PS0_b+0x38b)[0x4b6c64b] /usr/local/fsl/lib/fslview/bin/fslview(_ZN17ApplicationWindow8fileOpenEv+0x3 e)[0x80a9392] /usr/local/fsl/lib/fslview/bin/fslview(_ZN21ApplicationWindowBase9qt_invokeE iP8QUObject+0x74)[0x8143df0] /usr/local/fsl/lib/fslview/bin/fslview(_ZN17ApplicationWindow9qt_invokeEiP8Q UObject+0x31)[0x8161ad9] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEP15QConnectio nListP8QUObject+0x16a)[0x499de2a] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEi+0xbd)[0x499 e95d] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9activatedEv+0x29)[0x4d4df29] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction18internalActivationEv+0x50)[0x 4af9a30] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9qt_invokeEiP8QUObject+0x10c)[0 x4d4de1c] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEP15QConnectio nListP8QUObject+0x16a)[0x499de2a] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal6signalERK8QVariant+0x8e)[0x4d2 97ae] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal8activateEv+0x77)[0x49bc5f7] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QPopupMenu17mouseReleaseEventEP11QMou seEvent+0x475)[0x4abd3f5] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget5eventEP6QEvent+0x3c5)[0x49db2d 5] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication14internalNotifyEP7QObjec tP6QEvent+0x9b)[0x49350fb] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication6notifyEP7QObjectP6QEvent +0x287)[0x4936757] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QETWidget19translateMouseEventEPK7_XEv ent+0xf40)[0x48cd8f0] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication15x11ProcessEventEP7_XEve nt+0x5e6)[0x48cba26] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop13processEventsEj+0x4eb)[0x 48dd6ab] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop9enterLoopEv+0x42)[0x494e67 2] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop4execEv+0x26)[0x494e536] /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication4execEv+0x1f)[0x4934c0f] /usr/local/fsl/lib/fslview/bin/fslview(main+0x11ca)[0x809e77a] /lib/libc.so.6(__libc_start_main+0xdc)[0xbbef2c] /usr/local/fsl/lib/fslview/bin/fslview(_ZN6QFrame10paintEventEP11QPaintEvent +0x3d)[0x809cba1] ======= Memory map: ======== 00101000-001ff000 r-xp 00000000 08:06 2638398 /usr/lib/libX11.so.6.2.0 001ff000-00203000 rwxp 000fe000 08:06 2638398 /usr/lib/libX11.so.6.2.0 0024c000-0026b000 r-xp 00000000 08:06 623523 /lib/libexpat.so.0.5.0 0026b000-0026d000 rwxp 0001e000 08:06 623523 /lib/libexpat.so.0.5.0 0026f000-0027e000 r-xp 00000000 08:06 2638403 /usr/lib/libXext.so.6.4.0 0027e000-0027f000 rwxp 0000e000 08:06 2638403 /usr/lib/libXext.so.6.4.0 00281000-002fe000 r-xp 00000000 08:06 2638393 /usr/lib/libfreetype.so.6.3.10 002fe000-00301000 rwxp 0007d000 08:06 2638393 /usr/lib/libfreetype.so.6.3.10 00303000-0032a000 r-xp 00000000 08:06 2638394 /usr/lib/libfontconfig.so.1.1.0 0032a000-00332000 rwxp 00027000 08:06 2638394 /usr/lib/libfontconfig.so.1.1.0 00334000-00359000 r-xp 00000000 08:06 2638395 /usr/lib/libpng12.so.0.10.0 00359000-0035a000 rwxp 00024000 08:06 2638395 /usr/lib/libpng12.so.0.10.0 0035c000-00364000 r-xp 00000000 08:06 2638399 /usr/lib/libXrender.so.1.3.0 00364000-00365000 rwxp 00007000 08:06 2638399 /usr/lib/libXrender.so.1.3.0 00367000-00370000 r-xp 00000000 08:06 2638408 /usr/lib/libXcursor.so.1.0.2 00370000-00371000 rwxp 00008000 08:06 2638408 /usr/lib/libXcursor.so.1.0.2 00373000-00375000 r-xp 00000000 08:06 2638404 /usr/lib/libXinerama.so.1.0.0 00375000-00376000 rwxp 00001000 08:06 2638404 /usr/lib/libXinerama.so.1.0.0 00378000-0037c000 r-xp 00000000 08:06 2638407 /usr/lib/libXfixes.so.3.1.0 0037c000-0037d000 rwxp 00003000 08:06 2638407 /usr/lib/libXfixes.so.3.1.0 0037f000-00382000 r-xp 00000000 08:06 2638406 /usr/lib/libXrandr.so.2.0.0 00382000-00383000 rwxp 00002000 08:06 2638406 /usr/lib/libXrandr.so.2.0.0 008db000-0092f000 r-xp 00000000 08:06 2621533 /usr/lib/libXt.so.6.0.0 0092f000-00933000 rwxp 00054000 08:06 2621533 /usr/lib/libXt.so.6.0.0 00955000-009bf000 r-xp 00000000 08:06 2638493 /usr/lib/libGL.so.1.2 009bf000-009c2000 rwxp 00069000 08:06 2638493 /usr/lib/libGL.so.1.2 009c2000-009c3000 rwxp 009c2000 00:00 0 009c7000-009c9000 r-xp 00000000 08:06 2635087 /usr/lib/libscim-x11utils-1.0.so.8.1.0 009c9000-009ca000 rwxp 00001000 08:06 2635087 /usr/lib/libscim-x11utils-1.0.so.8.1.0 ------=_NextPart_000_0022_01C7A11E.96C45AD0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hello everyone,

 

I’ve installed FSL from the Fedora Core 4 = package and am getting errors with FSLView.  I’m not sure if I’ve = completed installation properly, I honestly was expecting an RPM package but nonetheless… I’ve simply unzipped and un-tar-ed the FSL = package and then linked it to /usr/local/fsl.  Lastly I followed the = “Running” instructions (for bash shell in .bashrc) from the Downloading and = Installing page. Is this all there is to it?  Or have I missed something = important?

When I run FSLView the program crashes (and = disappears from the screen) when I try to do File->Open.  The output in the = terminal window is included below. Does this error mean anything to anyone?  My = .scim folder doesn’t even contain a global file.  Any help would be much = appreciated.

 

Beau

 

/usr/local/fsl/bin/fslview: line 6:  1119 Aborted           =       

${FSLDIR}/lib/fslview/bin/fslview = $@

ScimInputContextPlugin()

WARNING: please edit ~/.scim/global and change /DefaultConfigModule to kconfig

*** glibc detected *** /usr/local/fsl/lib/fslview/bin/fslview: free():  =

invalid pointer: 0x08d512e0 = ***

=3D=3D=3D=3D=3D=3D=3D Backtrace: = =3D=3D=3D=3D=3D=3D=3D=3D=3D

/lib/libc.so.6[0xc0f09d]

/lib/libc.so.6(cfree+0x90)[0xc126f0]=

/usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0x444c871]

/usr/lib/qt-3.3/plugins/inputmethods/libqscim.so(_ZN4s= cim23QScimInputContextGlobal10initializeEv+0xfd2)[0xb65fd3a2]<= /span>

/usr/lib/qt-3.3/plugins/inputmethods/libqscim.so(_ZN4s= cim17QScimInputContextC1Ev+0x31e)[0xb65fe3fe]

/usr/lib/qt-3.3/plugins/inputmethods/libqscim.so(_ZN22= ScimInputContextPlugin6createERK7QString+0x7e)[0xb65f75de]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN26QInputContextPl= uginPrivate6createERK7QString+0x1b)[0x4ba7c5b]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN20QInputContextFa= ctory6createERK7QStringP7QWidget+0x5d)[0x4ba78ed]

/usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so= (_ZN18QMultiInputContext17changeInputMethodE7QString+0xc3)[0xb668c51b]

/usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so= (_ZN18QMultiInputContext5slaveEv+0x43)[0xb668c6cf]

/usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so= (_ZN18QMultiInputContext15setHolderWidgetEP7QWidget+0x26)[0xb668c93e]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN20QInputContextFa= ctory6createERK7QStringP7QWidget+0x75)[0x4ba7905]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget18createI= nputContextEv+0x72)[0x48fdb22]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget17resetIn= putContextEv+0x22)[0x48fddc2]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit7setTex= tERK7QString+0x22)[0x4a7d102]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit12setMa= xLengthEi+0x36)[0x4a75fd6]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog4ini= tEv+0x36c)[0x4b5618c]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialogC1ER= K7QStringS2_P7QWidgetPKcb+0xa2)[0x4b68732]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog15ge= tOpenFileNameERK7QStringS2_P7QWidgetPKcS2_PS0_b+0x38b)[0x4b6c64b]

/usr/local/fsl/lib/fslview/bin/fslview(_ZN17Applicatio= nWindow8fileOpenEv+0x3e)[0x80a9392]

/usr/local/fsl/lib/fslview/bin/fslview(_ZN21Applicatio= nWindowBase9qt_invokeEiP8QUObject+0x74)[0x8143df0]

/usr/local/fsl/lib/fslview/bin/fslview(_ZN17Applicatio= nWindow9qt_invokeEiP8QUObject+0x31)[0x8161ad9]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activat= e_signalEP15QConnectionListP8QUObject+0x16a)[0x499de2a]=

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activat= e_signalEi+0xbd)[0x499e95d]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9activate= dEv+0x29)[0x4d4df29]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction18interna= lActivationEv+0x50)[0x4af9a30]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9qt_invok= eEiP8QUObject+0x10c)[0x4d4de1c]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activat= e_signalEP15QConnectionListP8QUObject+0x16a)[0x499de2a]=

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal6signalER= K8QVariant+0x8e)[0x4d297ae]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal8activate= Ev+0x77)[0x49bc5f7]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QPopupMenu17mou= seReleaseEventEP11QMouseEvent+0x475)[0x4abd3f5]<= /p>

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget5eventEP6= QEvent+0x3c5)[0x49db2d5]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication14i= nternalNotifyEP7QObjectP6QEvent+0x9b)[0x49350fb]=

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication6no= tifyEP7QObjectP6QEvent+0x287)[0x4936757]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QETWidget19trans= lateMouseEventEPK7_XEvent+0xf40)[0x48cd8f0]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication15x= 11ProcessEventEP7_XEvent+0x5e6)[0x48cba26]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop13pro= cessEventsEj+0x4eb)[0x48dd6ab]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop9ente= rLoopEv+0x42)[0x494e672]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop4exec= Ev+0x26)[0x494e536]

/usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication4ex= ecEv+0x1f)[0x4934c0f]

/usr/local/fsl/lib/fslview/bin/fslview(main+0x11ca)[0x= 809e77a]

/lib/libc.so.6(__libc_start_main+0xdc)[0xbbef2c]<= /o:p>

/usr/local/fsl/lib/fslview/bin/fslview(_ZN6QFrame10pai= ntEventEP11QPaintEvent+0x3d)[0x809cba1]

=3D=3D=3D=3D=3D=3D=3D Memory map: = =3D=3D=3D=3D=3D=3D=3D=3D

00101000-001ff000 r-xp 00000000 08:06 = 2638398    /usr/lib/libX11.so.6.2.0

001ff000-00203000 rwxp 000fe000 08:06 = 2638398    /usr/lib/libX11.so.6.2.0

0024c000-0026b000 r-xp 00000000 08:06 = 623523     /lib/libexpat.so.0.5.0

0026b000-0026d000 rwxp 0001e000 08:06 = 623523     /lib/libexpat.so.0.5.0

0026f000-0027e000 r-xp 00000000 08:06 = 2638403   =  /usr/lib/libXext.so.6.4.0

0027e000-0027f000 rwxp 0000e000 08:06 = 2638403    /usr/lib/libXext.so.6.4.0

00281000-002fe000 r-xp 00000000 08:06 = 2638393     = /usr/lib/libfreetype.so.6.3.10

002fe000-00301000 rwxp 0007d000 08:06 = 2638393     = /usr/lib/libfreetype.so.6.3.10

00303000-0032a000 r-xp 00000000 08:06 = 2638394     = /usr/lib/libfontconfig.so.1.1.0

0032a000-00332000 rwxp 00027000 08:06 = 2638394     = /usr/lib/libfontconfig.so.1.1.0

00334000-00359000 r-xp 00000000 08:06 = 2638395    /usr/lib/libpng12.so.0.10.0

00359000-0035a000 rwxp 00024000 08:06 = 2638395    /usr/lib/libpng12.so.0.10.0

0035c000-00364000 r-xp 00000000 08:06 = 2638399    /usr/lib/libXrender.so.1.3.0

00364000-00365000 rwxp 00007000 08:06 = 2638399    /usr/lib/libXrender.so.1.3.0

00367000-00370000 r-xp 00000000 08:06 = 2638408    /usr/lib/libXcursor.so.1.0.2

00370000-00371000 rwxp 00008000 08:06 = 2638408    /usr/lib/libXcursor.so.1.0.2

00373000-00375000 r-xp 00000000 08:06 = 2638404    /usr/lib/libXinerama.so.1.0.0

00375000-00376000 rwxp 00001000 08:06 2638404  =   /usr/lib/libXinerama.so.1.0.0

00378000-0037c000 r-xp 00000000 08:06 = 2638407    /usr/lib/libXfixes.so.3.1.0

0037c000-0037d000 rwxp 00003000 08:06 = 2638407    /usr/lib/libXfixes.so.3.1.0

0037f000-00382000 r-xp 00000000 08:06 = 2638406    /usr/lib/libXrandr.so.2.0.0

00382000-00383000 rwxp 00002000 08:06 = 2638406    /usr/lib/libXrandr.so.2.0.0

008db000-0092f000 r-xp 00000000 08:06 = 2621533    /usr/lib/libXt.so.6.0.0

0092f000-00933000 rwxp 00054000 08:06 = 2621533    /usr/lib/libXt.so.6.0.0

00955000-009bf000 r-xp 00000000 08:06 = 2638493    /usr/lib/libGL.so.1.2

009bf000-009c2000 rwxp 00069000 08:06 = 2638493    /usr/lib/libGL.so.1.2

009c2000-009c3000 rwxp 009c2000 00:00 = 0

009c7000-009c9000 r-xp 00000000 08:06 = 2635087     = /usr/lib/libscim-x11utils-1.0.so.8.1.0

009c9000-009ca000 rwxp 00001000 08:06 = 2635087     = /usr/lib/libscim-x11utils-1.0.so.8.1.0

------=_NextPart_000_0022_01C7A11E.96C45AD0-- ========================================================================= Date: Tue, 29 May 2007 03:12:43 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Johnston <[log in to unmask]> Subject: Flipping Image Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="ISO-8859-1" Hi I am running into some problem flipping images using avwswapdim. We're us= ing CATNAP DTI to create fa, adc and cmap images. We than use avwcreatehd to convert these images to analyze format, but during this conversion, the y= orientation is set to opposite for some reason. We than use avwswapdim to= re-orientate these images. fa and adc images (datatype =3D14) works perf= ectly but for cmap (datatype=3D128), avwswapdim complains about no-support for = this particular datatype. Is there another way to flip datatype 128 ? Thanks, Dave ========================================================================= Date: Tue, 29 May 2007 05:53:06 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: TBSS specks on mean_FA_skeleton.nii.gz Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Hello, My mean_FA_skeleton is very pretty, but there are several roughly 1-3 pixel green islands or specks=20 which don't seem to be part of any tract. Just wondering=20 if anyone knows how they make their way in? Best, Marc ========================================================================= Date: Tue, 29 May 2007 08:10:04 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: FSLView Error In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=WINDOWS-1252; delsp=yes; format=flowed Content-Transfer-Encoding: quoted-printable What OS are you running? On 28 May 2007, at 18:51, Beau Sapach wrote: > Hello everyone, > > > > I=92ve installed FSL from the Fedora Core 4 package and am getting =20 > errors with FSLView. I=92m not sure if I=92ve completed installation =20= > properly, I honestly was expecting an RPM package but nonetheless=85 =20= > I=92ve simply unzipped and un-tar-ed the FSL package and then linked =20= > it to /usr/local/fsl. Lastly I followed the =93Running=94 = instructions =20 > (for bash shell in .bashrc) from the Downloading and Installing =20 > page. Is this all there is to it? Or have I missed something =20 > important? > > When I run FSLView the program crashes (and disappears from the =20 > screen) when I try to do File->Open. The output in the terminal =20 > window is included below. Does this error mean anything to anyone? =20= > My .scim folder doesn=92t even contain a global file. Any help would =20= > be much appreciated. > > > > Beau > > > > /usr/local/fsl/bin/fslview: line 6: 1119 Aborted > > ${FSLDIR}/lib/fslview/bin/fslview $@ > > ScimInputContextPlugin() > > WARNING: please edit ~/.scim/global and change /DefaultConfigModule =20= > to kconfig > > *** glibc detected *** /usr/local/fsl/lib/fslview/bin/fslview: free(): > > invalid pointer: 0x08d512e0 *** > > =3D=3D=3D=3D=3D=3D=3D Backtrace: =3D=3D=3D=3D=3D=3D=3D=3D=3D > > /lib/libc.so.6[0xc0f09d] > > /lib/libc.so.6(cfree+0x90)[0xc126f0] > > /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0x444c871] > > /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so=20 > (_ZN4scim23QScimInputContextGlobal10initializeEv+0xfd2)[0xb65fd3a2] > > /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so=20 > (_ZN4scim17QScimInputContextC1Ev+0x31e)[0xb65fe3fe] > > /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so=20 > (_ZN22ScimInputContextPlugin6createERK7QString+0x7e)[0xb65f75de] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN26QInputContextPluginPrivate6createERK7QString+0x1b)[0x4ba7c5b] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x5d)[0x4ba78ed] > > /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so=20 > (_ZN18QMultiInputContext17changeInputMethodE7QString+0xc3)[0xb668c51b] > > /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so=20 > (_ZN18QMultiInputContext5slaveEv+0x43)[0xb668c6cf] > > /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so=20 > (_ZN18QMultiInputContext15setHolderWidgetEP7QWidget+0x26)[0xb668c93e] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x75)[0x4ba7905] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget18createInputContextEv=20 > +0x72)[0x48fdb22] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget17resetInputContextEv=20 > +0x22)[0x48fddc2] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit7setTextERK7QString=20 > +0x22)[0x4a7d102] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit12setMaxLengthEi=20 > +0x36)[0x4a75fd6] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog4initEv+0x36c)=20 > [0x4b5618c] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN11QFileDialogC1ERK7QStringS2_P7QWidgetPKcb+0xa2)[0x4b68732] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN11QFileDialog15getOpenFileNameERK7QStringS2_P7QWidgetPKcS2_PS0_b=20= > +0x38b)[0x4b6c64b] > > /usr/local/fsl/lib/fslview/bin/fslview=20 > (_ZN17ApplicationWindow8fileOpenEv+0x3e)[0x80a9392] > > /usr/local/fsl/lib/fslview/bin/fslview=20 > (_ZN21ApplicationWindowBase9qt_invokeEiP8QUObject+0x74)[0x8143df0] > > /usr/local/fsl/lib/fslview/bin/fslview=20 > (_ZN17ApplicationWindow9qt_invokeEiP8QUObject+0x31)[0x8161ad9] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a)=20 > [0x499de2a] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEi=20 > +0xbd)[0x499e95d] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9activatedEv+0x29)=20 > [0x4d4df29] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction18internalActivationEv=20 > +0x50)[0x4af9a30] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9qt_invokeEiP8QUObject=20 > +0x10c)[0x4d4de1c] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a)=20 > [0x499de2a] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal6signalERK8QVariant=20 > +0x8e)[0x4d297ae] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal8activateEv+0x77)=20 > [0x49bc5f7] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN10QPopupMenu17mouseReleaseEventEP11QMouseEvent+0x475)[0x4abd3f5] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget5eventEP6QEvent+0x3c5)=20 > [0x49db2d5] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN12QApplication14internalNotifyEP7QObjectP6QEvent+0x9b)[0x49350fb] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN12QApplication6notifyEP7QObjectP6QEvent+0x287)[0x4936757] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN9QETWidget19translateMouseEventEPK7_XEvent+0xf40)[0x48cd8f0] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3=20 > (_ZN12QApplication15x11ProcessEventEP7_XEvent+0x5e6)[0x48cba26] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop13processEventsEj=20 > +0x4eb)[0x48dd6ab] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop9enterLoopEv+0x42)=20 > [0x494e672] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop4execEv+0x26)=20 > [0x494e536] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication4execEv+0x1f)=20 > [0x4934c0f] > > /usr/local/fsl/lib/fslview/bin/fslview(main+0x11ca)[0x809e77a] > > /lib/libc.so.6(__libc_start_main+0xdc)[0xbbef2c] > > /usr/local/fsl/lib/fslview/bin/fslview=20 > (_ZN6QFrame10paintEventEP11QPaintEvent+0x3d)[0x809cba1] > > =3D=3D=3D=3D=3D=3D=3D Memory map: =3D=3D=3D=3D=3D=3D=3D=3D > > 00101000-001ff000 r-xp 00000000 08:06 2638398 /usr/lib/libX11.so.=20= > 6.2.0 > > 001ff000-00203000 rwxp 000fe000 08:06 2638398 /usr/lib/libX11.so.=20= > 6.2.0 > > 0024c000-0026b000 r-xp 00000000 08:06 623523 /lib/libexpat.so.=20 > 0.5.0 > > 0026b000-0026d000 rwxp 0001e000 08:06 623523 /lib/libexpat.so.=20 > 0.5.0 > > 0026f000-0027e000 r-xp 00000000 08:06 2638403 /usr/lib/=20 > libXext.so.6.4.0 > > 0027e000-0027f000 rwxp 0000e000 08:06 2638403 /usr/lib/=20 > libXext.so.6.4.0 > > 00281000-002fe000 r-xp 00000000 08:06 2638393 /usr/lib/=20 > libfreetype.so.6.3.10 > > 002fe000-00301000 rwxp 0007d000 08:06 2638393 /usr/lib/=20 > libfreetype.so.6.3.10 > > 00303000-0032a000 r-xp 00000000 08:06 2638394 /usr/lib/=20 > libfontconfig.so.1.1.0 > > 0032a000-00332000 rwxp 00027000 08:06 2638394 /usr/lib/=20 > libfontconfig.so.1.1.0 > > 00334000-00359000 r-xp 00000000 08:06 2638395 /usr/lib/=20 > libpng12.so.0.10.0 > > 00359000-0035a000 rwxp 00024000 08:06 2638395 /usr/lib/=20 > libpng12.so.0.10.0 > > 0035c000-00364000 r-xp 00000000 08:06 2638399 /usr/lib/=20 > libXrender.so.1.3.0 > > 00364000-00365000 rwxp 00007000 08:06 2638399 /usr/lib/=20 > libXrender.so.1.3.0 > > 00367000-00370000 r-xp 00000000 08:06 2638408 /usr/lib/=20 > libXcursor.so.1.0.2 > > 00370000-00371000 rwxp 00008000 08:06 2638408 /usr/lib/=20 > libXcursor.so.1.0.2 > > 00373000-00375000 r-xp 00000000 08:06 2638404 /usr/lib/=20 > libXinerama.so.1.0.0 > > 00375000-00376000 rwxp 00001000 08:06 2638404 /usr/lib/=20 > libXinerama.so.1.0.0 > > 00378000-0037c000 r-xp 00000000 08:06 2638407 /usr/lib/=20 > libXfixes.so.3.1.0 > > 0037c000-0037d000 rwxp 00003000 08:06 2638407 /usr/lib/=20 > libXfixes.so.3.1.0 > > 0037f000-00382000 r-xp 00000000 08:06 2638406 /usr/lib/=20 > libXrandr.so.2.0.0 > > 00382000-00383000 rwxp 00002000 08:06 2638406 /usr/lib/=20 > libXrandr.so.2.0.0 > > 008db000-0092f000 r-xp 00000000 08:06 2621533 /usr/lib/libXt.so.=20 > 6.0.0 > > 0092f000-00933000 rwxp 00054000 08:06 2621533 /usr/lib/libXt.so.=20 > 6.0.0 > > 00955000-009bf000 r-xp 00000000 08:06 2638493 /usr/lib/libGL.so.1.2 > > 009bf000-009c2000 rwxp 00069000 08:06 2638493 /usr/lib/libGL.so.1.2 > > 009c2000-009c3000 rwxp 009c2000 00:00 0 > > 009c7000-009c9000 r-xp 00000000 08:06 2635087 /usr/lib/libscim-=20 > x11utils-1.0.so.8.1.0 > > 009c9000-009ca000 rwxp 00001000 08:06 2635087 /usr/lib/libscim-=20 > x11utils-1.0.so.8.1.0 > > ------------------------------------------------------------------------=20= --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------=20= --- ========================================================================= Date: Tue, 29 May 2007 08:12:13 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: TBSS specks on mean_FA_skeleton.nii.gz In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Just due to noise and or complex topology. In general you shouldn't worry about them. Don't forget that TBSS will be thresholding this, normally between 2000 and 3000. You can simulate this effect by loading the skeleton into FSLView and setting the lower intensity display range to the threshold. Steve. On 29 May 2007, at 05:53, Marc Dubin wrote: > Hello, > > My mean_FA_skeleton is very pretty, but there > are several roughly 1-3 pixel green islands or specks > which don't seem to be part of any tract. Just wondering > if anyone knows how they make their way in? > > Best, > Marc ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Tue, 29 May 2007 08:13:21 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Flipping Image In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi - this sounds like an invalid Analyze datatype maybe. Why not persuade CATNAP to output it as a different datatype? Cheers. On 29 May 2007, at 03:12, David Johnston wrote: > Hi > > I am running into some problem flipping images using avwswapdim. > We're using > CATNAP DTI to create fa, adc and cmap images. We than use > avwcreatehd to > convert these images to analyze format, but during this conversion, > the y > orientation is set to opposite for some reason. We than use > avwswapdim to > re-orientate these images. fa and adc images (datatype =14) works > perfectly > but for cmap (datatype=128), avwswapdim complains about no-support > for this > particular datatype. Is there another way to flip datatype 128 ? > > Thanks, Dave ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Tue, 29 May 2007 17:45:39 +1000 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Chang <[log in to unmask]> Subject: Re: Flipping Image In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_206609_2119823.1180424739307" ------=_Part_206609_2119823.1180424739307 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi, Thanks for the response. I think datatype 128 is type ARGB. I have no problem viewing this analyze file using other image application but due to the way we're processing our batch images (CATNAP have no support for analyze format file currently), the quickest way is to generate the header file and flip the image through linux bash script. Thanks, Dave On 5/29/07, Steve Smith <[log in to unmask]> wrote: > > Hi - this sounds like an invalid Analyze datatype maybe. Why not > persuade CATNAP to output it as a different datatype? > Cheers. > > > On 29 May 2007, at 03:12, David Johnston wrote: > > > Hi > > > > I am running into some problem flipping images using avwswapdim. > > We're using > > CATNAP DTI to create fa, adc and cmap images. We than use > > avwcreatehd to > > convert these images to analyze format, but during this conversion, > > the y > > orientation is set to opposite for some reason. We than use > > avwswapdim to > > re-orientate these images. fa and adc images (datatype =14) works > > perfectly > > but for cmap (datatype=128), avwswapdim complains about no-support > > for this > > particular datatype. Is there another way to flip datatype 128 ? > > > > Thanks, Dave > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > ------=_Part_206609_2119823.1180424739307 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi,

Thanks for the response. I think datatype 128 is type ARGB. I have no problem viewing this analyze file using other image application but due to the way we're processing our batch images (CATNAP have no support for analyze format file currently), the quickest way is to generate the header file and flip the image through linux bash script.

Thanks, Dave

On 5/29/07, Steve Smith <[log in to unmask]> wrote:
Hi - this sounds like an invalid Analyze datatype maybe. Why not
persuade CATNAP to output it as a different datatype?
Cheers.


On 29 May 2007, at 03:12, David Johnston wrote:

> Hi
>
> I am running into some problem flipping images using avwswapdim.
> We're using
> CATNAP DTI to create fa, adc and cmap images. We than use
> avwcreatehd to
> convert these images to analyze format, but during this conversion,
> the y
> orientation is set to opposite for some reason. We than use
> avwswapdim to
> re-orientate these images. fa and adc images  (datatype =14) works
> perfectly
> but for cmap (datatype=128), avwswapdim complains about no-support
> for this
> particular datatype. Is there another way to flip datatype 128 ?
>
> Thanks, Dave


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---

------=_Part_206609_2119823.1180424739307-- ========================================================================= Date: Tue, 29 May 2007 08:56:09 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Mark Jenkinson <[log in to unmask]> Subject: Re: Flipping Image In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Hi, I'm afraid that FSL has never supported RGB and is unlikely to - sorry. As this type stores three 8-bit values per voxel, that will also make it very difficult to convert to another type. The easiest way is probably to write a header where the 4th dimension is 3 (or 3 * number of time points) and the datatype is 8-bit (datatype 2). This would then allow it to be read in, but would require reshaping to get into another form if that's what you want, which could be done using matlab. Hope this helps in some way. All the best, Mark David Chang wrote: > Hi, > > Thanks for the response. I think datatype 128 is type ARGB. I have no > problem viewing this analyze file using other image application but > due to the way we're processing our batch images (CATNAP have no > support for analyze format file currently), the quickest way is to > generate the header file and flip the image through linux bash script. > > Thanks, Dave > > On 5/29/07, *Steve Smith* <[log in to unmask] > > wrote: > > Hi - this sounds like an invalid Analyze datatype maybe. Why not > persuade CATNAP to output it as a different datatype? > Cheers. > > > On 29 May 2007, at 03:12, David Johnston wrote: > > > Hi > > > > I am running into some problem flipping images using avwswapdim. > > We're using > > CATNAP DTI to create fa, adc and cmap images. We than use > > avwcreatehd to > > convert these images to analyze format, but during this conversion, > > the y > > orientation is set to opposite for some reason. We than use > > avwswapdim to > > re-orientate these images. fa and adc images (datatype =14) works > > perfectly > > but for cmap (datatype=128), avwswapdim complains about no-support > > for this > > particular datatype. Is there another way to flip datatype 128 ? > > > > Thanks, Dave > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] > http://www.fmrib.ox.ac.uk/~steve > > ------------------------------------------------------------------------ > --- > > ========================================================================= Date: Tue, 29 May 2007 11:37:13 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Mark Jenkinson <[log in to unmask]> Subject: Re: Flipping Image In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.2) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit PS: Just to clarify - setting the 4th dimension to be 3 (if there is a single image) won't give you something that makes sense in FSLView or other programs. You would definitely need to do a funky reordering (e.g. in MATLAB) to produce something that looks correct. You could also do the reordering in FSL using a script it to split up the individual channels using avwroi and avwcreatehd, by making dim1=3, dim2=Nx*Ny, dim3=Nz, dim4=Nt initially, then splitting into three images (splitting the x dimension) and then replacing the headers with new headers that have the standard sizes (dim1=Nx, dim2=Ny, dim3=Nz, dim4=Nt). On 29 May 2007, at 08:56, Mark Jenkinson wrote: > Hi, > > I'm afraid that FSL has never supported RGB and is unlikely to - > sorry. > As this type stores three 8-bit values per voxel, that will also > make it very > difficult to convert to another type. The easiest way is probably > to write > a header where the 4th dimension is 3 (or 3 * number of time > points) and > the datatype is 8-bit (datatype 2). This would then allow it to be > read in, but > would require reshaping to get into another form if that's what you > want, > which could be done using matlab. > > Hope this helps in some way. > All the best, > Mark > > > David Chang wrote: > >> Hi, >> >> Thanks for the response. I think datatype 128 is type ARGB. I have >> no problem viewing this analyze file using other image application >> but due to the way we're processing our batch images (CATNAP have >> no support for analyze format file currently), the quickest way is >> to generate the header file and flip the image through linux bash >> script. >> >> Thanks, Dave >> >> On 5/29/07, *Steve Smith* <[log in to unmask] >> > wrote: >> >> Hi - this sounds like an invalid Analyze datatype maybe. Why not >> persuade CATNAP to output it as a different datatype? >> Cheers. >> >> >> On 29 May 2007, at 03:12, David Johnston wrote: >> >> > Hi >> > >> > I am running into some problem flipping images using >> avwswapdim. >> > We're using >> > CATNAP DTI to create fa, adc and cmap images. We than use >> > avwcreatehd to >> > convert these images to analyze format, but during this >> conversion, >> > the y >> > orientation is set to opposite for some reason. We than use >> > avwswapdim to >> > re-orientate these images. fa and adc images (datatype =14) >> works >> > perfectly >> > but for cmap (datatype=128), avwswapdim complains about no- >> support >> > for this >> > particular datatype. Is there another way to flip datatype >> 128 ? >> > >> > Thanks, Dave >> >> >> >> --------------------------------------------------------------------- >> --- >> --- >> Stephen M. Smith, Professor of Biomedical Engineering >> Associate Director, Oxford University FMRIB Centre >> >> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >> +44 (0) 1865 222726 (fax 222717) >> [log in to unmask] >> http://www.fmrib.ox.ac.uk/~steve >> >> >> --------------------------------------------------------------------- >> --- >> --- >> >> ========================================================================= Date: Tue, 29 May 2007 08:08:05 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: TBSS specks on mean_FA_skeleton.nii.gz In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_73900_23588160.1180440485333" ------=_Part_73900_23588160.1180440485333 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Thank you, Steve. I will give this a try.Best, Marc On 5/29/07, Steve Smith <[log in to unmask]> wrote: > > Just due to noise and or complex topology. In general you shouldn't > worry about them. Don't forget that TBSS will be thresholding this, > normally between 2000 and 3000. You can simulate this effect by > loading the skeleton into FSLView and setting the lower intensity > display range to the threshold. > > Steve. > > > On 29 May 2007, at 05:53, Marc Dubin wrote: > > > Hello, > > > > My mean_FA_skeleton is very pretty, but there > > are several roughly 1-3 pixel green islands or specks > > which don't seem to be part of any tract. Just wondering > > if anyone knows how they make their way in? > > > > Best, > > Marc > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > -- 646 831 8886 [log in to unmask] ------=_Part_73900_23588160.1180440485333 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Thank you, Steve. I will give this a try.
Best,
Marc

On 5/29/07, Steve Smith <[log in to unmask] > wrote:
Just due to noise and or complex topology. In general you shouldn't
worry about them. Don't forget that TBSS will be thresholding this,
normally between 2000 and 3000. You can simulate this effect by
loading the skeleton into FSLView and setting the lower intensity
display range to the threshold.

Steve.


On 29 May 2007, at 05:53, Marc Dubin wrote:

> Hello,
>
> My mean_FA_skeleton is very pretty, but there
> are several roughly 1-3 pixel green islands or specks
> which don't seem to be part of any tract. Just wondering
> if anyone knows how they make their way in?
>
> Best,
> Marc


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/ ~steve
------------------------------------------------------------------------
---



--
646 831 8886
[log in to unmask]
------=_Part_73900_23588160.1180440485333-- ========================================================================= Date: Tue, 29 May 2007 12:18:39 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Kideckel <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_Part_258249_9376502.1180455519181" ------=_Part_258249_9376502.1180455519181 Content-Type: multipart/alternative; boundary="----=_Part_258250_19260454.1180455519181" ------=_Part_258250_19260454.1180455519181 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. when i run dtifit, and view the FA image, it look like a non-diffusion weighted image. do these bvecs and bvals look okay? Thank you. David On 5/28/07, David Kideckel <[log in to unmask]> wrote: > > hello. i have tried your previous suggestions, and think that the problem > may be in the bvecs file, but i do not know for sure. how would i check for > this? > thanks again. > David > > On 5/25/07, Tim Behrens <[log in to unmask]> wrote: > > > > Hi - if you run bedpost_datacheck on your directory it might tell you > > what's wrong - for example make sure the number of entriess in bvals and > > bvecs is the same as the number of volumes in your data. > > > > T > > On 24 May 2007, at 00:17, David Kideckel wrote: > > > > hi. thank you for your support. once i run bet from the command line > > (using cygwin), it does not seem to be working properly. but when i use the > > gui, it does... > > anyway, i am trying to rum dtifit from the command line and get the > > following message: > > 0 256 0 256 0 50 > > 0 slices processed > > Aborted > > > > Any idea why this is happening? Did I enter something wrong? > > > > Thanks. > > David > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > If you select dtifit from the drop down menu on the same page, all` > > > the information about file specifications is there. Don't forget to end your > > > bvecs and bvals files with carriage returns. > > > Cheers > > > > > > T > > > On 23 May 2007, at 18:53, David Kideckel wrote: > > > > > > Hi. That was one of the firs things I did do. I will try to be more > > > specific with my question. I want to do TBSS, and no tractography. So, would > > > I do the following in this order: > > > Eddy Correct > > > BET > > > DTI Fit > > > > > > If so, where is it in the manual specified how to create and input the > > > bvecs and bvals files? I follow the link it gives in the FAQ section, but I > > > can't seem to find this info. > > > > > > Thanks again for your help. > > > David > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > Hi - if you go to http://www.fmrib.ox.ac.uk/fsl/fdt/index.html > > > > > > > > and select "processing pipeline", all should be clear. > > > > > > > > T > > > > > > > > On 23 May 2007, at 17:22, David Kideckel wrote: > > > > > > > > Hi. Thank you for your response. Can you please send me this > > > > information. I can't seem to find it on the FSL website. > > > > Thanks you. > > > > David > > > > > > > > On 5/23/07, Steve Smith < [log in to unmask]> wrote: > > > > > > > > > > Hi, > > > > > > > > > > Indeed - you should run eddy_correct and then bet and then dtifit > > > > > to > > > > > produce the FA image - see the FDT manual pages for more details. > > > > > Sorry the TBSS page isn't clearer about that - we'll fix that > > > > > shortly. > > > > > > > > > > Cheers. > > > > > > > > > > > > > > > > > > > > On 22 May 2007, at 18:51, David Kideckel wrote: > > > > > > > > > > > Dear FSLers, > > > > > > > > > > > > I am new to TBSS and am wondering what the preprocessing steps > > > > > are > > > > > > to do prior to TBSS. This information seems to be different on > > > > > the > > > > > > FAQ section of the FSL website and the technical report on TBSS. > > > > > > > > > > > > > > > > > Is it correct to assume that to compare 2 groups with diffusion > > > > > > data, I would need to run: > > > > > > Eddy correct > > > > > > Registration > > > > > > DTIFit > > > > > > TBSS > > > > > > > > > > > > I am unclear as to whether I need to run BET (brain extraction > > > > > > tool) and/or BEDPOST. > > > > > > > > > > > > Any guidance would be much appreciated. > > > > > > > > > > > > David Kideckel, PhD Candidate > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > --- > > > > > Stephen M. Smith, Professor of Biomedical Engineering > > > > > Associate Director, Oxford University FMRIB Centre > > > > > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > > > +44 (0) 1865 222726 (fax 222717) > > > > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > > > ------------------------------------------------------------------------ > > > > > > > > > > --- > > > > > > > > > > > > > > > > > > > > > > > > > > > > ------=_Part_258250_19260454.1180455519181 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. when i run dtifit, and view the FA image, it look like a non-diffusion weighted image. do these bvecs and bvals look okay?

Thank you.

David
On 5/28/07, David Kideckel <[log in to unmask]> wrote:
hello. i have tried your previous suggestions, and think that the problem may be in the bvecs file, but i do not know for sure. how would i check for this?
thanks again.
David


On 5/25/07, Tim Behrens <[log in to unmask]> wrote:
Hi - if you run bedpost_datacheck on your directory it might tell you what's wrong - 
for example make sure the number of entriess in bvals and bvecs is the same as the number of volumes in your data.

T

On 24 May 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the command line (using cygwin), it does not seem to be working properly. but when i use the gui, it does...
anyway, i am trying to rum dtifit from the command line and get the following message:
0 256 0 256 0 50
0 slices processed
Aborted <core dumped>

Any idea why this is happening? Did I enter something wrong?

Thanks.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
If you select dtifit from the drop down menu on the same page, all` the information about file specifications is there. Don't forget to end your bvecs and bvals files with carriage returns.

Cheers

T

On 23 May 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try to be more specific with my question. I want to do TBSS, and no tractography. So, would I do the following in this order:
Eddy Correct
BET
DTI Fit

If so, where is it in the manual specified how to create and input the bvecs and bvals files? I follow the link it gives in the FAQ section, but I can't seem to find this info.

Thanks again for your help.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you go to 
http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be clear.

T

On 23 May 2007, at 17:22, David Kideckel wrote:

Hi. Thank you for your response. Can you please send me this information. I can't seem to find it on the FSL website.
Thanks you.
David

On 5/23/07, Steve Smith < [log in to unmask]> wrote:
Hi,

Indeed - you should run eddy_correct and then bet and then dtifit to
produce the FA image - see the FDT manual pages for more details.
Sorry the TBSS page isn't clearer about that - we'll fix that shortly.

Cheers.



On 22 May 2007, at 18:51, David Kideckel wrote:

> Dear FSLers,
>
> I am new to TBSS and am wondering what the preprocessing steps are
> to do prior to TBSS. This information seems to be different on the
> FAQ section of the FSL website and the technical report on TBSS.
>
> Is it correct to assume that to compare 2 groups with diffusion
> data, I would need to run:
> Eddy correct
> Registration
> DTIFit
> TBSS
>
> I am unclear as to whether I need to run BET (brain extraction
> tool) and/or BEDPOST.
>
> Any guidance would be much appreciated.
>
> David Kideckel, PhD Candidate


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---








------=_Part_258250_19260454.1180455519181-- ------=_Part_258249_9376502.1180455519181 Content-Type: application/octet-stream; name=bvecs Content-Transfer-Encoding: base64 X-Attachment-Id: f_f2aknyhk Content-Disposition: attachment; filename="bvecs" MCAwIDAuNjUyNzU1OTMyMzA2IC0wLjU3MTc4MjIxMzcyOSAtMC43MDk2ODgyMzkwNDIgMC4wNzQy NjA2MjI3MDEgLTAuMDE2ODc1MzU3NjM5IC0wLjAwOTQxMzUwMTY1MCAwLjc3MjU5NjU2NDkzOSAw LjI2MzE1NzgzOTA3NSAwLjQwMDc1MTk3MzQ5MSAwLjgzOTgyMzY4MjQzNyAwLjAwOTQxMzQ1MDE5 NyAwLjY3NzU0ODE5NDAwMCAwLjYwOTMwMjIxOTUwNiAtMC45NzU2NjQ4NzQ0NDUgMC4wMzAyNjQ5 MTIwOTMgMC45ODQ5MzgyMzAyOTkgMC4yOTgxMTY2NzE3ODQgLTAuMjgwMTQ0NDU5MjU4IC0wLjUw OTc4MDM3Mzg2MyAtMC42ODQ1NTQ0NDgyODQgLTAuMzkzMTQxMDkxMDQ0IC0wLjg3NDA2MDU5Mzc0 MiAgLTAuNTg3ODI1MTQwMDgxICANCjAgMCAtMC43NDY0MTQ4MTQzOTAgLTAuMjQ0MjI1MzQ2ODEz IC0wLjY2OTE0MzQ4OTkyOSAwLjE0OTI3Nzg4NzgyMCAtMC45OTc5NTkwNzYwOTEgMC42ODc4OTgz NzExNTQgMC41MjMzNjI0NTkzNzIgMC45NTk0MTg1MDg1MTQgMC4xMDgzNzk3NDMwMDYgLTAuMjc2 NzU0OTQ0MTAwIC0wLjY4Nzg5ODQwMTY2MyAwLjUzODE2MTg4NDU2NSAtMC4yNjY0NDIzODY2ODYg MC4xNTI2ODEwNjc4NTggMC44MTUxNTQ2MjgyMjggLTAuMDA4ODAwNjc3OTI3IC0wLjY5MDg0Mjky ODQyNSAtMC4xMTcwNTkyNDQ0NzQgMC44NTk5NzcyMDczMzIgMC40NDA4Mjg1MTUxMzYgLTAuNzU4 MzY4MzYyNDY0IC0wLjI0ODE0MjUzNzAyOCAwLjQ3NjkxMTkzNjA2NA0KMCAwIC0wLjEyOTUxNjg2 MjYwNyAtMC43ODMyMTA3NTA3MTYgLTAuMjIwNDMwNDcyNjAxIC0wLjk4NjAwMjc3NDkwNyAtMC4w NjE1ODY1NjMwNzQgMC43MjU3NDU5NzI3NDEgMC4zNTk0MjQ5MzUwOTAgMC4xMDEzMTE3ODI0MDQg MC45MDk3NTM2NDA4NTUgMC40NjcwMTQ4NjQxNDcgLTAuNzI1NzQ1OTQ0NDkwIC0wLjUwMTMwODUx ODU4OCAtMC43NDY4MzI4MTkyMzIgLTAuMTU3MzczODk5NjU1IC0wLjU3ODQ1MjIxNjg0NiAtMC4x NzI2ODI0NTU4NjYgMC42NTg2ODIzOTU1ODIgMC45NTI3OTM4OTk2NjEgMC4wMjM3MzEyNzI0NzUg MC41ODA1NjQ3NDg4MjQgMC41MTk5MjA2NzYwMTMgMC40MTc2NjQxNzEwNTggLTAuNjUzNDY1MDc5 MzQ5DQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0K ------=_Part_258249_9376502.1180455519181 Content-Type: application/octet-stream; name=bvals Content-Transfer-Encoding: base64 X-Attachment-Id: f_f2akoeu3 Content-Disposition: attachment; filename="bvals" MCAwIDEwMDAgMTAwMCAxMDAwIDEwMDAgMTAwMCAxMDAwIDEwMDAgMTAwMCAxMDAwIDEwMDAgMTAw MCAxMDAwIDEwMDAgMTAwMCAxMDAwIDEwMDAgMTAwMCAxMDAwIDEwMDAgMTAwMCAxMDAwIDEwMDAg MTAwMA0KDQoNCg== ------=_Part_258249_9376502.1180455519181-- ========================================================================= Date: Tue, 29 May 2007 17:30:00 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Saad Jbabdi <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/alternative; boundary=Apple-Mail-20-285538127 --Apple-Mail-20-285538127 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed they should work fine. i don't know why your FA image looks like you describe. maybe you could upload your tar-gzipped data onto our server: www.fmrib.ox.ac.uk/cgi-bin/upload.cgi and send the reference number. we will have a look. cheers, saad On 29 May 2007, at 17:18, David Kideckel wrote: > hi. when i run dtifit, and view the FA image, it look like a non- > diffusion weighted image. do these bvecs and bvals look okay? > > Thank you. > > David > On 5/28/07, David Kideckel <[log in to unmask]> wrote: > hello. i have tried your previous suggestions, and think that the > problem may be in the bvecs file, but i do not know for sure. how > would i check for this? > thanks again. > David > > > On 5/25/07, Tim Behrens <[log in to unmask]> wrote: > Hi - if you run bedpost_datacheck on your directory it might tell > you what's wrong - > for example make sure the number of entriess in bvals and bvecs is > the same as the number of volumes in your data. > > T > > On 24 May 2007, at 00:17, David Kideckel wrote: > >> hi. thank you for your support. once i run bet from the command >> line (using cygwin), it does not seem to be working properly. but >> when i use the gui, it does... >> anyway, i am trying to rum dtifit from the command line and get >> the following message: >> 0 256 0 256 0 50 >> 0 slices processed >> Aborted >> >> Any idea why this is happening? Did I enter something wrong? >> >> Thanks. >> David >> >> On 5/23/07, Tim Behrens < [log in to unmask]> wrote: >> If you select dtifit from the drop down menu on the same page, >> all` the information about file specifications is there. Don't >> forget to end your bvecs and bvals files with carriage returns. >> >> Cheers >> >> T >> >> On 23 May 2007, at 18:53, David Kideckel wrote: >> >>> Hi. That was one of the firs things I did do. I will try to be >>> more specific with my question. I want to do TBSS, and no >>> tractography. So, would I do the following in this order: >>> Eddy Correct >>> BET >>> DTI Fit >>> >>> If so, where is it in the manual specified how to create and >>> input the bvecs and bvals files? I follow the link it gives in >>> the FAQ section, but I can't seem to find this info. >>> >>> Thanks again for your help. >>> David >>> >>> On 5/23/07, Tim Behrens < [log in to unmask]> wrote: >>> Hi - if you go to >>> http://www.fmrib.ox.ac.uk/fsl/fdt/index.html >>> >>> and select "processing pipeline", all should be clear. >>> >>> T >>> >>> On 23 May 2007, at 17:22, David Kideckel wrote: >>> >>>> Hi. Thank you for your response. Can you please send me this >>>> information. I can't seem to find it on the FSL website. >>>> Thanks you. >>>> David >>>> >>>> On 5/23/07, Steve Smith < [log in to unmask]> wrote: >>>> Hi, >>>> >>>> Indeed - you should run eddy_correct and then bet and then >>>> dtifit to >>>> produce the FA image - see the FDT manual pages for more details. >>>> Sorry the TBSS page isn't clearer about that - we'll fix that >>>> shortly. >>>> >>>> Cheers. >>>> >>>> >>>> >>>> On 22 May 2007, at 18:51, David Kideckel wrote: >>>> >>>> > Dear FSLers, >>>> > >>>> > I am new to TBSS and am wondering what the preprocessing steps >>>> are >>>> > to do prior to TBSS. This information seems to be different on >>>> the >>>> > FAQ section of the FSL website and the technical report on TBSS. >>>> > >>>> > Is it correct to assume that to compare 2 groups with diffusion >>>> > data, I would need to run: >>>> > Eddy correct >>>> > Registration >>>> > DTIFit >>>> > TBSS >>>> > >>>> > I am unclear as to whether I need to run BET (brain extraction >>>> > tool) and/or BEDPOST. >>>> > >>>> > Any guidance would be much appreciated. >>>> > >>>> > David Kideckel, PhD Candidate >>>> >>>> >>>> ------------------------------------------------------------------- >>>> ----- >>>> --- >>>> Stephen M. Smith, Professor of Biomedical Engineering >>>> Associate Director, Oxford University FMRIB Centre >>>> >>>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >>>> +44 (0) 1865 222726 (fax 222717) >>>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >>>> ------------------------------------------------------------------- >>>> ----- >>>> --- >>>> >>> >>> >> >> > > > > > ------------------------------------------------------------------------ --- Saad Jbabdi, Postdoctoral Research Assistant, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222545 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~saad ------------------------------------------------------------------------ --- --Apple-Mail-20-285538127 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=ISO-8859-1
they should work = fine.
i don't know why your FA image looks like you = describe.
maybe you could upload your tar-gzipped data onto = our server: www.fmrib.ox.ac.uk/c= gi-bin/upload.cgi
and send the reference number. we will = have a look.

cheers,
saad
<= BR>
On 29 May 2007, at 17:18, David Kideckel wrote:

hi. when i = run dtifit, and view the FA image, it look like a non-diffusion weighted = image. do these bvecs and bvals look okay?

Thank = you.

David
On 5/28/07, David Kideckel <[log in to unmask]> = wrote:
hello. i have tried your previous suggestions, and think that the = problem may be in the bvecs file, but i do not know for sure. how would = i check for this?
thanks again.
David


On = 5/25/07, Tim Behrens <[log in to unmask]> = wrote:
Hi - if you run bedpost_datacheck on your directory it might = tell you what's wrong -=A0
for example make sure the number of = entriess in bvals and bvecs is the same as the number of volumes in your = data.

T

On 24 May = 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the = command line (using cygwin), it does not seem to be working properly. = but when i use the gui, it does...
anyway, i am trying to rum dtifit = from the command line and get the following message:
0 256 0 256 0 = 50
0 slices processed
Aborted <core dumped>

Any idea = why this is happening? Did I enter something wrong? =

Thanks.
David

On = 5/23/07, Tim Behrens < [log in to unmask]> = wrote:
If you select dtifit from the drop down menu on the same = page, all` the information about file specifications is there. Don't = forget to end your bvecs and bvals files with carriage returns. =

=
Cheers

T

On 23 May = 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try = to be more specific with my question. I want to do TBSS, and no = tractography. So, would I do the following in this order:
Eddy = Correct
BET
DTI Fit

If so, where is it in the manual = specified how to create and input the bvecs and bvals files? I follow = the link it gives in the FAQ section, but I can't seem to find this = info.

Thanks again for your help.
David

On 5/23/07, Tim = Behrens < [log in to unmask]> = wrote:
Hi - if you go to=A0
= http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be = clear.

T
=

On 23 May 2007, at 17:22, David Kideckel wrote: =

Hi. Thank you for your response. Can = you please send me this information. I can't seem to find it on the FSL = website.
Thanks you.
David

On 5/23/07, Steve = Smith < = [log in to unmask]> wrote:
Hi,

Indeed = - you should run eddy_correct and then bet and then dtifit to =
produce the FA image - see the FDT manual pages for more = details.
Sorry the TBSS page isn't clearer about that - we'll fix = that shortly.

Cheers.



On 22 May 2007, at 18:51, = David Kideckel wrote:

> Dear FSLers,
>
> I am new = to TBSS and am wondering what the preprocessing steps are
> to do = prior to TBSS. This information seems to be different on the
> = FAQ section of the FSL website and the technical report on TBSS. =
>
> Is it correct to assume that to compare 2 groups with = diffusion
> data, I would need to run:
> Eddy = correct
> Registration
> DTIFit
> = TBSS
>
> I am unclear as to whether I need to run BET (brain = extraction
> tool) and/or BEDPOST.
>
> Any guidance = would be much appreciated.
>
> David Kideckel, PhD Candidate =


--------------------------------------------------------------= ----------
---
Stephen M. Smith, Professor of Biomedical = Engineering
Associate Director,=A0=A0Oxford University FMRIB = Centre

FMRIB, JR Hospital, Headington, Oxford=A0=A0OX3 9DU, UK =
+44 (0) 1865 222726=A0=A0(fax 222717)
[log in to unmask]=A0=A0=A0=A0= = http://www.fmrib.ox.ac.uk/~steve
----------------------------------= -------------------------------------- =
---



=




=

<bvecs>
<= SPAN>
<bvals>

Saad Jbabdi,=A0
Postdoctoral Research Assistant,=A0=A0
Oxford University FMRIB Centre

FMRIB, = JR Hospital, Headington, Oxford=A0=A0OX3 9DU, UK
+44 (0) 1865 222545=A0=A0(fax 222717)


=

= --Apple-Mail-20-285538127-- ========================================================================= Date: Tue, 29 May 2007 12:36:16 -0400 Reply-To: FSL - FMRIB's Software Library <
[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Kideckel <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_258694_28218120.1180456576920" ------=_Part_258694_28218120.1180456576920 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. thank you. the ref number is 713292. i sent the FA image after DTIFIT. Do you want me to send any other ones? David On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote: > > they should work fine. > i don't know why your FA image looks like you describe. > maybe you could upload your tar-gzipped data onto our server: > www.fmrib.ox.ac.uk/cgi-bin/upload.cgi > and send the reference number. we will have a look. > > cheers, > saad > > On 29 May 2007, at 17:18, David Kideckel wrote: > > hi. when i run dtifit, and view the FA image, it look like a non-diffusion > weighted image. do these bvecs and bvals look okay? > > Thank you. > > David > On 5/28/07, David Kideckel <[log in to unmask]> wrote: > > > > hello. i have tried your previous suggestions, and think that the > > problem may be in the bvecs file, but i do not know for sure. how would i > > check for this? > > thanks again. > > David > > > > On 5/25/07, Tim Behrens <[log in to unmask]> wrote: > > > > > > Hi - if you run bedpost_datacheck on your directory it might tell you > > > what's wrong - for example make sure the number of entriess in bvals > > > and bvecs is the same as the number of volumes in your data. > > > > > > T > > > On 24 May 2007, at 00:17, David Kideckel wrote: > > > > > > hi. thank you for your support. once i run bet from the command line > > > (using cygwin), it does not seem to be working properly. but when i use the > > > gui, it does... > > > anyway, i am trying to rum dtifit from the command line and get the > > > following message: > > > 0 256 0 256 0 50 > > > 0 slices processed > > > Aborted > > > > > > Any idea why this is happening? Did I enter something wrong? > > > > > > Thanks. > > > David > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > If you select dtifit from the drop down menu on the same page, all` > > > > the information about file specifications is there. Don't forget to end your > > > > bvecs and bvals files with carriage returns. > > > > Cheers > > > > > > > > T > > > > On 23 May 2007, at 18:53, David Kideckel wrote: > > > > > > > > Hi. That was one of the firs things I did do. I will try to be more > > > > specific with my question. I want to do TBSS, and no tractography. So, would > > > > I do the following in this order: > > > > Eddy Correct > > > > BET > > > > DTI Fit > > > > > > > > If so, where is it in the manual specified how to create and input > > > > the bvecs and bvals files? I follow the link it gives in the FAQ section, > > > > but I can't seem to find this info. > > > > > > > > Thanks again for your help. > > > > David > > > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > > > Hi - if you go to http://www.fmrib.ox.ac.uk/fsl/fdt/index.html > > > > > > > > > > and select "processing pipeline", all should be clear. > > > > > > > > > > T > > > > > > > > > > On 23 May 2007, at 17:22, David Kideckel wrote: > > > > > > > > > > Hi. Thank you for your response. Can you please send me this > > > > > information. I can't seem to find it on the FSL website. > > > > > Thanks you. > > > > > David > > > > > > > > > > On 5/23/07, Steve Smith < [log in to unmask]> wrote: > > > > > > > > > > > > Hi, > > > > > > > > > > > > Indeed - you should run eddy_correct and then bet and then > > > > > > dtifit to > > > > > > produce the FA image - see the FDT manual pages for more > > > > > > details. > > > > > > Sorry the TBSS page isn't clearer about that - we'll fix that > > > > > > shortly. > > > > > > > > > > > > Cheers. > > > > > > > > > > > > > > > > > > > > > > > > On 22 May 2007, at 18:51, David Kideckel wrote: > > > > > > > > > > > > > Dear FSLers, > > > > > > > > > > > > > > I am new to TBSS and am wondering what the preprocessing steps > > > > > > are > > > > > > > to do prior to TBSS. This information seems to be different on > > > > > > the > > > > > > > FAQ section of the FSL website and the technical report on > > > > > > TBSS. > > > > > > > > > > > > > > Is it correct to assume that to compare 2 groups with > > > > > > diffusion > > > > > > > data, I would need to run: > > > > > > > Eddy correct > > > > > > > Registration > > > > > > > DTIFit > > > > > > > TBSS > > > > > > > > > > > > > > I am unclear as to whether I need to run BET (brain extraction > > > > > > > > > > > > > tool) and/or BEDPOST. > > > > > > > > > > > > > > Any guidance would be much appreciated. > > > > > > > > > > > > > > David Kideckel, PhD Candidate > > > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > > --- > > > > > > Stephen M. Smith, Professor of Biomedical Engineering > > > > > > Associate Director, Oxford University FMRIB Centre > > > > > > > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > > > > +44 (0) 1865 222726 (fax 222717) > > > > > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > > > > ------------------------------------------------------------------------ > > > > > > > > > > > > --- > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------------- > Saad Jbabdi, > Postdoctoral Research Assistant, > Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222545 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~saad > > --------------------------------------------------------------------------- > > > > > ------=_Part_258694_28218120.1180456576920 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. thank you. the ref number is 713292. i sent the FA image after DTIFIT. Do you want me to send any other ones?
David

On 5/29/07, Saad Jbabdi < [log in to unmask]> wrote:
they should work fine.
i don't know why your FA image looks like you describe.
maybe you could upload your tar-gzipped data onto our server: www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
and send the reference number. we will have a look.

cheers,
saad

On 29 May 2007, at 17:18, David Kideckel wrote:

hi. when i run dtifit, and view the FA image, it look like a non-diffusion weighted image. do these bvecs and bvals look okay?

Thank you.

David
On 5/28/07, David Kideckel < [log in to unmask]> wrote:
hello. i have tried your previous suggestions, and think that the problem may be in the bvecs file, but i do not know for sure. how would i check for this?
thanks again.
David


On 5/25/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you run bedpost_datacheck on your directory it might tell you what's wrong - 
for example make sure the number of entriess in bvals and bvecs is the same as the number of volumes in your data.

T

On 24 May 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the command line (using cygwin), it does not seem to be working properly. but when i use the gui, it does...
anyway, i am trying to rum dtifit from the command line and get the following message:
0 256 0 256 0 50
0 slices processed
Aborted <core dumped>

Any idea why this is happening? Did I enter something wrong?

Thanks.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
If you select dtifit from the drop down menu on the same page, all` the information about file specifications is there. Don't forget to end your bvecs and bvals files with carriage returns.

Cheers

T

On 23 May 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try to be more specific with my question. I want to do TBSS, and no tractography. So, would I do the following in this order:
Eddy Correct
BET
DTI Fit

If so, where is it in the manual specified how to create and input the bvecs and bvals files? I follow the link it gives in the FAQ section, but I can't seem to find this info.

Thanks again for your help.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you go to 
http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be clear.

T

On 23 May 2007, at 17:22, David Kideckel wrote:

Hi. Thank you for your response. Can you please send me this information. I can't seem to find it on the FSL website.
Thanks you.
David

On 5/23/07, Steve Smith < [log in to unmask] > wrote:
Hi,

Indeed - you should run eddy_correct and then bet and then dtifit to
produce the FA image - see the FDT manual pages for more details.
Sorry the TBSS page isn't clearer about that - we'll fix that shortly.

Cheers.



On 22 May 2007, at 18:51, David Kideckel wrote:

> Dear FSLers,
>
> I am new to TBSS and am wondering what the preprocessing steps are
> to do prior to TBSS. This information seems to be different on the
> FAQ section of the FSL website and the technical report on TBSS.
>
> Is it correct to assume that to compare 2 groups with diffusion
> data, I would need to run:
> Eddy correct
> Registration
> DTIFit
> TBSS
>
> I am unclear as to whether I need to run BET (brain extraction
> tool) and/or BEDPOST.
>
> Any guidance would be much appreciated.
>
> David Kideckel, PhD Candidate


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---








<bvecs>
<bvals>

---------------------------------------------------------------------------
Saad Jbabdi, 
Postdoctoral Research Assistant,  
Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222545  (fax 222717)
---------------------------------------------------------------------------





------=_Part_258694_28218120.1180456576920-- ========================================================================= Date: Tue, 29 May 2007 12:40:51 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Kideckel <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_258811_25451369.1180456851354" ------=_Part_258811_25451369.1180456851354 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. i uploaded another FA image that looks slightly better, but still pretty bad. the reference number is 886189. Thanks again! David On 5/29/07, David Kideckel <[log in to unmask]> wrote: > > hi. thank you. the ref number is 713292. i sent the FA image after DTIFIT. > Do you want me to send any other ones? > David > > On 5/29/07, Saad Jbabdi < [log in to unmask]> wrote: > > > > they should work fine. > > i don't know why your FA image looks like you describe. > > maybe you could upload your tar-gzipped data onto our server: > > www.fmrib.ox.ac.uk/cgi-bin/upload.cgi > > and send the reference number. we will have a look. > > > > cheers, > > saad > > > > On 29 May 2007, at 17:18, David Kideckel wrote: > > > > hi. when i run dtifit, and view the FA image, it look like a > > non-diffusion weighted image. do these bvecs and bvals look okay? > > > > Thank you. > > > > David > > On 5/28/07, David Kideckel < [log in to unmask]> wrote: > > > > > > hello. i have tried your previous suggestions, and think that the > > > problem may be in the bvecs file, but i do not know for sure. how would i > > > check for this? > > > thanks again. > > > David > > > > > > On 5/25/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > Hi - if you run bedpost_datacheck on your directory it might tell > > > > you what's wrong - for example make sure the number of entriess in > > > > bvals and bvecs is the same as the number of volumes in your data. > > > > > > > > T > > > > On 24 May 2007, at 00:17, David Kideckel wrote: > > > > > > > > hi. thank you for your support. once i run bet from the command line > > > > (using cygwin), it does not seem to be working properly. but when i use the > > > > gui, it does... > > > > anyway, i am trying to rum dtifit from the command line and get the > > > > following message: > > > > 0 256 0 256 0 50 > > > > 0 slices processed > > > > Aborted > > > > > > > > Any idea why this is happening? Did I enter something wrong? > > > > > > > > Thanks. > > > > David > > > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > > > If you select dtifit from the drop down menu on the same page, > > > > > all` the information about file specifications is there. Don't forget to end > > > > > your bvecs and bvals files with carriage returns. > > > > > Cheers > > > > > > > > > > T > > > > > On 23 May 2007, at 18:53, David Kideckel wrote: > > > > > > > > > > Hi. That was one of the firs things I did do. I will try to be > > > > > more specific with my question. I want to do TBSS, and no tractography. So, > > > > > would I do the following in this order: > > > > > Eddy Correct > > > > > BET > > > > > DTI Fit > > > > > > > > > > If so, where is it in the manual specified how to create and input > > > > > the bvecs and bvals files? I follow the link it gives in the FAQ section, > > > > > but I can't seem to find this info. > > > > > > > > > > Thanks again for your help. > > > > > David > > > > > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > > > > > Hi - if you go to http://www.fmrib.ox.ac.uk/fsl/fdt/index.html > > > > > > > > > > > > and select "processing pipeline", all should be clear. > > > > > > > > > > > > T > > > > > > > > > > > > On 23 May 2007, at 17:22, David Kideckel wrote: > > > > > > > > > > > > Hi. Thank you for your response. Can you please send me this > > > > > > information. I can't seem to find it on the FSL website. > > > > > > Thanks you. > > > > > > David > > > > > > > > > > > > On 5/23/07, Steve Smith < [log in to unmask] > wrote: > > > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > Indeed - you should run eddy_correct and then bet and then > > > > > > > dtifit to > > > > > > > produce the FA image - see the FDT manual pages for more > > > > > > > details. > > > > > > > Sorry the TBSS page isn't clearer about that - we'll fix that > > > > > > > shortly. > > > > > > > > > > > > > > Cheers. > > > > > > > > > > > > > > > > > > > > > > > > > > > > On 22 May 2007, at 18:51, David Kideckel wrote: > > > > > > > > > > > > > > > Dear FSLers, > > > > > > > > > > > > > > > > I am new to TBSS and am wondering what the preprocessing > > > > > > > steps are > > > > > > > > to do prior to TBSS. This information seems to be different > > > > > > > on the > > > > > > > > FAQ section of the FSL website and the technical report on > > > > > > > TBSS. > > > > > > > > > > > > > > > > Is it correct to assume that to compare 2 groups with > > > > > > > diffusion > > > > > > > > data, I would need to run: > > > > > > > > Eddy correct > > > > > > > > Registration > > > > > > > > DTIFit > > > > > > > > TBSS > > > > > > > > > > > > > > > > I am unclear as to whether I need to run BET (brain > > > > > > > extraction > > > > > > > > tool) and/or BEDPOST. > > > > > > > > > > > > > > > > Any guidance would be much appreciated. > > > > > > > > > > > > > > > > David Kideckel, PhD Candidate > > > > > > > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > > > --- > > > > > > > Stephen M. Smith, Professor of Biomedical Engineering > > > > > > > Associate Director, Oxford University FMRIB Centre > > > > > > > > > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > > > > > +44 (0) 1865 222726 (fax 222717) > > > > > > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > > > > > ------------------------------------------------------------------------ > > > > > > > > > > > > > > --- > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------------- > > Saad Jbabdi, > > Postdoctoral Research Assistant, > > Oxford University FMRIB Centre > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > +44 (0) 1865 222545 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/~saad > > --------------------------------------------------------------------------- > > > > > > > > > > > > > ------=_Part_258811_25451369.1180456851354 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. i uploaded another FA image that looks slightly better, but still pretty bad. the reference number is 886189.
Thanks again!
David

On 5/29/07, David Kideckel <[log in to unmask]> wrote:
hi. thank you. the ref number is 713292. i sent the FA image after DTIFIT. Do you want me to send any other ones?
David


On 5/29/07, Saad Jbabdi < [log in to unmask]> wrote:
they should work fine.
i don't know why your FA image looks like you describe.
maybe you could upload your tar-gzipped data onto our server: www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
and send the reference number. we will have a look.

cheers,
saad

On 29 May 2007, at 17:18, David Kideckel wrote:

hi. when i run dtifit, and view the FA image, it look like a non-diffusion weighted image. do these bvecs and bvals look okay?

Thank you.

David
On 5/28/07, David Kideckel < [log in to unmask]> wrote:
hello. i have tried your previous suggestions, and think that the problem may be in the bvecs file, but i do not know for sure. how would i check for this?
thanks again.
David


On 5/25/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you run bedpost_datacheck on your directory it might tell you what's wrong - 
for example make sure the number of entriess in bvals and bvecs is the same as the number of volumes in your data.

T

On 24 May 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the command line (using cygwin), it does not seem to be working properly. but when i use the gui, it does...
anyway, i am trying to rum dtifit from the command line and get the following message:
0 256 0 256 0 50
0 slices processed
Aborted <core dumped>

Any idea why this is happening? Did I enter something wrong?

Thanks.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
If you select dtifit from the drop down menu on the same page, all` the information about file specifications is there. Don't forget to end your bvecs and bvals files with carriage returns.

Cheers

T

On 23 May 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try to be more specific with my question. I want to do TBSS, and no tractography. So, would I do the following in this order:
Eddy Correct
BET
DTI Fit

If so, where is it in the manual specified how to create and input the bvecs and bvals files? I follow the link it gives in the FAQ section, but I can't seem to find this info.

Thanks again for your help.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you go to 
http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be clear.

T

On 23 May 2007, at 17:22, David Kideckel wrote:

Hi. Thank you for your response. Can you please send me this information. I can't seem to find it on the FSL website.
Thanks you.
David

On 5/23/07, Steve Smith < [log in to unmask] > wrote:
Hi,

Indeed - you should run eddy_correct and then bet and then dtifit to
produce the FA image - see the FDT manual pages for more details.
Sorry the TBSS page isn't clearer about that - we'll fix that shortly.

Cheers.



On 22 May 2007, at 18:51, David Kideckel wrote:

> Dear FSLers,
>
> I am new to TBSS and am wondering what the preprocessing steps are
> to do prior to TBSS. This information seems to be different on the
> FAQ section of the FSL website and the technical report on TBSS.
>
> Is it correct to assume that to compare 2 groups with diffusion
> data, I would need to run:
> Eddy correct
> Registration
> DTIFit
> TBSS
>
> I am unclear as to whether I need to run BET (brain extraction
> tool) and/or BEDPOST.
>
> Any guidance would be much appreciated.
>
> David Kideckel, PhD Candidate


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---








<bvecs>
<bvals>

---------------------------------------------------------------------------
Saad Jbabdi, 
Postdoctoral Research Assistant,  
Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222545  (fax 222717)
---------------------------------------------------------------------------






------=_Part_258811_25451369.1180456851354-- ========================================================================= Date: Tue, 29 May 2007 18:05:00 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Saad Jbabdi <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/alternative; boundary=Apple-Mail-21-287637253 --Apple-Mail-21-287637253 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed what is the difference between the two FA images you sent? the most likely explanation for your results is that the data you have do not correspond to the bvecs you provide to dtifit. could you upload the actual data (not the result of dtifit) ? saad. On 29 May 2007, at 17:40, David Kideckel wrote: > hi. i uploaded another FA image that looks slightly better, but > still pretty bad. the reference number is 886189. > Thanks again! > David > > On 5/29/07, David Kideckel <[log in to unmask]> wrote: > hi. thank you. the ref number is 713292. i sent the FA image after > DTIFIT. Do you want me to send any other ones? > David > > > On 5/29/07, Saad Jbabdi < [log in to unmask]> wrote: > they should work fine. > i don't know why your FA image looks like you describe. > maybe you could upload your tar-gzipped data onto our server: > www.fmrib.ox.ac.uk/cgi-bin/upload.cgi > and send the reference number. we will have a look. > > cheers, > saad > > On 29 May 2007, at 17:18, David Kideckel wrote: > >> hi. when i run dtifit, and view the FA image, it look like a non- >> diffusion weighted image. do these bvecs and bvals look okay? >> >> Thank you. >> >> David >> On 5/28/07, David Kideckel < [log in to unmask]> wrote: >> hello. i have tried your previous suggestions, and think that the >> problem may be in the bvecs file, but i do not know for sure. how >> would i check for this? >> thanks again. >> David >> >> >> On 5/25/07, Tim Behrens < [log in to unmask]> wrote: >> Hi - if you run bedpost_datacheck on your directory it might tell >> you what's wrong - >> for example make sure the number of entriess in bvals and bvecs is >> the same as the number of volumes in your data. >> >> T >> >> On 24 May 2007, at 00:17, David Kideckel wrote: >> >>> hi. thank you for your support. once i run bet from the command >>> line (using cygwin), it does not seem to be working properly. but >>> when i use the gui, it does... >>> anyway, i am trying to rum dtifit from the command line and get >>> the following message: >>> 0 256 0 256 0 50 >>> 0 slices processed >>> Aborted >>> >>> Any idea why this is happening? Did I enter something wrong? >>> >>> Thanks. >>> David >>> >>> On 5/23/07, Tim Behrens < [log in to unmask]> wrote: >>> If you select dtifit from the drop down menu on the same page, >>> all` the information about file specifications is there. Don't >>> forget to end your bvecs and bvals files with carriage returns. >>> >>> Cheers >>> >>> T >>> >>> On 23 May 2007, at 18:53, David Kideckel wrote: >>> >>>> Hi. That was one of the firs things I did do. I will try to be >>>> more specific with my question. I want to do TBSS, and no >>>> tractography. So, would I do the following in this order: >>>> Eddy Correct >>>> BET >>>> DTI Fit >>>> >>>> If so, where is it in the manual specified how to create and >>>> input the bvecs and bvals files? I follow the link it gives in >>>> the FAQ section, but I can't seem to find this info. >>>> >>>> Thanks again for your help. >>>> David >>>> >>>> On 5/23/07, Tim Behrens < [log in to unmask]> wrote: >>>> Hi - if you go to >>>> http://www.fmrib.ox.ac.uk/fsl/fdt/index.html >>>> >>>> and select "processing pipeline", all should be clear. >>>> >>>> T >>>> >>>> On 23 May 2007, at 17:22, David Kideckel wrote: >>>> >>>>> Hi. Thank you for your response. Can you please send me this >>>>> information. I can't seem to find it on the FSL website. >>>>> Thanks you. >>>>> David >>>>> >>>>> On 5/23/07, Steve Smith < [log in to unmask] > wrote: >>>>> Hi, >>>>> >>>>> Indeed - you should run eddy_correct and then bet and then >>>>> dtifit to >>>>> produce the FA image - see the FDT manual pages for more details. >>>>> Sorry the TBSS page isn't clearer about that - we'll fix that >>>>> shortly. >>>>> >>>>> Cheers. >>>>> >>>>> >>>>> >>>>> On 22 May 2007, at 18:51, David Kideckel wrote: >>>>> >>>>> > Dear FSLers, >>>>> > >>>>> > I am new to TBSS and am wondering what the preprocessing >>>>> steps are >>>>> > to do prior to TBSS. This information seems to be different >>>>> on the >>>>> > FAQ section of the FSL website and the technical report on TBSS. >>>>> > >>>>> > Is it correct to assume that to compare 2 groups with diffusion >>>>> > data, I would need to run: >>>>> > Eddy correct >>>>> > Registration >>>>> > DTIFit >>>>> > TBSS >>>>> > >>>>> > I am unclear as to whether I need to run BET (brain extraction >>>>> > tool) and/or BEDPOST. >>>>> > >>>>> > Any guidance would be much appreciated. >>>>> > >>>>> > David Kideckel, PhD Candidate >>>>> >>>>> >>>>> ------------------------------------------------------------------ >>>>> ------ >>>>> --- >>>>> Stephen M. Smith, Professor of Biomedical Engineering >>>>> Associate Director, Oxford University FMRIB Centre >>>>> >>>>> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >>>>> +44 (0) 1865 222726 (fax 222717) >>>>> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >>>>> ------------------------------------------------------------------ >>>>> ------ >>>>> --- >>>>> >>>> >>>> >>> >>> >> >> >> >> >> > > ---------------------------------------------------------------------- > ----- > Saad Jbabdi, > Postdoctoral Research Assistant, > Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222545 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~saad > ---------------------------------------------------------------------- > ----- > > > > > > ------------------------------------------------------------------------ --- Saad Jbabdi, Postdoctoral Research Assistant, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222545 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~saad ------------------------------------------------------------------------ --- --Apple-Mail-21-287637253 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=ISO-8859-1
what is the difference = between the two FA images you sent?

the most likely explanation = for your results is that the data you have do not correspond to the = bvecs you provide to dtifit.
could you upload the actual data = (not the result of dtifit) ?

saad.


On 29 May 2007, at = 17:40, David Kideckel wrote:

hi. i = uploaded another FA image that looks slightly better, but still pretty = bad. the reference number is 886189.
Thanks = again!
David

On 5/29/07, David Kideckel <[log in to unmask]> = wrote:
hi. thank you. the ref number is 713292. i sent the FA image = after DTIFIT. Do you want me to send any other ones?
David


On = 5/29/07, Saad Jbabdi < [log in to unmask]> = wrote:
they should work fine.
i don't know why your = FA image looks like you describe.
maybe you could upload your = tar-gzipped data onto our server: = www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
and send the = reference number. we will have a = look.

cheers,
saad

<= SPAN>
On 29 May 2007, at 17:18, David Kideckel wrote: =

hi. when i = run dtifit, and view the FA image, it look like a non-diffusion weighted = image. do these bvecs and bvals look okay?

Thank = you.

David
On 5/28/07, David Kideckel < = [log in to unmask]> wrote:
hello. i have = tried your previous suggestions, and think that the problem may be in = the bvecs file, but i do not know for sure. how would i check for this? =
thanks again.
David


On 5/25/07, Tim = Behrens < [log in to unmask]> = wrote:
Hi - if you run bedpost_datacheck on your directory it might = tell you what's wrong -=A0
for example make sure the number of = entriess in bvals and bvecs is the same as the number of volumes in your = data.

T

On 24 May = 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the = command line (using cygwin), it does not seem to be working properly. = but when i use the gui, it does...
anyway, i am trying to rum dtifit = from the command line and get the following message:
0 256 0 256 0 = 50
0 slices processed
Aborted <core dumped>

Any idea = why this is happening? Did I enter something wrong? =

Thanks.
David

On = 5/23/07, Tim Behrens < [log in to unmask]> = wrote:
If you select dtifit from the drop down menu on the same = page, all` the information about file specifications is there. Don't = forget to end your bvecs and bvals files with carriage returns. =

= Cheers

T

On 23 May = 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try = to be more specific with my question. I want to do TBSS, and no = tractography. So, would I do the following in this order:
Eddy = Correct
BET
DTI Fit

If so, where is it in the manual = specified how to create and input the bvecs and bvals files? I follow = the link it gives in the FAQ section, but I can't seem to find this = info.

Thanks again for your help.
David

On 5/23/07, Tim = Behrens < [log in to unmask]> = wrote:
Hi - if you go to=A0
= http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be = clear.

T
=

On 23 May 2007, at 17:22, David Kideckel wrote: =

Hi. Thank you for your response. Can = you please send me this information. I can't seem to find it on the FSL = website.
Thanks you.
David

On 5/23/07, Steve = Smith < = [log in to unmask] > wrote:
Hi,

Indeed = - you should run eddy_correct and then bet and then dtifit to =
produce the FA image - see the FDT manual pages for more = details.
Sorry the TBSS page isn't clearer about that - we'll fix = that shortly.

Cheers.



On 22 May 2007, at 18:51, = David Kideckel wrote:

> Dear FSLers,
>
> I am new = to TBSS and am wondering what the preprocessing steps are
> to do = prior to TBSS. This information seems to be different on the
> = FAQ section of the FSL website and the technical report on TBSS. =
>
> Is it correct to assume that to compare 2 groups with = diffusion
> data, I would need to run:
> Eddy = correct
> Registration
> DTIFit
> = TBSS
>
> I am unclear as to whether I need to run BET (brain = extraction
> tool) and/or BEDPOST.
>
> Any guidance = would be much appreciated.
>
> David Kideckel, PhD Candidate =


--------------------------------------------------------------= ----------
---
Stephen M. Smith, Professor of Biomedical = Engineering
Associate Director,=A0=A0Oxford University FMRIB = Centre

FMRIB, JR Hospital, Headington, Oxford=A0=A0OX3 9DU, UK =
+44 (0) 1865 222726=A0=A0(fax 222717)
[log in to unmask]=A0=A0=A0= =A0 = http://www.fmrib.ox.ac.uk/~steve
----------------------------------= -------------------------------------- =
---



=




=

<bvecs><= /DIV>
<bvals>

--------------------------------------------------------------------= -------
Saad Jbabdi,=A0
Postdoctoral Research Assistant,=A0=A0
=
Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford=A0=A0OX3 9DU, = UK
+44 (0) 1865 = 222545=A0=A0(fax 222717)
--------------------------------------------------------------------= -------



=


=


Saad Jbabdi,=A0
Postdoctoral Research Assistant,=A0=A0
Oxford University FMRIB Centre

FMRIB, = JR Hospital, Headington, Oxford=A0=A0OX3 9DU, UK
+44 (0) 1865 222545=A0=A0(fax 222717)


=

= --Apple-Mail-21-287637253-- ========================================================================= Date: Tue, 29 May 2007 13:16:40 -0400 Reply-To: FSL - FMRIB's Software Library <
[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Kideckel <[log in to unmask]> Subject: Re: TBSS Pipeline In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_259379_2445206.1180459000179" ------=_Part_259379_2445206.1180459000179 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hi, the ref number is 913750 (it seems to be taking a while to upload due to the file size) . it is the 4D volume. I wonder if it is the fact that I made my headers with SPM99, then created a 4D image using FSL commands (avwmerge -a and then ran the eddy correct script: eddy_correct . This shouldn't be a problem though, right? Thanks again! David On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote: > > what is the difference between the two FA images you sent? > > the most likely explanation for your results is that the data you have do > not correspond to the bvecs you provide to dtifit. > could you upload the actual data (not the result of dtifit) ? > > saad. > > > On 29 May 2007, at 17:40, David Kideckel wrote: > > hi. i uploaded another FA image that looks slightly better, but still > pretty bad. the reference number is 886189. > Thanks again! > David > > On 5/29/07, David Kideckel <[log in to unmask]> wrote: > > > > hi. thank you. the ref number is 713292. i sent the FA image after > > DTIFIT. Do you want me to send any other ones? > > David > > > > On 5/29/07, Saad Jbabdi < [log in to unmask]> wrote: > > > > > > they should work fine. > > > i don't know why your FA image looks like you describe. > > > maybe you could upload your tar-gzipped data onto our server: > > > www.fmrib.ox.ac.uk/cgi-bin/upload.cgi > > > and send the reference number. we will have a look. > > > > > > cheers, > > > saad > > > > > > On 29 May 2007, at 17:18, David Kideckel wrote: > > > > > > hi. when i run dtifit, and view the FA image, it look like a > > > non-diffusion weighted image. do these bvecs and bvals look okay? > > > > > > Thank you. > > > > > > David > > > On 5/28/07, David Kideckel < [log in to unmask]> wrote: > > > > > > > > hello. i have tried your previous suggestions, and think that the > > > > problem may be in the bvecs file, but i do not know for sure. how would i > > > > check for this? > > > > thanks again. > > > > David > > > > > > > > On 5/25/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > > > Hi - if you run bedpost_datacheck on your directory it might tell > > > > > you what's wrong - for example make sure the number of entriess > > > > > in bvals and bvecs is the same as the number of volumes in your data. > > > > > > > > > > T > > > > > On 24 May 2007, at 00:17, David Kideckel wrote: > > > > > > > > > > hi. thank you for your support. once i run bet from the command > > > > > line (using cygwin), it does not seem to be working properly. but when i use > > > > > the gui, it does... > > > > > anyway, i am trying to rum dtifit from the command line and get > > > > > the following message: > > > > > 0 256 0 256 0 50 > > > > > 0 slices processed > > > > > Aborted > > > > > > > > > > Any idea why this is happening? Did I enter something wrong? > > > > > > > > > > Thanks. > > > > > David > > > > > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > > > > > If you select dtifit from the drop down menu on the same page, > > > > > > all` the information about file specifications is there. Don't forget to end > > > > > > your bvecs and bvals files with carriage returns. > > > > > > Cheers > > > > > > > > > > > > T > > > > > > On 23 May 2007, at 18:53, David Kideckel wrote: > > > > > > > > > > > > Hi. That was one of the firs things I did do. I will try to be > > > > > > more specific with my question. I want to do TBSS, and no tractography. So, > > > > > > would I do the following in this order: > > > > > > Eddy Correct > > > > > > BET > > > > > > DTI Fit > > > > > > > > > > > > If so, where is it in the manual specified how to create and > > > > > > input the bvecs and bvals files? I follow the link it gives in the FAQ > > > > > > section, but I can't seem to find this info. > > > > > > > > > > > > Thanks again for your help. > > > > > > David > > > > > > > > > > > > On 5/23/07, Tim Behrens < [log in to unmask]> wrote: > > > > > > > > > > > > > > Hi - if you go to http://www.fmrib.ox.ac.uk/fsl/fdt/index.html > > > > > > > > > > > > > > and select "processing pipeline", all should be clear. > > > > > > > > > > > > > > T > > > > > > > > > > > > > > On 23 May 2007, at 17:22, David Kideckel wrote: > > > > > > > > > > > > > > Hi. Thank you for your response. Can you please send me this > > > > > > > information. I can't seem to find it on the FSL website. > > > > > > > Thanks you. > > > > > > > David > > > > > > > > > > > > > > On 5/23/07, Steve Smith < [log in to unmask] > wrote: > > > > > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > Indeed - you should run eddy_correct and then bet and then > > > > > > > > dtifit to > > > > > > > > produce the FA image - see the FDT manual pages for more > > > > > > > > details. > > > > > > > > Sorry the TBSS page isn't clearer about that - we'll fix > > > > > > > > that shortly. > > > > > > > > > > > > > > > > Cheers. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > On 22 May 2007, at 18:51, David Kideckel wrote: > > > > > > > > > > > > > > > > > Dear FSLers, > > > > > > > > > > > > > > > > > > I am new to TBSS and am wondering what the preprocessing > > > > > > > > steps are > > > > > > > > > to do prior to TBSS. This information seems to be > > > > > > > > different on the > > > > > > > > > FAQ section of the FSL website and the technical report on > > > > > > > > TBSS. > > > > > > > > > > > > > > > > > > Is it correct to assume that to compare 2 groups with > > > > > > > > diffusion > > > > > > > > > data, I would need to run: > > > > > > > > > Eddy correct > > > > > > > > > Registration > > > > > > > > > DTIFit > > > > > > > > > TBSS > > > > > > > > > > > > > > > > > > I am unclear as to whether I need to run BET (brain > > > > > > > > extraction > > > > > > > > > tool) and/or BEDPOST. > > > > > > > > > > > > > > > > > > Any guidance would be much appreciated. > > > > > > > > > > > > > > > > > > David Kideckel, PhD Candidate > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > > > > --- > > > > > > > > Stephen M. Smith, Professor of Biomedical Engineering > > > > > > > > Associate Director, Oxford University FMRIB Centre > > > > > > > > > > > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > > > > > > +44 (0) 1865 222726 (fax 222717) > > > > > > > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > > > > > > > > ------------------------------------------------------------------------ > > > > > > > > > > > > > > > > --- > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------------- > > > Saad Jbabdi, > > > Postdoctoral Research Assistant, > > > Oxford University FMRIB Centre > > > > > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > > > +44 (0) 1865 222545 (fax 222717) > > > [log in to unmask] http://www.fmrib.ox.ac.uk/~saad > > > --------------------------------------------------------------------------- > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------------- > Saad Jbabdi, > Postdoctoral Research Assistant, > Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222545 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~saad > > --------------------------------------------------------------------------- > > > > > ------=_Part_259379_2445206.1180459000179 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline hi, the ref number is 913750 (it seems to be taking a while to upload due to the file size) . it is the 4D volume. I wonder if it is the fact that I made my headers with SPM99, then created a 4D image using FSL commands (avwmerge -a <output> <input> and then ran the eddy correct script: eddy_correct <input> <output>. This shouldn't be a problem though, right?
Thanks again!
David

On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote:
what is the difference between the two FA images you sent?

the most likely explanation for your results is that the data you have do not correspond to the bvecs you provide to dtifit.
could you upload the actual data (not the result of dtifit) ?

saad.


On 29 May 2007, at 17:40, David Kideckel wrote:

hi. i uploaded another FA image that looks slightly better, but still pretty bad. the reference number is 886189.
Thanks again!
David

On 5/29/07, David Kideckel <[log in to unmask]> wrote:
hi. thank you. the ref number is 713292. i sent the FA image after DTIFIT. Do you want me to send any other ones?
David


On 5/29/07, Saad Jbabdi < [log in to unmask]> wrote:
they should work fine.
i don't know why your FA image looks like you describe.
maybe you could upload your tar-gzipped data onto our server: www.fmrib.ox.ac.uk/cgi-bin/upload.cgi
and send the reference number. we will have a look.

cheers,
saad

On 29 May 2007, at 17:18, David Kideckel wrote:

hi. when i run dtifit, and view the FA image, it look like a non-diffusion weighted image. do these bvecs and bvals look okay?

Thank you.

David
On 5/28/07, David Kideckel < [log in to unmask] > wrote:
hello. i have tried your previous suggestions, and think that the problem may be in the bvecs file, but i do not know for sure. how would i check for this?
thanks again.
David


On 5/25/07, Tim Behrens < [log in to unmask]> wrote:
Hi - if you run bedpost_datacheck on your directory it might tell you what's wrong - 
for example make sure the number of entriess in bvals and bvecs is the same as the number of volumes in your data.

T

On 24 May 2007, at 00:17, David Kideckel wrote:

hi. thank you for your support. once i run bet from the command line (using cygwin), it does not seem to be working properly. but when i use the gui, it does...
anyway, i am trying to rum dtifit from the command line and get the following message:
0 256 0 256 0 50
0 slices processed
Aborted <core dumped>

Any idea why this is happening? Did I enter something wrong?

Thanks.
David

On 5/23/07, Tim Behrens < [log in to unmask]> wrote:
If you select dtifit from the drop down menu on the same page, all` the information about file specifications is there. Don't forget to end your bvecs and bvals files with carriage returns.

Cheers

T

On 23 May 2007, at 18:53, David Kideckel wrote:

Hi. That was one of the firs things I did do. I will try to be more specific with my question. I want to do TBSS, and no tractography. So, would I do the following in this order:
Eddy Correct
BET
DTI Fit

If so, where is it in the manual specified how to create and input the bvecs and bvals files? I follow the link it gives in the FAQ section, but I can't seem to find this info.

Thanks again for your help.
David

On 5/23/07, Tim Behrens <[log in to unmask]" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)"> [log in to unmask]> wrote:
Hi - if you go to 
http://www.fmrib.ox.ac.uk/fsl/fdt/index.html

and select "processing pipeline", all should be clear.

T

On 23 May 2007, at 17:22, David Kideckel wrote:

Hi. Thank you for your response. Can you please send me this information. I can't seem to find it on the FSL website.
Thanks you.
David

On 5/23/07, Steve Smith < [log in to unmask] > wrote:
Hi,

Indeed - you should run eddy_correct and then bet and then dtifit to
produce the FA image - see the FDT manual pages for more details.
Sorry the TBSS page isn't clearer about that - we'll fix that shortly.

Cheers.



On 22 May 2007, at 18:51, David Kideckel wrote:

> Dear FSLers,
>
> I am new to TBSS and am wondering what the preprocessing steps are
> to do prior to TBSS. This information seems to be different on the
> FAQ section of the FSL website and the technical report on TBSS.
>
> Is it correct to assume that to compare 2 groups with diffusion
> data, I would need to run:
> Eddy correct
> Registration
> DTIFit
> TBSS
>
> I am unclear as to whether I need to run BET (brain extraction
> tool) and/or BEDPOST.
>
> Any guidance would be much appreciated.
>
> David Kideckel, PhD Candidate


------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]     http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---








<bvecs>
<bvals>

---------------------------------------------------------------------------
Saad Jbabdi, 
Postdoctoral Research Assistant,  
Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222545  (fax 222717)
---------------------------------------------------------------------------







---------------------------------------------------------------------------
Saad Jbabdi, 
Postdoctoral Research Assistant,  
Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222545   (fax 222717)
---------------------------------------------------------------------------





------=_Part_259379_2445206.1180459000179-- ========================================================================= Date: Tue, 29 May 2007 12:55:27 -0600 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Beau Sapach <[log in to unmask]> Subject: Re: FSLView Error In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Woops, sorry. It's Fedora Core 6. -----Original Message----- From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Steve Smith Sent: Tuesday, May 29, 2007 1:10 AM To: [log in to unmask] Subject: Re: [FSL] FSLView Error What OS are you running? On 28 May 2007, at 18:51, Beau Sapach wrote: > Hello everyone, > > > > I've installed FSL from the Fedora Core 4 package and am getting > errors with FSLView. I'm not sure if I've completed installation > properly, I honestly was expecting an RPM package but nonetheless. > I've simply unzipped and un-tar-ed the FSL package and then linked > it to /usr/local/fsl. Lastly I followed the "Running" instructions > (for bash shell in .bashrc) from the Downloading and Installing > page. Is this all there is to it? Or have I missed something > important? > > When I run FSLView the program crashes (and disappears from the > screen) when I try to do File->Open. The output in the terminal > window is included below. Does this error mean anything to anyone? > My .scim folder doesn't even contain a global file. Any help would > be much appreciated. > > > > Beau > > > > /usr/local/fsl/bin/fslview: line 6: 1119 Aborted > > ${FSLDIR}/lib/fslview/bin/fslview $@ > > ScimInputContextPlugin() > > WARNING: please edit ~/.scim/global and change /DefaultConfigModule > to kconfig > > *** glibc detected *** /usr/local/fsl/lib/fslview/bin/fslview: free(): > > invalid pointer: 0x08d512e0 *** > > ======= Backtrace: ========= > > /lib/libc.so.6[0xc0f09d] > > /lib/libc.so.6(cfree+0x90)[0xc126f0] > > /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0x444c871] > > /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so > (_ZN4scim23QScimInputContextGlobal10initializeEv+0xfd2)[0xb65fd3a2] > > /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so > (_ZN4scim17QScimInputContextC1Ev+0x31e)[0xb65fe3fe] > > /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so > (_ZN22ScimInputContextPlugin6createERK7QString+0x7e)[0xb65f75de] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN26QInputContextPluginPrivate6createERK7QString+0x1b)[0x4ba7c5b] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x5d)[0x4ba78ed] > > /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so > (_ZN18QMultiInputContext17changeInputMethodE7QString+0xc3)[0xb668c51b] > > /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so > (_ZN18QMultiInputContext5slaveEv+0x43)[0xb668c6cf] > > /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so > (_ZN18QMultiInputContext15setHolderWidgetEP7QWidget+0x26)[0xb668c93e] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x75)[0x4ba7905] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget18createInputContextEv > +0x72)[0x48fdb22] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget17resetInputContextEv > +0x22)[0x48fddc2] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit7setTextERK7QString > +0x22)[0x4a7d102] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit12setMaxLengthEi > +0x36)[0x4a75fd6] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog4initEv+0x36c) > [0x4b5618c] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN11QFileDialogC1ERK7QStringS2_P7QWidgetPKcb+0xa2)[0x4b68732] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN11QFileDialog15getOpenFileNameERK7QStringS2_P7QWidgetPKcS2_PS0_b > +0x38b)[0x4b6c64b] > > /usr/local/fsl/lib/fslview/bin/fslview > (_ZN17ApplicationWindow8fileOpenEv+0x3e)[0x80a9392] > > /usr/local/fsl/lib/fslview/bin/fslview > (_ZN21ApplicationWindowBase9qt_invokeEiP8QUObject+0x74)[0x8143df0] > > /usr/local/fsl/lib/fslview/bin/fslview > (_ZN17ApplicationWindow9qt_invokeEiP8QUObject+0x31)[0x8161ad9] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a) > [0x499de2a] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEi > +0xbd)[0x499e95d] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9activatedEv+0x29) > [0x4d4df29] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction18internalActivationEv > +0x50)[0x4af9a30] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9qt_invokeEiP8QUObject > +0x10c)[0x4d4de1c] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a) > [0x499de2a] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal6signalERK8QVariant > +0x8e)[0x4d297ae] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal8activateEv+0x77) > [0x49bc5f7] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN10QPopupMenu17mouseReleaseEventEP11QMouseEvent+0x475)[0x4abd3f5] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget5eventEP6QEvent+0x3c5) > [0x49db2d5] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN12QApplication14internalNotifyEP7QObjectP6QEvent+0x9b)[0x49350fb] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN12QApplication6notifyEP7QObjectP6QEvent+0x287)[0x4936757] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN9QETWidget19translateMouseEventEPK7_XEvent+0xf40)[0x48cd8f0] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3 > (_ZN12QApplication15x11ProcessEventEP7_XEvent+0x5e6)[0x48cba26] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop13processEventsEj > +0x4eb)[0x48dd6ab] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop9enterLoopEv+0x42) > [0x494e672] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop4execEv+0x26) > [0x494e536] > > /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication4execEv+0x1f) > [0x4934c0f] > > /usr/local/fsl/lib/fslview/bin/fslview(main+0x11ca)[0x809e77a] > > /lib/libc.so.6(__libc_start_main+0xdc)[0xbbef2c] > > /usr/local/fsl/lib/fslview/bin/fslview > (_ZN6QFrame10paintEventEP11QPaintEvent+0x3d)[0x809cba1] > > ======= Memory map: ======== > > 00101000-001ff000 r-xp 00000000 08:06 2638398 /usr/lib/libX11.so. > 6.2.0 > > 001ff000-00203000 rwxp 000fe000 08:06 2638398 /usr/lib/libX11.so. > 6.2.0 > > 0024c000-0026b000 r-xp 00000000 08:06 623523 /lib/libexpat.so. > 0.5.0 > > 0026b000-0026d000 rwxp 0001e000 08:06 623523 /lib/libexpat.so. > 0.5.0 > > 0026f000-0027e000 r-xp 00000000 08:06 2638403 /usr/lib/ > libXext.so.6.4.0 > > 0027e000-0027f000 rwxp 0000e000 08:06 2638403 /usr/lib/ > libXext.so.6.4.0 > > 00281000-002fe000 r-xp 00000000 08:06 2638393 /usr/lib/ > libfreetype.so.6.3.10 > > 002fe000-00301000 rwxp 0007d000 08:06 2638393 /usr/lib/ > libfreetype.so.6.3.10 > > 00303000-0032a000 r-xp 00000000 08:06 2638394 /usr/lib/ > libfontconfig.so.1.1.0 > > 0032a000-00332000 rwxp 00027000 08:06 2638394 /usr/lib/ > libfontconfig.so.1.1.0 > > 00334000-00359000 r-xp 00000000 08:06 2638395 /usr/lib/ > libpng12.so.0.10.0 > > 00359000-0035a000 rwxp 00024000 08:06 2638395 /usr/lib/ > libpng12.so.0.10.0 > > 0035c000-00364000 r-xp 00000000 08:06 2638399 /usr/lib/ > libXrender.so.1.3.0 > > 00364000-00365000 rwxp 00007000 08:06 2638399 /usr/lib/ > libXrender.so.1.3.0 > > 00367000-00370000 r-xp 00000000 08:06 2638408 /usr/lib/ > libXcursor.so.1.0.2 > > 00370000-00371000 rwxp 00008000 08:06 2638408 /usr/lib/ > libXcursor.so.1.0.2 > > 00373000-00375000 r-xp 00000000 08:06 2638404 /usr/lib/ > libXinerama.so.1.0.0 > > 00375000-00376000 rwxp 00001000 08:06 2638404 /usr/lib/ > libXinerama.so.1.0.0 > > 00378000-0037c000 r-xp 00000000 08:06 2638407 /usr/lib/ > libXfixes.so.3.1.0 > > 0037c000-0037d000 rwxp 00003000 08:06 2638407 /usr/lib/ > libXfixes.so.3.1.0 > > 0037f000-00382000 r-xp 00000000 08:06 2638406 /usr/lib/ > libXrandr.so.2.0.0 > > 00382000-00383000 rwxp 00002000 08:06 2638406 /usr/lib/ > libXrandr.so.2.0.0 > > 008db000-0092f000 r-xp 00000000 08:06 2621533 /usr/lib/libXt.so. > 6.0.0 > > 0092f000-00933000 rwxp 00054000 08:06 2621533 /usr/lib/libXt.so. > 6.0.0 > > 00955000-009bf000 r-xp 00000000 08:06 2638493 /usr/lib/libGL.so.1.2 > > 009bf000-009c2000 rwxp 00069000 08:06 2638493 /usr/lib/libGL.so.1.2 > > 009c2000-009c3000 rwxp 009c2000 00:00 0 > > 009c7000-009c9000 r-xp 00000000 08:06 2635087 /usr/lib/libscim- > x11utils-1.0.so.8.1.0 > > 009c9000-009ca000 rwxp 00001000 08:06 2635087 /usr/lib/libscim- > x11utils-1.0.so.8.1.0 > > ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Tue, 29 May 2007 15:04:18 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Yaroslav Halchenko <[log in to unmask]> Subject: Motion correction / example_func (registration) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Dear FSL People, Could you please clarify the situation a bit for me and correct me where I am wrong. Looking at feat.tcl and report.log I got an impression (please correct me if I am wrong), that 1. mcflirt takes middle of the sequence as the reference image, and it doesn't care about optional parameter such as alternative_example_func and there is no parameter to provide alternative 'target_vol_number': ,--------------------------------------------------- | set target_vol_number [ expr $total_volumes / 2 ] `--- 2. example_func (which is used later on for registration) is extracted prior the run of mcflirt, and by default (unless alternative_example_func is provided) the very first volume is taken to be the reference ,----------------------------------------------------------------------------- | fsl:exec "${FSLDIR}/bin/avwroi $funcdata example_func $target_vol_number 1" `--- 3. mcflirt is ran (so now example_func is not really an example of real data) The problem I see is that if there is considerable motion between 1st and 'middle' volume, all the registration of the results is skewed, because statistics are computed on motion corrected data taking middle volume as the reference; and registration is based on the 1st volume prior to motion correction. Having an option to specify which volume to use for motion correction is very beneficial imho especially if field correction is used and most probably field mapping happens right after BOLD sequence, so that field map and BOLD as aligned correctly. In such cases, I guess, taking the last volume as the reference for mcflirt and registration should be the right thing to do. -- .-. =------------------------------ /v\ ----------------------------= Keep in touch // \\ ([log in to unmask])onerussian.com Yaroslav Halchenko /( )\ ICQ#: 60653192 Linux User ^^-^^ [175555] ========================================================================= Date: Tue, 29 May 2007 15:07:59 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Yaroslav Halchenko <[log in to unmask]> Subject: Re: Motion correction / example_func (registration) -- fixed ;-) In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Heh.... my bad -- feat does take target_vol_number as the volume number for example_func -- how could I miss that. Sorry about the noise P.S. Still, option to have target_vol_number defined manually instead of being computed might be beneficial for the unwarping I guess... On Tue, 29 May 2007, Yaroslav Halchenko wrote: > Dear FSL People, > Could you please clarify the situation a bit for me and correct me where > I am wrong. > Looking at feat.tcl and report.log I got an impression (please > correct me if I am wrong), that -- .-. =------------------------------ /v\ ----------------------------= Keep in touch // \\ ([log in to unmask])onerussian.com Yaroslav Halchenko /( )\ ICQ#: 60653192 Linux User ^^-^^ [175555] ========================================================================= Date: Tue, 29 May 2007 22:06:20 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Andreas Bartsch <[log in to unmask]> Subject: AW: [FSL] TBSS Pipeline MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi, I haven't seen your data but this can in fact be a problem due to the = volume specific scaling SPM99 introduces. Switch to DICOM -> NIFTI and = give it a second trial to see if it still looks weird. Cheers- Andreas ________________________________ Von: FSL - FMRIB's Software Library im Auftrag von David Kideckel Gesendet: Di 29.05.2007 19:16 An: [log in to unmask] Betreff: Re: [FSL] TBSS Pipeline hi, the ref number is 913750 (it seems to be taking a while to upload = due to the file size) . it is the 4D volume. I wonder if it is the fact = that I made my headers with SPM99, then created a 4D image using FSL = commands (avwmerge -a and then ran the eddy correct = script: eddy_correct . This shouldn't be a problem = though, right?=20 Thanks again! David On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote:=20 what is the difference between the two FA images you sent? the most likely explanation for your results is that the data you have = do not correspond to the bvecs you provide to dtifit.=20 could you upload the actual data (not the result of dtifit) ? =09 saad. =09 On 29 May 2007, at 17:40, David Kideckel wrote:=20 hi. i uploaded another FA image that looks slightly better, but still = pretty bad. the reference number is 886189. Thanks again! David =09 =09 On 5/29/07, David Kideckel <[log in to unmask]> wrote:=20 hi. thank you. the ref number is 713292. i sent the FA image after = DTIFIT. Do you want me to send any other ones? David=20 =09 =09 =09 On 5/29/07, Saad Jbabdi < [log in to unmask] = > wrote:=20 they should work fine. i don't know why your FA image looks like you describe. maybe you could upload your tar-gzipped data onto our server: = www.fmrib.ox.ac.uk/cgi-bin/upload.cgi and send the reference number. we will have a look. cheers, saad =09 On 29 May 2007, at 17:18, David Kideckel wrote:=20 hi. when i run dtifit, and view the FA image, it look like a = non-diffusion weighted image. do these bvecs and bvals look okay? =09 Thank you. =09 David =09 On 5/28/07, David Kideckel < [log in to unmask] = > wrote:=20 hello. i have tried your previous suggestions, and think that the = problem may be in the bvecs file, but i do not know for sure. how would = i check for this?=20 thanks again. David=20 =09 =09 =09 On 5/25/07, Tim Behrens < [log in to unmask] = > wrote:=20 Hi - if you run bedpost_datacheck on your directory it might tell = you what's wrong - =20 for example make sure the number of entriess in bvals and bvecs is = the same as the number of volumes in your data.=20 T=20 =09 =09 On 24 May 2007, at 00:17, David Kideckel wrote: hi. thank you for your support. once i run bet from the command = line (using cygwin), it does not seem to be working properly. but when i = use the gui, it does...=20 anyway, i am trying to rum dtifit from the command line and get the = following message:=20 0 256 0 256 0 50 0 slices processed=20 Aborted =09 Any idea why this is happening? Did I enter something wrong?=20 =09 Thanks. David =09 =09 On 5/23/07, Tim Behrens < [log in to unmask] = > wrote:=20 If you select dtifit from the drop down menu on the same page, all` = the information about file specifications is there. Don't forget to end = your bvecs and bvals files with carriage returns.=20 Cheers T=20 =09 =09 On 23 May 2007, at 18:53, David Kideckel wrote: Hi. That was one of the firs things I did do. I will try to be more = specific with my question. I want to do TBSS, and no tractography. So, = would I do the following in this order:=20 Eddy Correct BET DTI Fit =09 If so, where is it in the manual specified how to create and input = the bvecs and bvals files? I follow the link it gives in the FAQ = section, but I can't seem to find this info.=20 =09 Thanks again for your help. David =09 =09 On 5/23/07, Tim Behrens < [log in to unmask] = > wrote:=20 Hi - if you go to =20 http://www.fmrib.ox.ac.uk/fsl/fdt/index.html =09 and select "processing pipeline", all should be clear. =09 T =09 On 23 May 2007, at 17:22, David Kideckel wrote:=20 Hi. Thank you for your response. Can you please send me this = information. I can't seem to find it on the FSL website.=20 Thanks you.=20 David =09 =09 On 5/23/07, Steve Smith < [log in to unmask] = > wrote:=20 Hi, =09 Indeed - you should run eddy_correct and then bet and then dtifit = to=20 produce the FA image - see the FDT manual pages for more details. Sorry the TBSS page isn't clearer about that - we'll fix that = shortly.=20 =09 Cheers. =09 =09 =09 On 22 May 2007, at 18:51, David Kideckel wrote:=20 =09 > Dear FSLers, > > I am new to TBSS and am wondering what the preprocessing steps = are > to do prior to TBSS. This information seems to be different on = the=20 > FAQ section of the FSL website and the technical report on TBSS.=20 > > Is it correct to assume that to compare 2 groups with diffusion > data, I would need to run: > Eddy correct > Registration=20 > DTIFit > TBSS > > I am unclear as to whether I need to run BET (brain extraction=20 > tool) and/or BEDPOST. > > Any guidance would be much appreciated. > > David Kideckel, PhD Candidate=20 =09 =09 = ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre =09 FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK=20 +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve = =20 = ------------------------------------------------------------------------ = --- =09 =09 =09 =09 = -------------------------------------------------------------------------= -- Saad Jbabdi,=20 Postdoctoral Research Assistant, =20 =09 Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222545 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~saad = -------------------------------------------------------------------------= --=20 =09 = -------------------------------------------------------------------------= -- Saad Jbabdi,=20 Postdoctoral Research Assistant, =20 Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222545 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~saad = =20 = -------------------------------------------------------------------------= -- ========================================================================= Date: Tue, 29 May 2007 16:24:41 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: David Kideckel <[log in to unmask]> Subject: Re: AW: [FSL] TBSS Pipeline In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_263160_14622448.1180470281089" ------=_Part_263160_14622448.1180470281089 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hi. thanks. that's interesting... how do i switch to that? david On 5/29/07, Andreas Bartsch <[log in to unmask]> wrote: > > Hi, > I haven't seen your data but this can in fact be a problem due to the > volume specific scaling SPM99 introduces. Switch to DICOM -> NIFTI and give > it a second trial to see if it still looks weird. > Cheers- > Andreas > > ________________________________ > > Von: FSL - FMRIB's Software Library im Auftrag von David Kideckel > Gesendet: Di 29.05.2007 19:16 > An: [log in to unmask] > Betreff: Re: [FSL] TBSS Pipeline > > > hi, the ref number is 913750 (it seems to be taking a while to upload due > to the file size) . it is the 4D volume. I wonder if it is the fact that I > made my headers with SPM99, then created a 4D image using FSL commands > (avwmerge -a and then ran the eddy correct script: > eddy_correct . This shouldn't be a problem though, right? > Thanks again! > David > > > On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote: > > what is the difference between the two FA images you sent? > > the most likely explanation for your results is that the data you > have do not correspond to the bvecs you provide to dtifit. > could you upload the actual data (not the result of dtifit) ? > > > saad. > > > > On 29 May 2007, at 17:40, David Kideckel wrote: > > > hi. i uploaded another FA image that looks slightly > better, but still pretty bad. the reference number is 886189. > Thanks again! > David > > > On 5/29/07, David Kideckel <[log in to unmask]> > wrote: > > hi. thank you. the ref number is 713292. i sent > the FA image after DTIFIT. Do you want me to send any other ones? > David > > > > On 5/29/07, Saad Jbabdi < [log in to unmask] [log in to unmask]> > wrote: > > they should work fine. > i don't know why your FA image looks like > you describe. > maybe you could upload your tar-gzipped > data onto our server: www.fmrib.ox.ac.uk/cgi-bin/upload.cgi > and send the reference number. we will > have a look. > > cheers, > saad > > > On 29 May 2007, at 17:18, David Kideckel > wrote: > > > hi. when i run dtifit, and view > the FA image, it look like a non-diffusion weighted image. do these bvecs > and bvals look okay? > > Thank you. > > David > > On 5/28/07, David Kideckel < > [log in to unmask] > wrote: > > hello. i have tried your previous > suggestions, and think that the problem may be in the bvecs file, but i do > not know for sure. how would i check for this? > thanks again. > David > > > > On 5/25/07, Tim Behrens < > [log in to unmask] > wrote: > > Hi - if you run bedpost_datacheck > on your directory it might tell you what's wrong - > for example make sure the number > of entriess in bvals and bvecs is the same as the number of volumes in your > data. > > T > > > On 24 May 2007, at 00:17, David > Kideckel wrote: > > > hi. thank you for your support. > once i run bet from the command line (using cygwin), it does not seem to be > working properly. but when i use the gui, it does... > anyway, i am trying to rum dtifit > from the command line and get the following message: > 0 256 0 256 0 50 > 0 slices processed > Aborted > > Any idea why this is happening? > Did I enter something wrong? > > Thanks. > David > > > On 5/23/07, Tim Behrens < > [log in to unmask] > wrote: > > If you select dtifit from the drop > down menu on the same page, all` the information about file specifications > is there. Don't forget to end your bvecs and bvals files with carriage > returns. > > Cheers > > T > > > On 23 May 2007, at 18:53, David > Kideckel wrote: > > > Hi. That was one of the firs > things I did do. I will try to be more specific with my question. I want to > do TBSS, and no tractography. So, would I do the following in this order: > Eddy Correct > BET > DTI Fit > > If so, where is it in the manual > specified how to create and input the bvecs and bvals files? I follow the > link it gives in the FAQ section, but I can't seem to find this info. > > Thanks again for your help. > David > > > On 5/23/07, Tim Behrens < > [log in to unmask] > wrote: > > Hi - if you go to > > http://www.fmrib.ox.ac.uk/fsl/fdt/index.html > > > and select "processing pipeline", > all should be clear. > > > T > > > On 23 May 2007, at 17:22, David > Kideckel wrote: > > > Hi. Thank you for your response. > Can you please send me this information. I can't seem to find it on the FSL > website. > Thanks you. > David > > > On 5/23/07, Steve Smith < > [log in to unmask] > wrote: > > Hi, > > Indeed - you should run > eddy_correct and then bet and then dtifit to > produce the FA image - see the FDT > manual pages for more details. > Sorry the TBSS page isn't clearer > about that - we'll fix that shortly. > > Cheers. > > > > On 22 May 2007, at 18:51, David > Kideckel wrote: > > > Dear FSLers, > > > > I am new to TBSS and am > wondering what the preprocessing steps are > > to do prior to TBSS. This > information seems to be different on the > > FAQ section of the FSL website > and the technical report on TBSS. > > > > Is it correct to assume that to > compare 2 groups with diffusion > > data, I would need to run: > > Eddy correct > > Registration > > DTIFit > > TBSS > > > > I am unclear as to whether I > need to run BET (brain extraction > > tool) and/or BEDPOST. > > > > Any guidance would be much > appreciated. > > > > David Kideckel, PhD Candidate > > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of > Biomedical Engineering > Associate Director, Oxford > University FMRIB Centre > > FMRIB, JR Hospital, Headington, > Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] > http://www.fmrib.ox.ac.uk/~steve > > ------------------------------------------------------------------------ > --- > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------------- > Saad Jbabdi, > Postdoctoral Research Assistant, > > Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, > Oxford OX3 9DU, UK > +44 (0) 1865 222545 (fax 222717) > [log in to unmask] > http://www.fmrib.ox.ac.uk/~saad > > --------------------------------------------------------------------------- > > > > > > > > > > > --------------------------------------------------------------------------- > Saad Jbabdi, > Postdoctoral Research Assistant, > Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222545 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~saad < > http://www.fmrib.ox.ac.uk/~saad> > > --------------------------------------------------------------------------- > ------=_Part_263160_14622448.1180470281089 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline hi. thanks.  that's interesting... how do i switch to that?
dav= id

On 5/29/07, Andreas Bartsch < [log in to unmask]> wrote:
Hi,
I haven't seen your= data but this can in fact be a problem due to the volume specific scaling = SPM99 introduces. Switch to DICOM -> NIFTI and give it a second trial to= see if it still looks weird.
Cheers-
Andreas

________________________________

Von: = FSL - FMRIB's Software Library im Auftrag von David Kideckel
Gesende= t: Di 29.05.2007 19:16
An: FSL@JIS= CMAIL.AC.UK
Betreff: Re: [FSL] TBSS Pipeline


hi, the ref number is 9= 13750 (it seems to be taking a while to upload due to the file size) . it i= s the 4D volume. I wonder if it is the fact that I made my headers with SPM= 99, then created a 4D image using FSL commands (avwmerge -a <output> = <input> and then ran the eddy correct script: eddy_correct <input&= gt; <output>. This shouldn't be a problem though, right?
Thanks again!
David


On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote:

&= nbsp;       what is the difference betwe= en the two FA images you sent?

      &= nbsp; the most likely explanation for your results is that the data yo= u have do not correspond to the bvecs you provide to dtifit.
        could you upload the ac= tual data (not the result of dtifit) ?


    &= nbsp;   saad.



     &= nbsp;  On 29 May 2007, at 17:40, David Kideckel wrote:

            &n= bsp;   hi. i uploaded another FA image that looks slightly b= etter, but still pretty bad. the reference number is 886189.
            = ;    Thanks again!
     &nb= sp;          David

           &nbs= p;    On 5/29/07, David Kideckel <[log in to unmask]">[log in to unmask]> wrote:

&n= bsp;            = ;           hi. than= k you. the ref number is 713292. i sent the FA image after DTIFIT. Do you w= ant me to send any other ones?
            = ;            Da= vid



         &n= bsp;            = ;  On 5/29/07, Saad Jbabdi < [log in to unmask] <mailto:[log in to unmask] > > wrote:

        = ;            &n= bsp;           they = should work fine.
         =             &nb= sp;          i don't = know why your FA image looks like you describe.
    =             &nb= sp;            =    maybe you could upload your tar-gzipped data onto our ser= ver:=20 www.fmrib.ox.ac.uk= /cgi-bin/upload.cgi
        =             &nb= sp;           and se= nd the reference number. we will have a look.

   &nbs= p;            &= nbsp;           &nbs= p;   cheers,
            = ;            &n= bsp;       saad


  &= nbsp;           &nbs= p;            &= nbsp;    On 29 May 2007, at 17:18, David Kideckel wrote= :


          &n= bsp;            = ;            &n= bsp;    hi. when i run dtifit, and view the FA image, i= t look like a non-diffusion weighted image. do these bvecs and bvals look o= kay?

           &= nbsp;           &nbs= p;            &= nbsp;   Thank you.

     &nbs= p;            &= nbsp;           &nbs= p;         David

 =             &nb= sp;            =             &nb= sp; On 5/28/07, David Kideckel < [log in to unmask] <mailto:[log in to unmask]> > wrote:

  &n= bsp;            = ;            &n= bsp;            = ;hello. i have tried your previous suggestions, and think that the problem = may be in the bvecs file, but i do not know for sure. how would i check for= this?
            = ;            &n= bsp;            = ;   thanks again.
      &nb= sp;            =             &nb= sp;        David



&nb= sp;            =             &nb= sp;            =   On 5/25/07, Tim Behrens < [log in to unmask] <mailto:[log in to unmask]> > wrote:

   &n= bsp;            = ;            &n= bsp;           Hi - = if you run bedpost_datacheck on your directory it might tell you what's= wrong -
            = ;            &n= bsp;            = ;   for example make sure the number of entriess in bvals an= d bvecs is the same as the number of volumes in your data.

 &nb= sp;            =             &nb= sp;            =  T


         &n= bsp;            = ;            &n= bsp;     On 24 May 2007, at 00:17, David Kideckel = wrote:


          &nb= sp;            =             &nb= sp;    hi. thank you for your support. once i run bet f= rom the command line (using cygwin), it does not seem to be working properl= y. but when i use the gui, it does...
     &nbs= p;            &= nbsp;           &nbs= p;         anyway, i am trying= to rum dtifit from the command line and get the following message:
            = ;            &n= bsp;            = ;   0 256 0 256 0 50
      =             &nb= sp;            =          0 slices processed            &n= bsp;            = ;            &n= bsp;  Aborted <core dumped>

    =             &nb= sp;            =            Any idea = why this is happening? Did I enter something wrong?

           &= nbsp;           &nbs= p;            &= nbsp;   Thanks.
       = ;            &n= bsp;            = ;        David


 &nb= sp;            =             &nb= sp;            =  On 5/23/07, Tim Behrens < [log in to unmask] <mailto:[log in to unmask]> > wrote:

   &n= bsp;            = ;            &n= bsp;           If yo= u select dtifit from the drop down menu on the same page, all` the informat= ion about file specifications is there. Don't forget to end your bvecs = and bvals files with carriage returns.

           &= nbsp;           &nbs= p;            &= nbsp;   Cheers

      &n= bsp;            = ;            &n= bsp;        T


 &nbs= p;            &= nbsp;           &nbs= p;            &= nbsp;On 23 May 2007, at 18:53, David Kideckel wrote:


  = ;            &n= bsp;            = ;            &n= bsp;Hi. That was one of the firs things I did do. I will try to be more spe= cific with my question. I want to do TBSS, and no tractography. So, would I= do the following in this order:
            = ;            &n= bsp;            = ;   Eddy Correct
      &nbs= p;            &= nbsp;           &nbs= p;        BET
   =             &nb= sp;            =             DTI= Fit

          &nb= sp;            =             &nb= sp;    If so, where is it in the manual specified how t= o create and input the bvecs and bvals files? I follow the link it gives in= the FAQ section, but I can't seem to find this info.

           &= nbsp;           &nbs= p;            &= nbsp;   Thanks again for your help.
   &nb= sp;            =             &nb= sp;           David<= br>

          &nbs= p;            &= nbsp;           &nbs= p;    On 5/23/07, Tim Behrens < [log in to unmask]"> [log in to unmask] <mailto:[log in to unmask]> > wrote:

   &n= bsp;            = ;            &n= bsp;           Hi - = if you go to
          = ;            &n= bsp;            = ;      http://www.fmrib.o= x.ac.uk/fsl/fdt/index.html


     &nb= sp;            =             &nb= sp;         and select "p= rocessing pipeline", all should be clear.


           &= nbsp;           &nbs= p;            &= nbsp;   T


      &nb= sp;            =             &nb= sp;        On 23 May 2007, at 17:22= , David Kideckel wrote:


      &nbs= p;            &= nbsp;           &nbs= p;        Hi. Thank you for your re= sponse. Can you please send me this information. I can't seem to find i= t on the FSL website.
            = ;            &n= bsp;            = ;   Thanks you.
       = ;            &n= bsp;            = ;        David


 &nb= sp;            =             &nb= sp;            =  On 5/23/07, Steve Smith < [log in to unmask] <mailto:[log in to unmask]> > wrote:

    &n= bsp;            = ;            &n= bsp;          Hi,

=             &nb= sp;            =             &nb= sp;  Indeed - you should run eddy_correct and then bet and then d= tifit to
            = ;            &n= bsp;            = ;   produce the FA image - see the FDT manual pages for more= details.
          &n= bsp;            = ;            &n= bsp;    Sorry the TBSS page isn't clearer about tha= t - we'll fix that shortly.

           &= nbsp;           &nbs= p;            &= nbsp;   Cheers.



     = ;            &n= bsp;            = ;          On 22 May 2007= , at 18:51, David Kideckel wrote:

     &nbs= p;            &= nbsp;           &nbs= p;         > Dear FSLers,
            = ;            &n= bsp;            = ;   >
        =             &nb= sp;            =        > I am new to TBSS and am wond= ering what the preprocessing steps are
     &nb= sp;            =             &nb= sp;         > to do prior t= o TBSS. This information seems to be different on the
            = ;            &n= bsp;            = ;   > FAQ section of the FSL website and the technical re= port on TBSS.
         &nbs= p;            &= nbsp;           &nbs= p;     >
      = ;            &n= bsp;            = ;         > Is it correct t= o assume that to compare 2 groups with diffusion
            = ;            &n= bsp;            = ;   > data, I would need to run:
   &nb= sp;            =             &nb= sp;           > E= ddy correct
          =             &nb= sp;            =      > Registration
    =             &nb= sp;            =            > DTIF= it
            = ;            &n= bsp;            = ;   > TBSS
       &= nbsp;           &nbs= p;            &= nbsp;       >
   &n= bsp;            = ;            &n= bsp;           > = I am unclear as to whether I need to run BET (brain extraction
 &nb= sp;            =             &nb= sp;            =  > tool) and/or BEDPOST.
            = ;            &n= bsp;            = ;   >
        =             &nb= sp;            =        > Any guidance would be much a= ppreciated.
          =             &nb= sp;            =      >
      &= nbsp;           &nbs= p;            &= nbsp;        > David Kideckel, P= hD Candidate


          &nb= sp;            =             &nb= sp;    ------------------------------------------------= ------------------------
        = ;            &n= bsp;            = ;       ---
    &= nbsp;           &nbs= p;            &= nbsp;          Stephen M.= Smith, Professor of Biomedical Engineering
            = ;            &n= bsp;            = ;   Associate Director,  Oxford University FMRIB C= entre

          &n= bsp;            = ;            &n= bsp;    FMRIB, JR Hospital, Headington, Oxford &nb= sp;OX3 9DU, UK
         &nb= sp;            =             &nb= sp;     +44 (0) 1865 222726  (fax 222717= )
            = ;            &n= bsp;            = ;   [log in to unmask] c.uk     http://www.fmrib.ox.ac.uk/~steve < http://www.fmrib.ox.ac.uk/~steve>
     &= nbsp;           &nbs= p;            &= nbsp;         ----------------= --------------------------------------------------------
  &nb= sp;            =             &nb= sp;            = ---











        = ;            &n= bsp;            = ;       <bvecs>

  = ;            &n= bsp;            = ;            &n= bsp;<bvals>



       =             &nb= sp;            = ---------------------------------------------------------------------------
            = ;            &n= bsp;       Saad Jbabdi,
  &= nbsp;           &nbs= p;            &= nbsp;    Postdoctoral Research Assistant,

 =             &nb= sp;            =       Oxford University FMRIB Centre

&= nbsp;           &nbs= p;            &= nbsp;      FMRIB, JR Hospital, Headington, Ox= ford  OX3 9DU, UK
            = ;            &n= bsp;       +44 (0) 1865 222545 &nbs= p;(fax 222717)
         &nb= sp;            =           [log in to unmask]">[log in to unmask]     http://www.fmrib.ox.ac.uk/~saad
           &= nbsp;           &nbs= p;        -------------------------= --------------------------------------------------





<= br>


        ------------= ---------------------------------------------------------------
        Saad Jbabdi,
 &= nbsp;      Postdoctoral Research Assistant,        Oxford University FMRIB C= entre

        FMRIB, JR Hosp= ital, Headington, Oxford  OX3 9DU, UK
    =     +44 (0) 1865 222545   (fax 222717)
        [log in to unmask]     http://www.fmrib.ox.ac.uk/~saad <http://www.fmrib.ox.ac.uk/~saad >
        ---------------= ------------------------------------------------------------

------=_Part_263160_14622448.1180470281089-- ========================================================================= Date: Tue, 29 May 2007 23:12:45 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Riccardo Della Nave <[log in to unmask]> Subject: Re: Export jpeg from FSLVIEW In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Content-Disposition: inline If you have KDE try KSnapshot with "Capture mode: Section of Window" On Monday 28 May 2007 19:47:43 Marc Dubin wrote: > Thanks, Matt! > > On 5/28/07, Matt <[log in to unmask]> wrote: > > Marc, > > > > There isn't a way to do that; you need to take a screen capture. > > > > Peace, > > > > Matt. > > > > -----Original Message----- > > From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On > > Behalf Of Marc Dubin > > Sent: Monday, May 28, 2007 1:31 PM > > To: [log in to unmask] > > Subject: [FSL] Export jpeg from FSLVIEW > > > > Does anyone know how to export one of the images from fslview > > in jpeg or gif or tiff format? > > > > Thanks, > > Marc ========================================================================= Date: Tue, 29 May 2007 22:39:04 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Bradley Goodyear <[log in to unmask]> Subject: FEAT on multiple processors Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain Is there a set up I need to go through to permit FEAT to take advantage o= f the multiple processors=20 on a G5 or Intel Mac? -Brad ========================================================================= Date: Tue, 29 May 2007 22:47:10 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: FEAT on multiple processors In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi - we're planning that for the next release we'll make it easy to use SGE to achieve this. Cheers. On 29 May 2007, at 22:39, Bradley Goodyear wrote: > Is there a set up I need to go through to permit FEAT to take > advantage of the multiple processors > on a G5 or Intel Mac? > > -Brad ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Tue, 29 May 2007 23:50:07 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Andreas Bartsch <[log in to unmask]> Subject: AW: [FSL] AW: [FSL] TBSS Pipeline MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi, first you could check if the volumes are indeed scaled differently. = Alternatively, you could write a little matlab or R code to rescale your = volumes uniformely - see a previous post of mine. But I guess you are = asking for a converter: see the quite recent postings to this list. = Freesurfer comes with mri_convert which is quite neat, and Jolinda has = written the fast and cool converter MRIConvert (coincidential naming;). Cheers- Andreas ________________________________ Von: FSL - FMRIB's Software Library im Auftrag von David Kideckel Gesendet: Di 29.05.2007 22:24 An: [log in to unmask] Betreff: Re: [FSL] AW: [FSL] TBSS Pipeline hi. thanks. that's interesting... how do i switch to that? david On 5/29/07, Andreas Bartsch < [log in to unmask] = > wrote:=20 Hi, I haven't seen your data but this can in fact be a problem due to the = volume specific scaling SPM99 introduces. Switch to DICOM -> NIFTI and = give it a second trial to see if it still looks weird.=20 Cheers- Andreas =09 ________________________________ =09 Von: FSL - FMRIB's Software Library im Auftrag von David Kideckel Gesendet: Di 29.05.2007 19:16 An: [log in to unmask] Betreff: Re: [FSL] TBSS Pipeline =09 =09 hi, the ref number is 913750 (it seems to be taking a while to upload = due to the file size) . it is the 4D volume. I wonder if it is the fact = that I made my headers with SPM99, then created a 4D image using FSL = commands (avwmerge -a and then ran the eddy correct = script: eddy_correct . This shouldn't be a problem = though, right?=20 Thanks again! David =09 =09 On 5/29/07, Saad Jbabdi <[log in to unmask]> wrote: =09 what is the difference between the two FA images you sent? =09 the most likely explanation for your results is that the data = you have do not correspond to the bvecs you provide to dtifit.=20 could you upload the actual data (not the result of dtifit) ? =09 =09 saad. =09 =09 =09 On 29 May 2007, at 17:40, David Kideckel wrote: =09 =09 hi. i uploaded another FA image that looks slightly = better, but still pretty bad. the reference number is 886189.=20 Thanks again! David =09 =09 On 5/29/07, David Kideckel <[log in to unmask]> = wrote: =09 hi. thank you. the ref number is 713292. i sent = the FA image after DTIFIT. Do you want me to send any other ones?=20 David =09 =09 =09 On 5/29/07, Saad Jbabdi < [log in to unmask] = > wrote: =09 they should work fine. i don't know why your FA image looks = like you describe. maybe you could upload your tar-gzipped = data onto our server: www.fmrib.ox.ac.uk/cgi-bin/upload.cgi and send the reference number. we will = have a look. =09 cheers,=20 saad =09 =09 On 29 May 2007, at 17:18, David = Kideckel wrote: =09 =09 hi. when i run dtifit, and view = the FA image, it look like a non-diffusion weighted image. do these = bvecs and bvals look okay?=20 =09 Thank you. =09 David =09 On 5/28/07, David Kideckel < = [log in to unmask] > wrote: =09 hello. i have tried your = previous suggestions, and think that the problem may be in the bvecs = file, but i do not know for sure. how would i check for this?=20 thanks again. David =09 =09 =09 On 5/25/07, Tim Behrens < = [log in to unmask] > wrote: =09 Hi - if you run = bedpost_datacheck on your directory it might tell you what's wrong -=20 for example make sure the = number of entriess in bvals and bvecs is the same as the number of = volumes in your data. =09 T =09 =09 On 24 May 2007, at 00:17, David = Kideckel wrote:=20 =09 =09 hi. thank you for your support. = once i run bet from the command line (using cygwin), it does not seem to = be working properly. but when i use the gui, it does... anyway, i am trying to rum = dtifit from the command line and get the following message:=20 0 256 0 256 0 50 0 slices processed Aborted =09 Any idea why this is happening? = Did I enter something wrong?=20 =09 Thanks. David =09 =09 On 5/23/07, Tim Behrens < = [log in to unmask] > wrote: =09 If you select dtifit from the = drop down menu on the same page, all` the information about file = specifications is there. Don't forget to end your bvecs and bvals files = with carriage returns.=20 =09 Cheers =09 T =09 =09 On 23 May 2007, at 18:53, David = Kideckel wrote: =09 =09 Hi. That was one of the firs = things I did do. I will try to be more specific with my question. I want = to do TBSS, and no tractography. So, would I do the following in this = order:=20 Eddy Correct BET DTI Fit =09 If so, where is it in the = manual specified how to create and input the bvecs and bvals files? I = follow the link it gives in the FAQ section, but I can't seem to find = this info.=20 =09 Thanks again for your help. David =09 =09 On 5/23/07, Tim Behrens < = [log in to unmask] > wrote: =09 Hi - if you go to = http://www.fmrib.ox.ac.uk/fsl/fdt/index.html =09 =09 and select "processing = pipeline", all should be clear. =09 =09 T =09 =09 On 23 May 2007, at 17:22, David = Kideckel wrote: =09 =09 Hi. Thank you for your = response. Can you please send me this information. I can't seem to find = it on the FSL website.=20 Thanks you. David =09 =09 On 5/23/07, Steve Smith < = [log in to unmask] > wrote: =09 Hi, =09 Indeed - you should run = eddy_correct and then bet and then dtifit to=20 produce the FA image - see the = FDT manual pages for more details. Sorry the TBSS page isn't = clearer about that - we'll fix that shortly.=20 =09 Cheers. =09 =09 =09 On 22 May 2007, at 18:51, David = Kideckel wrote: =09 > Dear FSLers,=20 > > I am new to TBSS and am = wondering what the preprocessing steps are > to do prior to TBSS. This = information seems to be different on the=20 > FAQ section of the FSL = website and the technical report on TBSS. > > Is it correct to assume that = to compare 2 groups with diffusion=20 > data, I would need to run: > Eddy correct > Registration > DTIFit=20 > TBSS > > I am unclear as to whether I = need to run BET (brain extraction > tool) and/or BEDPOST.=20 > > Any guidance would be much = appreciated. > > David Kideckel, PhD Candidate = =09 =09 = ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of = Biomedical Engineering=20 Associate Director, Oxford = University FMRIB Centre =09 FMRIB, JR Hospital, Headington, = Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax = 222717)=20 [log in to unmask] = http://www.fmrib.ox.ac.uk/~steve < http://www.fmrib.ox.ac.uk/~steve = > = ------------------------------------------------------------------------ --- =09 =09 =09 =09 =09 =09 =09 =09 =09 =09 =09 =09 =09 =09 =09 = -------------------------------------------------------------------------= --=20 Saad Jbabdi, Postdoctoral Research Assistant, =09 Oxford University FMRIB Centre =09 FMRIB, JR Hospital, Headington, Oxford = OX3 9DU, UK=20 +44 (0) 1865 222545 (fax 222717) [log in to unmask] = http://www.fmrib.ox.ac.uk/~saad=20 = -------------------------------------------------------------------------= -- =09 =09 =09 =09 =09 =09 =09 =09 =09 = -------------------------------------------------------------------------= --=20 Saad Jbabdi, Postdoctoral Research Assistant, Oxford University FMRIB Centre =09 FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222545 (fax 222717)=20 [log in to unmask] http://www.fmrib.ox.ac.uk/~saad = = -------------------------------------------------------------------------= -- =09 ========================================================================= Date: Wed, 30 May 2007 03:25:59 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Peter Smale <[log in to unmask]> Subject: strange eigenvectors Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="ISO-8859-1" Hi all, I am getting strange eigenvector results with dtifit. I have uploaded a picture of the eigenvector lines at 230306.=20 As far as I know, the gradient direction information is all ok. I have obtained good results from other data from the same machine. The only difference with this data is that it comes off the scanner as an unaverag= ed mosaic (3 x 12+1 images). Regards, Peter ========================================================================= Date: Tue, 29 May 2007 19:23:52 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: ramya priya <[log in to unmask]> Subject: SIENAX MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-156544336-1180491832=:76320" Content-Transfer-Encoding: 8bit --0-156544336-1180491832=:76320 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hello, I have few questions regarding the calculation of brain volume using SIENAX. 1. Is there any difference in the brain volume between 2 sequences with and without any gap? Does SIENAX estimate the gap area into the total brain volume when there is gap between slices? 2.If we have a T1 weighted sequence, slice with 6mm thickness and 1 mm gap,does it make a difference on the brain volume calculation? 3. In our studies,we observed that few cases have bad results(brain vol less than 1200ml) even though the parameters of these cases are the same as other cases with good results(vol between 1300ml-1700ml)?what factors affect the calculation of the brain volume? Please let us know. Thank You, Priya --------------------------------- Get the free Yahoo! toolbar and rest assured with the added security of spyware protection. --0-156544336-1180491832=:76320 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 8bit
Hello,
 
I have few questions regarding the calculation of brain volume using SIENAX.
 
1. Is there any difference in the brain volume between 2 sequences with and without any gap? Does SIENAX estimate the gap area into the total brain volume when there is gap between slices?
 
2.If we have a T1 weighted sequence, slice with 6mm thickness and 1 mm gap,does it make a difference on the brain volume calculation?
 
3. In our studies,we observed that few cases have bad results(brain vol less than 1200ml) even though the parameters of these cases are the same as other cases with good results(vol between 1300ml-1700ml)?what factors affect the calculation of the brain volume? Please let us know.
 
Thank You,
Priya 


Get the free Yahoo! toolbar and rest assured with the added security of spyware protection. --0-156544336-1180491832=:76320-- ========================================================================= Date: Tue, 29 May 2007 20:38:13 -0600 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Bradley Goodyear <[log in to unmask]> Subject: Re: FEAT on multiple processors MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=response Content-Transfer-Encoding: 7bit Thanks Steve. I have read some discussion on the use of multiple processors already. Is it only available for some tools, and not FEAT at the moment? -BRad ----- Original Message ----- From: "Steve Smith" <[log in to unmask]> To: <[log in to unmask]> Sent: Tuesday, May 29, 2007 3:47 PM Subject: Re: [FSL] FEAT on multiple processors > Hi - we're planning that for the next release we'll make it easy to > use SGE to achieve this. > Cheers. > > > On 29 May 2007, at 22:39, Bradley Goodyear wrote: > >> Is there a set up I need to go through to permit FEAT to take >> advantage of the multiple processors >> on a G5 or Intel Mac? >> >> -Brad > > > ------------------------------------------------------------------------ > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ------------------------------------------------------------------------ > --- > > > ========================================================================= Date: Tue, 29 May 2007 23:07:29 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Yaroslav Halchenko <[log in to unmask]> Subject: Re: Motion correction / example_func (registration) -- fixed ;-) In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Sorry for the noise again. Now I recalled why the heck I started to mess with target_vol_number at the first place. MC of one of the subject revealed substantial amount of motion... and what is more interesting/problematic is that subject moved during TR of the 'middle volume', so that the brain got 'extended' and I got problems with MC itself since all volumes were aligned to the bad ones and registration of the bold to anatomical wasn't good either. So, an option to select volume to use for registration/MC would be great On Tue, 29 May 2007, Yaroslav Halchenko wrote: > Heh.... my bad -- feat does take target_vol_number as the volume number > for example_func -- how could I miss that. Sorry about the noise > P.S. Still, option to have target_vol_number defined manually instead of > being computed might be beneficial for the unwarping I guess... -- .-. =------------------------------ /v\ ----------------------------= Keep in touch // \\ ([log in to unmask])onerussian.com Yaroslav Halchenko /( )\ ICQ#: 60653192 Linux User ^^-^^ [175555] ========================================================================= Date: Wed, 30 May 2007 05:45:16 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: FEAT on multiple processors In-Reply-To: <000501c7a263$91e34080$657ba8c0@FROGGER> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit That's right - just TBSS and parts of FDT at the moment if I recall correctly. Cheers. On 30 May 2007, at 03:38, Bradley Goodyear wrote: > Thanks Steve. > I have read some discussion on the use of multiple processors already. > Is it only available for some tools, and not FEAT at the moment? > > -BRad > > ----- Original Message ----- From: "Steve Smith" > <[log in to unmask]> > To: <[log in to unmask]> > Sent: Tuesday, May 29, 2007 3:47 PM > Subject: Re: [FSL] FEAT on multiple processors > > >> Hi - we're planning that for the next release we'll make it easy >> to use SGE to achieve this. >> Cheers. >> On 29 May 2007, at 22:39, Bradley Goodyear wrote: >>> Is there a set up I need to go through to permit FEAT to take >>> advantage of the multiple processors >>> on a G5 or Intel Mac? >>> >>> -Brad >> --------------------------------------------------------------------- >> --- --- >> Stephen M. Smith, Professor of Biomedical Engineering >> Associate Director, Oxford University FMRIB Centre >> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >> +44 (0) 1865 222726 (fax 222717) >> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >> --------------------------------------------------------------------- >> --- --- >> ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Wed, 30 May 2007 06:15:33 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Motion correction / example_func (registration) -- fixed ;-) In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi - the easy fix would be to run mcflirt from the command line, using the -refvol option, and then feed the output into FEAT with MC turned off in FEAT. Cheers. On 30 May 2007, at 04:07, Yaroslav Halchenko wrote: > Sorry for the noise again. Now I recalled why the heck I started to > mess > with target_vol_number at the first place. > > MC of one of the subject revealed substantial amount of motion... and > what is more interesting/problematic is that subject moved during > TR of > the 'middle volume', so that the brain got 'extended' and I got > problems with MC itself since all volumes were aligned to the bad ones > and registration of the bold to anatomical wasn't good either. > > So, an option to select volume to use for registration/MC would be > great > On Tue, 29 May 2007, Yaroslav Halchenko wrote: > >> Heh.... my bad -- feat does take target_vol_number as the volume >> number >> for example_func -- how could I miss that. Sorry about the noise > >> P.S. Still, option to have target_vol_number defined manually >> instead of >> being computed might be beneficial for the unwarping I guess... > -- > .-. > =------------------------------ /v\ ----------------------------= > Keep in touch // \\ ([log in to unmask])onerussian.com > Yaroslav Halchenko /( )\ ICQ#: 60653192 > Linux User ^^-^^ [175555] ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Wed, 30 May 2007 06:19:52 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: SIENAX In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi, On 30 May 2007, at 03:23, ramya priya wrote: > I have few questions regarding the calculation of brain volume > using SIENAX. > > 1. Is there any difference in the brain volume between 2 sequences > with and without any gap? Does SIENAX estimate the gap area into > the total brain volume when there is gap between slices? _Almost_ always in MRI analysis there is no need to differentiate between imaging with and without gaps, as long as the total effective slice thickness (including any gap) is correctly set in the header. Programs like SIENAX will work fine as long as the image quality is ok. > 2.If we have a T1 weighted sequence, slice with 6mm thickness and > 1 mm gap,does it make a difference on the brain volume calculation? The gap is not a problem - you have an effective slice thickness of 7mm and the header should show 7mm for pixdim3. However, unrelated to the gap, 7mm is rather thick, so you will not get as accurate SIENAX results as you would with thinner slices. > 3. In our studies,we observed that few cases have bad results > (brain vol less than 1200ml) even though the parameters of these > cases are the same as other cases with good results(vol between > 1300ml-1700ml)?what factors affect the calculation of the brain > volume? Please let us know. Probably just because your slices are rather thick....you should run SIENAX with the -d option and look carefully at the different output stages to check what's going on - for more hints on this, see the segmentation lecture slides on the FSL course page. Cheers. ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Wed, 30 May 2007 10:09:24 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: FSLView Error In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi -the current release hasn't been tweaked for FC6 - I suspect that the next release will be. In the meantime hopefully you should be able to get a local sysadmin to help you compile from source. Cheers. On 29 May 2007, at 19:55, Beau Sapach wrote: > Woops, sorry. It's Fedora Core 6. > > -----Original Message----- > From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On > Behalf > Of Steve Smith > Sent: Tuesday, May 29, 2007 1:10 AM > To: [log in to unmask] > Subject: Re: [FSL] FSLView Error > > What OS are you running? > > > > On 28 May 2007, at 18:51, Beau Sapach wrote: > >> Hello everyone, >> >> >> >> I've installed FSL from the Fedora Core 4 package and am getting >> errors with FSLView. I'm not sure if I've completed installation >> properly, I honestly was expecting an RPM package but nonetheless. >> I've simply unzipped and un-tar-ed the FSL package and then linked >> it to /usr/local/fsl. Lastly I followed the "Running" instructions >> (for bash shell in .bashrc) from the Downloading and Installing >> page. Is this all there is to it? Or have I missed something >> important? >> >> When I run FSLView the program crashes (and disappears from the >> screen) when I try to do File->Open. The output in the terminal >> window is included below. Does this error mean anything to anyone? >> My .scim folder doesn't even contain a global file. Any help would >> be much appreciated. >> >> >> >> Beau >> >> >> >> /usr/local/fsl/bin/fslview: line 6: 1119 Aborted >> >> ${FSLDIR}/lib/fslview/bin/fslview $@ >> >> ScimInputContextPlugin() >> >> WARNING: please edit ~/.scim/global and change /DefaultConfigModule >> to kconfig >> >> *** glibc detected *** /usr/local/fsl/lib/fslview/bin/fslview: free >> (): >> >> invalid pointer: 0x08d512e0 *** >> >> ======= Backtrace: ========= >> >> /lib/libc.so.6[0xc0f09d] >> >> /lib/libc.so.6(cfree+0x90)[0xc126f0] >> >> /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0x444c871] >> >> /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so >> (_ZN4scim23QScimInputContextGlobal10initializeEv+0xfd2)[0xb65fd3a2] >> >> /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so >> (_ZN4scim17QScimInputContextC1Ev+0x31e)[0xb65fe3fe] >> >> /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so >> (_ZN22ScimInputContextPlugin6createERK7QString+0x7e)[0xb65f75de] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN26QInputContextPluginPrivate6createERK7QString+0x1b)[0x4ba7c5b] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x5d) >> [0x4ba78ed] >> >> /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so >> (_ZN18QMultiInputContext17changeInputMethodE7QString+0xc3) >> [0xb668c51b] >> >> /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so >> (_ZN18QMultiInputContext5slaveEv+0x43)[0xb668c6cf] >> >> /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so >> (_ZN18QMultiInputContext15setHolderWidgetEP7QWidget+0x26)[0xb668c93e] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x75) >> [0x4ba7905] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget18createInputContextEv >> +0x72)[0x48fdb22] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget17resetInputContextEv >> +0x22)[0x48fddc2] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit7setTextERK7QString >> +0x22)[0x4a7d102] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit12setMaxLengthEi >> +0x36)[0x4a75fd6] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog4initEv+0x36c) >> [0x4b5618c] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN11QFileDialogC1ERK7QStringS2_P7QWidgetPKcb+0xa2)[0x4b68732] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN11QFileDialog15getOpenFileNameERK7QStringS2_P7QWidgetPKcS2_PS0_b >> +0x38b)[0x4b6c64b] >> >> /usr/local/fsl/lib/fslview/bin/fslview >> (_ZN17ApplicationWindow8fileOpenEv+0x3e)[0x80a9392] >> >> /usr/local/fsl/lib/fslview/bin/fslview >> (_ZN21ApplicationWindowBase9qt_invokeEiP8QUObject+0x74)[0x8143df0] >> >> /usr/local/fsl/lib/fslview/bin/fslview >> (_ZN17ApplicationWindow9qt_invokeEiP8QUObject+0x31)[0x8161ad9] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a) >> [0x499de2a] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEi >> +0xbd)[0x499e95d] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9activatedEv+0x29) >> [0x4d4df29] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction18internalActivationEv >> +0x50)[0x4af9a30] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9qt_invokeEiP8QUObject >> +0x10c)[0x4d4de1c] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a) >> [0x499de2a] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal6signalERK8QVariant >> +0x8e)[0x4d297ae] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal8activateEv+0x77) >> [0x49bc5f7] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN10QPopupMenu17mouseReleaseEventEP11QMouseEvent+0x475)[0x4abd3f5] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget5eventEP6QEvent+0x3c5) >> [0x49db2d5] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN12QApplication14internalNotifyEP7QObjectP6QEvent+0x9b)[0x49350fb] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN12QApplication6notifyEP7QObjectP6QEvent+0x287)[0x4936757] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN9QETWidget19translateMouseEventEPK7_XEvent+0xf40)[0x48cd8f0] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >> (_ZN12QApplication15x11ProcessEventEP7_XEvent+0x5e6)[0x48cba26] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop13processEventsEj >> +0x4eb)[0x48dd6ab] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop9enterLoopEv+0x42) >> [0x494e672] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop4execEv+0x26) >> [0x494e536] >> >> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication4execEv+0x1f) >> [0x4934c0f] >> >> /usr/local/fsl/lib/fslview/bin/fslview(main+0x11ca)[0x809e77a] >> >> /lib/libc.so.6(__libc_start_main+0xdc)[0xbbef2c] >> >> /usr/local/fsl/lib/fslview/bin/fslview >> (_ZN6QFrame10paintEventEP11QPaintEvent+0x3d)[0x809cba1] >> >> ======= Memory map: ======== >> >> 00101000-001ff000 r-xp 00000000 08:06 2638398 /usr/lib/libX11.so. >> 6.2.0 >> >> 001ff000-00203000 rwxp 000fe000 08:06 2638398 /usr/lib/libX11.so. >> 6.2.0 >> >> 0024c000-0026b000 r-xp 00000000 08:06 623523 /lib/libexpat.so. >> 0.5.0 >> >> 0026b000-0026d000 rwxp 0001e000 08:06 623523 /lib/libexpat.so. >> 0.5.0 >> >> 0026f000-0027e000 r-xp 00000000 08:06 2638403 /usr/lib/ >> libXext.so.6.4.0 >> >> 0027e000-0027f000 rwxp 0000e000 08:06 2638403 /usr/lib/ >> libXext.so.6.4.0 >> >> 00281000-002fe000 r-xp 00000000 08:06 2638393 /usr/lib/ >> libfreetype.so.6.3.10 >> >> 002fe000-00301000 rwxp 0007d000 08:06 2638393 /usr/lib/ >> libfreetype.so.6.3.10 >> >> 00303000-0032a000 r-xp 00000000 08:06 2638394 /usr/lib/ >> libfontconfig.so.1.1.0 >> >> 0032a000-00332000 rwxp 00027000 08:06 2638394 /usr/lib/ >> libfontconfig.so.1.1.0 >> >> 00334000-00359000 r-xp 00000000 08:06 2638395 /usr/lib/ >> libpng12.so.0.10.0 >> >> 00359000-0035a000 rwxp 00024000 08:06 2638395 /usr/lib/ >> libpng12.so.0.10.0 >> >> 0035c000-00364000 r-xp 00000000 08:06 2638399 /usr/lib/ >> libXrender.so.1.3.0 >> >> 00364000-00365000 rwxp 00007000 08:06 2638399 /usr/lib/ >> libXrender.so.1.3.0 >> >> 00367000-00370000 r-xp 00000000 08:06 2638408 /usr/lib/ >> libXcursor.so.1.0.2 >> >> 00370000-00371000 rwxp 00008000 08:06 2638408 /usr/lib/ >> libXcursor.so.1.0.2 >> >> 00373000-00375000 r-xp 00000000 08:06 2638404 /usr/lib/ >> libXinerama.so.1.0.0 >> >> 00375000-00376000 rwxp 00001000 08:06 2638404 /usr/lib/ >> libXinerama.so.1.0.0 >> >> 00378000-0037c000 r-xp 00000000 08:06 2638407 /usr/lib/ >> libXfixes.so.3.1.0 >> >> 0037c000-0037d000 rwxp 00003000 08:06 2638407 /usr/lib/ >> libXfixes.so.3.1.0 >> >> 0037f000-00382000 r-xp 00000000 08:06 2638406 /usr/lib/ >> libXrandr.so.2.0.0 >> >> 00382000-00383000 rwxp 00002000 08:06 2638406 /usr/lib/ >> libXrandr.so.2.0.0 >> >> 008db000-0092f000 r-xp 00000000 08:06 2621533 /usr/lib/libXt.so. >> 6.0.0 >> >> 0092f000-00933000 rwxp 00054000 08:06 2621533 /usr/lib/libXt.so. >> 6.0.0 >> >> 00955000-009bf000 r-xp 00000000 08:06 2638493 /usr/lib/libGL.so. >> 1.2 >> >> 009bf000-009c2000 rwxp 00069000 08:06 2638493 /usr/lib/libGL.so. >> 1.2 >> >> 009c2000-009c3000 rwxp 009c2000 00:00 0 >> >> 009c7000-009c9000 r-xp 00000000 08:06 2635087 /usr/lib/libscim- >> x11utils-1.0.so.8.1.0 >> >> 009c9000-009ca000 rwxp 00001000 08:06 2635087 /usr/lib/libscim- >> x11utils-1.0.so.8.1.0 >> >> > > > ---------------------------------------------------------------------- > -- > --- > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > ---------------------------------------------------------------------- > -- > --- ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Wed, 30 May 2007 12:23:39 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: wolf zinke <[log in to unmask]> Subject: Re: FSLView Error In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit I guess it could bet a problem that you have the scim packages installed. Try to identify these packages and uninstall them and maybe you could start fslview. At least this was the case for me. best, wolf Steve Smith wrote: > Hi -the current release hasn't been tweaked for FC6 - I suspect that > the next release will be. In the meantime hopefully you should be able > to get a local sysadmin to help you compile from source. > Cheers. > > > On 29 May 2007, at 19:55, Beau Sapach wrote: > >> Woops, sorry. It's Fedora Core 6. >> >> -----Original Message----- >> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On >> Behalf >> Of Steve Smith >> Sent: Tuesday, May 29, 2007 1:10 AM >> To: [log in to unmask] >> Subject: Re: [FSL] FSLView Error >> >> What OS are you running? >> >> >> >> On 28 May 2007, at 18:51, Beau Sapach wrote: >> >>> Hello everyone, >>> >>> >>> >>> I've installed FSL from the Fedora Core 4 package and am getting >>> errors with FSLView. I'm not sure if I've completed installation >>> properly, I honestly was expecting an RPM package but nonetheless. >>> I've simply unzipped and un-tar-ed the FSL package and then linked >>> it to /usr/local/fsl. Lastly I followed the "Running" instructions >>> (for bash shell in .bashrc) from the Downloading and Installing >>> page. Is this all there is to it? Or have I missed something >>> important? >>> >>> When I run FSLView the program crashes (and disappears from the >>> screen) when I try to do File->Open. The output in the terminal >>> window is included below. Does this error mean anything to anyone? >>> My .scim folder doesn't even contain a global file. Any help would >>> be much appreciated. >>> >>> >>> >>> Beau >>> >>> >>> >>> /usr/local/fsl/bin/fslview: line 6: 1119 Aborted >>> >>> ${FSLDIR}/lib/fslview/bin/fslview $@ >>> >>> ScimInputContextPlugin() >>> >>> WARNING: please edit ~/.scim/global and change /DefaultConfigModule >>> to kconfig >>> >>> *** glibc detected *** /usr/local/fsl/lib/fslview/bin/fslview: free(): >>> >>> invalid pointer: 0x08d512e0 *** >>> >>> ======= Backtrace: ========= >>> >>> /lib/libc.so.6[0xc0f09d] >>> >>> /lib/libc.so.6(cfree+0x90)[0xc126f0] >>> >>> /usr/lib/libstdc++.so.6(_ZdlPv+0x21)[0x444c871] >>> >>> /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so >>> (_ZN4scim23QScimInputContextGlobal10initializeEv+0xfd2)[0xb65fd3a2] >>> >>> /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so >>> (_ZN4scim17QScimInputContextC1Ev+0x31e)[0xb65fe3fe] >>> >>> /usr/lib/qt-3.3/plugins/inputmethods/libqscim.so >>> (_ZN22ScimInputContextPlugin6createERK7QString+0x7e)[0xb65f75de] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN26QInputContextPluginPrivate6createERK7QString+0x1b)[0x4ba7c5b] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x5d)[0x4ba78ed] >>> >>> /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so >>> (_ZN18QMultiInputContext17changeInputMethodE7QString+0xc3)[0xb668c51b] >>> >>> /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so >>> (_ZN18QMultiInputContext5slaveEv+0x43)[0xb668c6cf] >>> >>> /usr/lib/qt-3.3/plugins/inputmethods/libqimsw-multi.so >>> (_ZN18QMultiInputContext15setHolderWidgetEP7QWidget+0x26)[0xb668c93e] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN20QInputContextFactory6createERK7QStringP7QWidget+0x75)[0x4ba7905] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget18createInputContextEv >>> +0x72)[0x48fdb22] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget17resetInputContextEv >>> +0x22)[0x48fddc2] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit7setTextERK7QString >>> +0x22)[0x4a7d102] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN9QLineEdit12setMaxLengthEi >>> +0x36)[0x4a75fd6] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN11QFileDialog4initEv+0x36c) >>> [0x4b5618c] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN11QFileDialogC1ERK7QStringS2_P7QWidgetPKcb+0xa2)[0x4b68732] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN11QFileDialog15getOpenFileNameERK7QStringS2_P7QWidgetPKcS2_PS0_b >>> +0x38b)[0x4b6c64b] >>> >>> /usr/local/fsl/lib/fslview/bin/fslview >>> (_ZN17ApplicationWindow8fileOpenEv+0x3e)[0x80a9392] >>> >>> /usr/local/fsl/lib/fslview/bin/fslview >>> (_ZN21ApplicationWindowBase9qt_invokeEiP8QUObject+0x74)[0x8143df0] >>> >>> /usr/local/fsl/lib/fslview/bin/fslview >>> (_ZN17ApplicationWindow9qt_invokeEiP8QUObject+0x31)[0x8161ad9] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a) >>> [0x499de2a] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QObject15activate_signalEi >>> +0xbd)[0x499e95d] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9activatedEv+0x29) >>> [0x4d4df29] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction18internalActivationEv >>> +0x50)[0x4af9a30] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QAction9qt_invokeEiP8QUObject >>> +0x10c)[0x4d4de1c] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN7QObject15activate_signalEP15QConnectionListP8QUObject+0x16a) >>> [0x499de2a] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal6signalERK8QVariant >>> +0x8e)[0x4d297ae] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QSignal8activateEv+0x77) >>> [0x49bc5f7] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN10QPopupMenu17mouseReleaseEventEP11QMouseEvent+0x475)[0x4abd3f5] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN7QWidget5eventEP6QEvent+0x3c5) >>> [0x49db2d5] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN12QApplication14internalNotifyEP7QObjectP6QEvent+0x9b)[0x49350fb] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN12QApplication6notifyEP7QObjectP6QEvent+0x287)[0x4936757] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN9QETWidget19translateMouseEventEPK7_XEvent+0xf40)[0x48cd8f0] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3 >>> (_ZN12QApplication15x11ProcessEventEP7_XEvent+0x5e6)[0x48cba26] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop13processEventsEj >>> +0x4eb)[0x48dd6ab] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop9enterLoopEv+0x42) >>> [0x494e672] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN10QEventLoop4execEv+0x26) >>> [0x494e536] >>> >>> /usr/lib/qt-3.3/lib/libqt-mt.so.3(_ZN12QApplication4execEv+0x1f) >>> [0x4934c0f] >>> >>> /usr/local/fsl/lib/fslview/bin/fslview(main+0x11ca)[0x809e77a] >>> >>> /lib/libc.so.6(__libc_start_main+0xdc)[0xbbef2c] >>> >>> /usr/local/fsl/lib/fslview/bin/fslview >>> (_ZN6QFrame10paintEventEP11QPaintEvent+0x3d)[0x809cba1] >>> >>> ======= Memory map: ======== >>> >>> 00101000-001ff000 r-xp 00000000 08:06 2638398 /usr/lib/libX11.so. >>> 6.2.0 >>> >>> 001ff000-00203000 rwxp 000fe000 08:06 2638398 /usr/lib/libX11.so. >>> 6.2.0 >>> >>> 0024c000-0026b000 r-xp 00000000 08:06 623523 /lib/libexpat.so. >>> 0.5.0 >>> >>> 0026b000-0026d000 rwxp 0001e000 08:06 623523 /lib/libexpat.so. >>> 0.5.0 >>> >>> 0026f000-0027e000 r-xp 00000000 08:06 2638403 /usr/lib/ >>> libXext.so.6.4.0 >>> >>> 0027e000-0027f000 rwxp 0000e000 08:06 2638403 /usr/lib/ >>> libXext.so.6.4.0 >>> >>> 00281000-002fe000 r-xp 00000000 08:06 2638393 /usr/lib/ >>> libfreetype.so.6.3.10 >>> >>> 002fe000-00301000 rwxp 0007d000 08:06 2638393 /usr/lib/ >>> libfreetype.so.6.3.10 >>> >>> 00303000-0032a000 r-xp 00000000 08:06 2638394 /usr/lib/ >>> libfontconfig.so.1.1.0 >>> >>> 0032a000-00332000 rwxp 00027000 08:06 2638394 /usr/lib/ >>> libfontconfig.so.1.1.0 >>> >>> 00334000-00359000 r-xp 00000000 08:06 2638395 /usr/lib/ >>> libpng12.so.0.10.0 >>> >>> 00359000-0035a000 rwxp 00024000 08:06 2638395 /usr/lib/ >>> libpng12.so.0.10.0 >>> >>> 0035c000-00364000 r-xp 00000000 08:06 2638399 /usr/lib/ >>> libXrender.so.1.3.0 >>> >>> 00364000-00365000 rwxp 00007000 08:06 2638399 /usr/lib/ >>> libXrender.so.1.3.0 >>> >>> 00367000-00370000 r-xp 00000000 08:06 2638408 /usr/lib/ >>> libXcursor.so.1.0.2 >>> >>> 00370000-00371000 rwxp 00008000 08:06 2638408 /usr/lib/ >>> libXcursor.so.1.0.2 >>> >>> 00373000-00375000 r-xp 00000000 08:06 2638404 /usr/lib/ >>> libXinerama.so.1.0.0 >>> >>> 00375000-00376000 rwxp 00001000 08:06 2638404 /usr/lib/ >>> libXinerama.so.1.0.0 >>> >>> 00378000-0037c000 r-xp 00000000 08:06 2638407 /usr/lib/ >>> libXfixes.so.3.1.0 >>> >>> 0037c000-0037d000 rwxp 00003000 08:06 2638407 /usr/lib/ >>> libXfixes.so.3.1.0 >>> >>> 0037f000-00382000 r-xp 00000000 08:06 2638406 /usr/lib/ >>> libXrandr.so.2.0.0 >>> >>> 00382000-00383000 rwxp 00002000 08:06 2638406 /usr/lib/ >>> libXrandr.so.2.0.0 >>> >>> 008db000-0092f000 r-xp 00000000 08:06 2621533 /usr/lib/libXt.so. >>> 6.0.0 >>> >>> 0092f000-00933000 rwxp 00054000 08:06 2621533 /usr/lib/libXt.so. >>> 6.0.0 >>> >>> 00955000-009bf000 r-xp 00000000 08:06 2638493 /usr/lib/libGL.so.1.2 >>> >>> 009bf000-009c2000 rwxp 00069000 08:06 2638493 /usr/lib/libGL.so.1.2 >>> >>> 009c2000-009c3000 rwxp 009c2000 00:00 0 >>> >>> 009c7000-009c9000 r-xp 00000000 08:06 2635087 /usr/lib/libscim- >>> x11utils-1.0.so.8.1.0 >>> >>> 009c9000-009ca000 rwxp 00001000 08:06 2635087 /usr/lib/libscim- >>> x11utils-1.0.so.8.1.0 >>> >>> >> >> >> ------------------------------------------------------------------------ >> --- >> Stephen M. Smith, Professor of Biomedical Engineering >> Associate Director, Oxford University FMRIB Centre >> >> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >> +44 (0) 1865 222726 (fax 222717) >> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >> ------------------------------------------------------------------------ >> --- > > > --------------------------------------------------------------------------- > > Stephen M. Smith, Professor of Biomedical Engineering > Associate Director, Oxford University FMRIB Centre > > FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve > --------------------------------------------------------------------------- > > > ========================================================================= Date: Wed, 30 May 2007 06:36:32 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: ramya priya <[log in to unmask]> Subject: Re: SIENAX In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-1648182324-1180532192=:51669" Content-Transfer-Encoding: 8bit --0-1648182324-1180532192=:51669 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, Thank you very much sir. Priya Steve Smith <[log in to unmask]> wrote: Hi, On 30 May 2007, at 03:23, ramya priya wrote: > I have few questions regarding the calculation of brain volume > using SIENAX. > > 1. Is there any difference in the brain volume between 2 sequences > with and without any gap? Does SIENAX estimate the gap area into > the total brain volume when there is gap between slices? _Almost_ always in MRI analysis there is no need to differentiate between imaging with and without gaps, as long as the total effective slice thickness (including any gap) is correctly set in the header. Programs like SIENAX will work fine as long as the image quality is ok. > 2.If we have a T1 weighted sequence, slice with 6mm thickness and > 1 mm gap,does it make a difference on the brain volume calculation? The gap is not a problem - you have an effective slice thickness of 7mm and the header should show 7mm for pixdim3. However, unrelated to the gap, 7mm is rather thick, so you will not get as accurate SIENAX results as you would with thinner slices. > 3. In our studies,we observed that few cases have bad results > (brain vol less than 1200ml) even though the parameters of these > cases are the same as other cases with good results(vol between > 1300ml-1700ml)?what factors affect the calculation of the brain > volume? Please let us know. Probably just because your slices are rather thick....you should run SIENAX with the -d option and look carefully at the different output stages to check what's going on - for more hints on this, see the segmentation lecture slides on the FSL course page. Cheers. ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- --------------------------------- Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. --0-1648182324-1180532192=:51669 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 8bit
Hi,
 
Thank you very much sir.
 
Priya

Steve Smith <[log in to unmask]> wrote:
Hi,

On 30 May 2007, at 03:23, ramya priya wrote:

> I have few questions regarding the calculation of brain volume
> using SIENAX.
>
> 1. Is there any difference in the brain volume between 2 sequences
> with and without any gap? Does SIENAX estimate the gap area into
> the total brain volume when there is gap between slices?

_Almost_ always in MRI analysis there is no need to differentiate
between imaging with and without gaps, as long as the total effective
slice thickness (including any gap) is correctly set in the header.
Programs like SIENAX will work fine as long as the image quality is ok.

> 2.If we have a T1 weighted sequence, slice with 6mm thickness and
> 1 mm gap,does it make a difference on the brain volume calculation?

The gap is not a problem - you have an effective slice thickness of
7mm and the header should show 7mm for pixdim3. However, unrelated to
the gap, 7mm is rather thick, so you will not get as accurate SIENAX
results as you would with thinner slices.

> 3. In our studies,we observed that few cases have bad results
> (brain vol less than 1200ml) even though the parameters of these
> cases are the same as other cases with good results(vol between
> 1300ml-1700ml)?what factors affect the calculation of the brain
> volume? Please let us know.

Probably just because your slices are rather thick....you should run
SIENAX with the -d option and look carefully at the different output
stages to check what's going on - for more hints on this, see the
segmentation lecture slides on the FSL course page.

Cheers.

------------------------------------------------------------------------
---
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
------------------------------------------------------------------------
---


Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. --0-1648182324-1180532192=:51669-- ========================================================================= Date: Wed, 30 May 2007 15:10:39 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Greg Burgess <[log in to unmask]> Subject: Re: FEAT on multiple processors Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" Any idea when we might expect the next version of FSL with enhanced multi= - processing abilities? --Greg ========================================================================= Date: Wed, 30 May 2007 16:12:42 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: FEAT on multiple processors In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit We are aiming for the end of July but that date may well slip..... Cheers. On 30 May 2007, at 15:10, Greg Burgess wrote: > Any idea when we might expect the next version of FSL with enhanced > multi- > processing abilities? > > --Greg ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Wed, 30 May 2007 15:56:25 -0500 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Peter Fried <[log in to unmask]> Subject: ICA Meta-analysis MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, I am having some frustrations while trying to conduct a meta-analysis of resting-state FMRI data. Hopefully someone can help. So far, I have processed the raw data using pre-stats in FEAT and run MELODIC on each subject in their native space. Following the history of posts on this issue, I have tried both of the following using a test case of 4 subjects: 1) Transformed each subject's filtered_func_data (feat output) file to standard space; concatenated the data together across time using avwmerge++ -z; run melodic on the new 5D file. Problem: melodic finishes in < 5 sec w/ no results. 2) Transformed each subject's melodic_IC file (melodic output - 30 components) to standard space; smoothed the data using a 5 mm Gaussian kernel (avwmaths++ -fmean -kernel gauss 5); concatenated the data together across time (avwmerge++ -z); run melodic on the new 5D file. Problem: melodic hangs in the command window while eating up most available CPU speed and RAM (after 30 minutes of 0 progress I killed the process). It certainly seems possible that method (2) may eventually work, but if it takes all available CPU and most RAM over an extended period of time for only 4 subjects, I'd hate to see what happens when I run it on 60+. I'm not sure what's going on with method (1) since I have no problem running melodic on an individual subject's filtered_func_data. Any suggestions on what I'm doing wrong or alternative methods to attempt would be much appreciated. Cheers, Peter Fried Center for Magnetic Resonance Research University of Minnesota ========================================================================= Date: Wed, 30 May 2007 23:26:16 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Zografos Caramanos <[log in to unmask]> Subject: bug when using SIENAX v-2.4 with -2 flag Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="ISO-8859-1" Hello! I think that I may have found a bug in SIENAX v-2.4 (FSL v-3.3.11). When I run a 3-mm, axial, T1-weighted scan with a 3-tissue segmentation, SIENAX v-2.4 gives you normalized GREY, WHITE, and BRAIN volumes: (SIENAX v-2.4, 3-tissue segmentation) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Class:=09CSF=09=09tissue 1=09tissue 2=09brain % Volumes:=09205237.1 =09566241.4 =09665635.7 =090.857188 ---------- convert brain volume into normalised volume --------------= tissue volume unnormalised-volume GREY 760365.64 566240.07 WHITE 893834.69 665633.72 BRAIN 1654200.33 1231873.79 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D (...excerpted from below) BUT, if I run the same scan with the -2 flag for a 2-tissue segmentation,= SIENAX v-2.4 does not seem to properly calculate the normalized BRAIN vol= ume: (SIENAX v-2.4, 2-tissue segmentation) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Class:=09CSF=09=09tissue 1=09brain percentage Volumes:=09205237.1 =091231946.5 =090.857195 ---------- convert brain volume into normalised volume --------------= BRAIN 1 1 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D (...excerpted from further below) Please note that this is not the case when I run the same scan using SIENAX v-2.3 (FSL v-3.2beta, FAST v-3.5), which gives slightly different values: (SIENAX v-2.3, 3-tissue segmentation) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Class:=09CSF=09=09tissue 1=09tissue 2=09brain % Volumes:=09195260.0 =09562279.6 =09662799.2 =090.862526 ---------- convert brain volume into normalised volume --------------= GREY 757578.8396899881 WHITE 893006.5301093496 BRAIN 1650585.3697993377 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D (...excerpted from further below) (SIENAX v-2.3, 2-tissue segmentation) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Class:=09CSF=09=09tissue 1=09brain percentage Volumes:=09194299.3 =091226099.5 =090.863208 ---------- convert brain volume into normalised volume --------------= BRAIN 1651960.9049744302 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D (...excerpted from further below) Please let me know if I am doing something wrong, or if there is some fix= to this problem. Cheers! Aki Zografos Caramanos, M.A. Magnetic Resonance Studies Unit Montreal Neurological Institute, McGill University (e-mail) [log in to unmask] =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D +++++++++++++++++++++++++++++++++++++++++++++++ SIENAX Version 2.4 with 3-tissue segmentation +++++++++++++++++++++++++++++++++++++++++++++++ -----------------------------------------------------------------------= SIENA - Structural Image Evaluation, using Normalisation, of Atrophy part of FSL www.fmrib.ox.ac.uk/fsl running cross-sectional atrophy measurement: sienax version 2.4 sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d ---------- extract brain --------------------------------------------= ---------- register to stdairach space using brain and skull --------= (do not worry about histogram warnings) VSCALING 1.3428326384 ---------- mask with std mask ---------------------------------------= ---------- segment tissue into types --------------------------------= FAST - FMRIB's Automated Segmentation Tool=09=09Version 3.53 Image: /tmp/fsl_z5FSyo_mc_carz6809.2007-02-08.Scan_nt1p_none_stdmaskbra= in T1-weighted image Imagesize : 192 x 256 x 60 Pixelsize : 0.976562 x 0.976562 x 3 Initial K-means segmentation... 8 main iterations ... Estimating PVE... Segmentation done successfully! Calculation time 234 seconds! Write segmentation image /tmp/fsl_z5FSyo.fastout/QQQQQ_seg.hdr Successf= ully! Write PVE image of tissue 0 Successfully! Write PVE image of tissue 1 Successfully! Write PVE image of tissue 2 Successfully! Class:=09=09CSF=09=09tissue 1=09tissue 2=09brain percentage Volumes:=09205237.1 =09566241.4 =09665635.7 =090.857188 ---------- convert brain volume into normalised volume --------------= tissue volume unnormalised-volume GREY 760365.64 566240.07 WHITE 893834.69 665633.72 BRAIN 1654200.33 1231873.79 +++++++++++++++++++++++++++++++++++++++++++++++ SIENAX Version 2.4 with 2-tissue segmentation +++++++++++++++++++++++++++++++++++++++++++++++ -----------------------------------------------------------------------= SIENA - Structural Image Evaluation, using Normalisation, of Atrophy part of FSL www.fmrib.ox.ac.uk/fsl running cross-sectional atrophy measurement: sienax version 2.4 sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d -2 ---------- extract brain --------------------------------------------= ---------- register to stdairach space using brain and skull --------= (do not worry about histogram warnings) VSCALING 1.3428326384 ---------- mask with std mask ---------------------------------------= ---------- segment tissue into types --------------------------------= FAST - FMRIB's Automated Segmentation Tool=09=09Version 3.53 Image: /tmp/fsl_69X29e_mc_carz6809.2007-02-08.Scan_nt1p_none_stdmaskbra= in T1-weighted image Imagesize : 192 x 256 x 60 Pixelsize : 0.976562 x 0.976562 x 3 Initial K-means segmentation... 8 main iterations ... Estimating PVE... Segmentation done successfully! Calculation time 234 seconds! Write segmentation image /tmp/fsl_69X29e.fastout/QQQQQ_seg.hdr Successf= ully! Write PVE image of tissue 0 Successfully! Write PVE image of tissue 1 Successfully! Class:=09=09CSF=09=09tissue 1=09brain percentage Volumes:=09205237.1 =091231946.5 =090.857195 ---------- convert brain volume into normalised volume --------------= BRAIN 1 1 +++++++++++++++++++++++++++++++++++++++++++++++ SIENAX Version 2.3 with 3-tissue segmentation +++++++++++++++++++++++++++++++++++++++++++++++ -----------------------------------------------------------------------= SIENA - Structural Image Evaluation, using Normalisation, of Atrophy part of FSL www.fmrib.ox.ac.uk/fsl running cross-sectional atrophy measurement: sienax version 2.3 sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d ---------- extract brain --------------------------------------------= ---------- register to talairach space using brain and skull --------= (do not worry about histogram warnings) WARNING: Both reference and input images have an sform matrix set The output image will use the sform from the reference image WARNING: Both reference and input images have an sform matrix set The output image will use the sform from the reference image WARNING: Both reference and input images have an sform matrix set The output image will use the sform from the reference image VSCALING 1.3473268622 ---------- mask with talairach mask ---------------------------------= ---------- segment tissue into types --------------------------------= FAST - FMRIB's Automated Segmentation Tool=09=09Version 3.5 Image: /tmp/fsl_mEUvrB_mc_carz6809.2007-02-08.Scan_nt1p_none_talmaskbra= in T1-weighted image Imagesize : 192 x 256 x 60 Pixelsize : 0.976562 x 0.976562 x 3 Initial K-means segmentation... 8 main iterations ... Estimating PVE... Segmentation done successfully! Calculation time 234 seconds! Write segmentation image /tmp/fsl_mEUvrB.fastout/QQQQQ_seg.hdr Successf= ully! Write PVE image of tissue 0 Successfully! Write PVE image of tissue 1 Successfully! Write PVE image of tissue 2 Successfully! Class:=09=09CSF=09=09tissue 1=09tissue 2=09brain percentage Volumes:=09195260.0 =09562279.6 =09662799.2 =090.862526 ---------- convert brain volume into normalised volume --------------= GREY 757578.8396899881 WHITE 893006.5301093496 BRAIN 1650585.3697993377 +++++++++++++++++++++++++++++++++++++++++++++++ SIENAX Version 2.3 with 2-tissue segmentation +++++++++++++++++++++++++++++++++++++++++++++++ -----------------------------------------------------------------------= SIENA - Structural Image Evaluation, using Normalisation, of Atrophy part of FSL www.fmrib.ox.ac.uk/fsl running cross-sectional atrophy measurement: sienax version 2.3 sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d -2 ---------- extract brain --------------------------------------------= ---------- register to talairach space using brain and skull --------= (do not worry about histogram warnings) WARNING: Both reference and input images have an sform matrix set The output image will use the sform from the reference image WARNING: Both reference and input images have an sform matrix set The output image will use the sform from the reference image WARNING: Both reference and input images have an sform matrix set The output image will use the sform from the reference image VSCALING 1.3473268622 ---------- mask with talairach mask ---------------------------------= ---------- segment tissue into types --------------------------------= FAST - FMRIB's Automated Segmentation Tool=09=09Version 3.5 Image: /tmp/fsl_AlcxrX_mc_carz6809.2007-02-08.Scan_nt1p_none_talmaskbra= in T1-weighted image Imagesize : 192 x 256 x 60 Pixelsize : 0.976562 x 0.976562 x 3 Initial K-means segmentation... 4 main iterations ... Estimating PVE... Segmentation done successfully! Calculation time 89 seconds! Write segmentation image /tmp/fsl_AlcxrX.fastout/QQQQQ_seg.hdr Successf= ully! Write PVE image of tissue 0 Successfully! Write PVE image of tissue 1 Successfully! Class:=09=09CSF=09=09tissue 1=09brain percentage Volumes:=09194299.3 =091226099.5 =090.863208 ---------- convert brain volume into normalised volume --------------= BRAIN 1651960.9049744302 ========================================================================= Date: Wed, 30 May 2007 23:38:15 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Zografos Caramanos <[log in to unmask]> Subject: SIENA v-2.4 vs v-2.3 Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="ISO-8859-1" Hello again! Can anyone please describe/summarize the differences between running SIEN= A v-2.3 (FSL v-3.2beta, FLIRT version 5.2, FAST Version 3.5) and the curren= t SIENA v-2.4 (FSL v-3.3.11, FLIRT version 5.4.2, FAST Version 3.53) -- or point me to such a description/summary. My Scan-Rescan results with v-2.4 seem to suggest that it is more robust than v-2.3, but each run of SIENA v-2.4 takes about 60 minutes as opposed= to about 20 minutes for SIENA v-2.3 (running on a Dual 2GHz PowerPC G5 with 2-GB DDR SDRAM and Mac OS X v-10.4.8) and I am curious as to what is responsible for this. Cheers! Aki Zografos Caramanos, M.A. Magnetic Resonance Studies Unit Montreal Neurological Institute, McGill University (e-mail) [log in to unmask] ========================================================================= Date: Wed, 30 May 2007 16:04:51 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: James Kyle <[log in to unmask]> Subject: mixing nii.gz and .hdr cope images in 2nd lvl analysis Mime-Version: 1.0 (Apple Message framework v752.2) Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed We have some first level analysis 1/2 was run with fsl3.2 and is in analyze format (.hdr/.img), the other 1/2 was run with 3.3 and is in compressed nifti format (nii.gz) Off the bat, I'm wondering if this is a "Bad Idea(TM)". I'm consistently getting an $fmri(VOLUME$rawstats) value of 0, which in turn causes gnuplot to error out when creating the tsplots. Looks something like this: > /Volumes/local/fsl-testing/bin/smoothest -d 13 -m mask -r stats/ > res4d > stats/smoothness > WARNING: Extreme smoothness detected in X - possibly biased global > estimate. > WARNING: Extreme smoothness detected in Y - possibly biased global > estimate. > WARNING: Extreme smoothness detected in Z - possibly biased global > estimate. > > rm -f stats/res4d* > > /Volumes/local/fsl-testing/bin/avwmaths stats/zstat1 -mas mask > thresh_zstat1 > > echo 0 > thresh_zstat1.vol > zstat1: DLH=nan VOLUME=0 RESELS=nan > > /Volumes/local/fsl-testing/bin/avwmaths stats/zstat2 -mas mask > thresh_zstat2 > > echo 0 > thresh_zstat2.vol > zstat2: DLH=nan VOLUME=0 RESELS=nan > .......... > > mkdir tsplot > > /Volumes/local/fsl-testing/bin/tsplot . -f filtered_func_data -o > tsplot > > gnuplot> plot 'tsplot/tsplot_zstat1.txt' using 1 title 'data' with > lines , 'tsplot/tsplot_zstat1.txt' using 2 title 'cope partial > model fit' with lines , 'tsplot/tsplot_zstat1.txt' using 3 title > 'full model fit' with lines > > > > ^ > line 0: Can't plot with an empty y range! > > convertb: Not a GIF image file (./magicsOMaZK). > convertb: Missing an image file name. Is this indicative of a problem with the design or could it be a result of our mixed input formats? -james ========================================================================= Date: Thu, 31 May 2007 00:23:23 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christian Beckmann <[log in to unmask]> Subject: Re: ICA Meta-analysis In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi if you want to concatenate the 4t dimension with avwmerge (or ++) you'll need to use the -t option. The final results should always be 4D, not 5D. If melodic does not finish when using option1 what exactly is the error message? Wrt both options note that when transforming all images (or IC maps) to a common space you don't need to upsample to 2mm resolution. If you have all the transformation matrices computed then you can transform a 4D data set into standard space while keeping it at e.g. 4x4x4mm resolution using flirt flirt -in 4D -ref ${FSLDIR}/etc/standard/avg152 -out 4Dstd - appyisoxfm 4 transformation.mat hope this helps Christian On 30 May 2007, at 21:56, Peter Fried wrote: > Hi, > > I am having some frustrations while trying to conduct a meta- > analysis of > resting-state FMRI data. Hopefully someone can help. > > So far, I have processed the raw data using pre-stats in FEAT and run > MELODIC on each subject in their native space. > > Following the history of posts on this issue, I have tried both of the > following using a test case of 4 subjects: > > 1) Transformed each subject's filtered_func_data (feat output) file to > standard space; concatenated the data together across time using > avwmerge++ -z; run melodic on the new 5D file. > > Problem: melodic finishes in < 5 sec w/ no results. > > 2) Transformed each subject's melodic_IC file (melodic output - 30 > components) to standard space; smoothed the data using a 5 mm Gaussian > kernel (avwmaths++ -fmean -kernel gauss 5); concatenated the data > together > across time (avwmerge++ -z); run melodic on the new 5D file. > > Problem: melodic hangs in the command window while eating up most > available CPU speed and RAM (after 30 minutes of 0 progress I > killed the > process). > > It certainly seems possible that method (2) may eventually work, > but if it > takes all available CPU and most RAM over an extended period of > time for > only 4 subjects, I'd hate to see what happens when I run it on 60+. > > I'm not sure what's going on with method (1) since I have no problem > running melodic on an individual subject's filtered_func_data. > > Any suggestions on what I'm doing wrong or alternative methods to > attempt > would be much appreciated. > > Cheers, > > Peter Fried > Center for Magnetic Resonance Research > University of Minnesota ____ Christian F. Beckmann University Research Lecturer Oxford University Centre for Functional MRI of the Brain (FMRIB) John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann tel: +44 1865 222551 fax: +44 1865 222717 ========================================================================= Date: Wed, 30 May 2007 19:05:56 -0500 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Paul Geha <[log in to unmask]> Subject: Motion correction for DTI Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="=====================_36037921==.ALT" --=====================_36037921==.ALT Content-Type: text/plain; charset="us-ascii"; format=flowed Hello, I have DTI data collected with 68 ( 8 no diffusion scans) different directions, with two consecutive scans, that are interleaved to cover the whole brain. In some of them the subject moved from one scan to the other so that when I reconstruct the image there is an obvious shifts between the 2. How can I correct for that. Can I use Flirt with 6 DOF only to register one to the other and then scramble them into one image? Is this transformation going to change the voxel values due to interpolation? thanks for the help Paul Paul Geha M.D. Northwestern University The Feinberg School of Medicine Department of Physiology M211 303 E. Chicago Ave. Chicago, IL 60611 Tel:312-503 2886 Fax: 312-503-5101 --=====================_36037921==.ALT Content-Type: text/html; charset="us-ascii" Hello,

I have DTI data collected with 68  ( 8 no diffusion scans) different directions, with two consecutive  scans, that are interleaved to cover the whole brain.  In some of them the subject moved from one scan to the other so that when I reconstruct the image there is an obvious shifts between the 2.  How can I correct for that.  Can I use Flirt  with 6 DOF only to register one to the other and then scramble them into one image?   Is this transformation going to change the voxel values due to interpolation?
thanks for the help
Paul




Paul Geha M.D.
Northwestern University
The Feinberg School of Medicine
Department of Physiology M211
303 E. Chicago Ave.
Chicago, IL 60611
Tel:312-503 2886
Fax: 312-503-5101

--=====================_36037921==.ALT-- ========================================================================= Date: Wed, 30 May 2007 20:58:43 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Re: Export jpeg from FSLVIEW In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_8573_16180734.1180573123289" ------=_Part_8573_16180734.1180573123289 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Thanks Matt, This is very helpful! I have been taking a snapshot in OS X which also works well. Best, Marc On 5/29/07, Riccardo Della Nave <[log in to unmask]> wrote: > > If you have KDE try KSnapshot with "Capture mode: Section of Window" > > On Monday 28 May 2007 19:47:43 Marc Dubin wrote: > > Thanks, Matt! > > > > On 5/28/07, Matt <[log in to unmask]> wrote: > > > Marc, > > > > > > There isn't a way to do that; you need to take a screen capture. > > > > > > Peace, > > > > > > Matt. > > > > > > -----Original Message----- > > > From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On > > > Behalf Of Marc Dubin > > > Sent: Monday, May 28, 2007 1:31 PM > > > To: [log in to unmask] > > > Subject: [FSL] Export jpeg from FSLVIEW > > > > > > Does anyone know how to export one of the images from fslview > > > in jpeg or gif or tiff format? > > > > > > Thanks, > > > Marc > -- 646 831 8886 [log in to unmask] ------=_Part_8573_16180734.1180573123289 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline

Thanks Matt,
This is very helpful!
I have been taking a snapshot in OS X
which also works well.
 
Best,
Marc

 
On 5/29/07, Riccardo Della Nave <[log in to unmask]> wrote:
If you have KDE try KSnapshot with "Capture mode: Section of Window"

On Monday 28 May 2007 19:47:43 Marc Dubin wrote:
> Thanks, Matt!
>
> On 5/28/07, Matt <[log in to unmask]> wrote:
> > Marc,
> >
> > There isn't a way to do that; you need to take a screen capture.
> >
> > Peace,
> >
> > Matt.
> >
> > -----Original Message-----
> > From: FSL - FMRIB's Software Library [mailto:[log in to unmask] ] On
> > Behalf Of Marc Dubin
> > Sent: Monday, May 28, 2007 1:31 PM
> > To: [log in to unmask]
> > Subject: [FSL] Export jpeg from FSLVIEW
> >
> > Does anyone know how to export one of the images from fslview
> > in jpeg or gif or tiff format?
> >
> > Thanks,
> > Marc



--
646 831 8886
[log in to unmask] ------=_Part_8573_16180734.1180573123289-- ========================================================================= Date: Thu, 31 May 2007 02:05:44 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: Retrieving pixelwise R-Squared in TBSS Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" Dear FSL Users, I'm using TBSS/randomise/Glm to do a voxelwise correlation analysis with a cognitive measure and was wondering how to=20 retrieve the FA's of a given voxel that go into the correlation for that voxel as well as the R-squared and p-value. Best, Marc ========================================================================= Date: Thu, 31 May 2007 02:55:06 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Marc Dubin <[log in to unmask]> Subject: TBSS: Overlaying mean_FA_skeleton on target FA RGB Map Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" Dear FSL Users, TO aid in the identification of specific tracts in the mean_FA_skeleton I= =20 was thinking of overlaying it on the RGB map of the registration target F= A=20 image. Have others found that to be a useful approach or are there other=20= approaches people have tried? Best, Marc ========================================================================= Date: Thu, 31 May 2007 08:31:02 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Andreas Bartsch <[log in to unmask]> Subject: AW: [FSL] Motion correction for DTI MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C7A34D.4257D08F" This is a multi-part message in MIME format. ------_=_NextPart_001_01C7A34D.4257D08F Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi, I guess it will depend on what you wanna do with the data. It won't be easy to combine the two runs - because of voxel intensity AND diffusion gradient issues. So I would probably analyze the two runs separately and combine them (e.g. by using the FA image or whatever) later on. Having said this, you may wanna try the CATNAP toolbox from the BIRN (which uses flirt, of course;). Cheers- Andreas ________________________________ Von: FSL - FMRIB's Software Library [mailto:[log in to unmask]] Im Auftrag von Paul Geha Gesendet: Donnerstag, 31. Mai 2007 02:06 An: [log in to unmask] Betreff: [FSL] Motion correction for DTI Hello,=20 I have DTI data collected with 68 ( 8 no diffusion scans) different directions, with two consecutive scans, that are interleaved to cover the whole brain. In some of them the subject moved from one scan to the other so that when I reconstruct the image there is an obvious shifts between the 2. How can I correct for that. Can I use Flirt with 6 DOF only to register one to the other and then scramble them into one image? Is this transformation going to change the voxel values due to interpolation?=20 thanks for the help Paul Paul Geha M.D.=20 Northwestern University=20 The Feinberg School of Medicine=20 Department of Physiology M211=20 303 E. Chicago Ave.=20 Chicago, IL 60611=20 Tel:312-503 2886 Fax: 312-503-5101=20 ------_=_NextPart_001_01C7A34D.4257D08F Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable
Hi,
I guess it will depend on what you wanna do = with the data.=20 It won't be easy to combine the two runs - because of voxel intensity = AND=20 diffusion gradient issues. So I would probably analyze the two runs = separately=20 and combine them (e.g. by using the FA image or whatever) later on. = Having said=20 this, you may wanna try the CATNAP toolbox from the BIRN (which uses = flirt, of=20 course;).
Cheers-
Andreas


Von: FSL - FMRIB's Software Library=20 [mailto:[log in to unmask]] Im Auftrag von Paul = Geha
Gesendet:=20 Donnerstag, 31. Mai 2007 02:06
An:=20 [log in to unmask]
Betreff: [FSL] Motion correction for=20 DTI

Hello,

I have DTI data collected with 68  ( 8 no = diffusion scans) different directions, with two consecutive  scans, = that=20 are interleaved to cover the whole brain.  In some of them the = subject=20 moved from one scan to the other so that when I reconstruct the image = there is=20 an obvious shifts between the 2.  How can I correct for that.  = Can I=20 use Flirt  with 6 DOF only to register one to the other and then = scramble=20 them into one image?   Is this transformation going to change = the=20 voxel values due to interpolation?
thanks for the=20 help
Paul




Paul Geha M.D.
Northwestern University
The = Feinberg School=20 of Medicine
Department of Physiology M211
303 E. Chicago Ave.=20
Chicago, IL 60611
Tel:312-503 2886
Fax: 312-503-5101=20

------_=_NextPart_001_01C7A34D.4257D08F-- ========================================================================= Date: Thu, 31 May 2007 07:39:20 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: mixing nii.gz and .hdr cope images in 2nd lvl analysis In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Yes - it's quite likely suboptimal or worse to mix things up in this way. I would recommend just re-running the first-level analyses with 3.3. Cheers. On 31 May 2007, at 00:04, James Kyle wrote: > We have some first level analysis 1/2 was run with fsl3.2 and is in > analyze format (.hdr/.img), the other 1/2 was run with 3.3 and is > in compressed nifti format (nii.gz) > > Off the bat, I'm wondering if this is a "Bad Idea(TM)". I'm > consistently getting an $fmri(VOLUME$rawstats) value of 0, which in > turn causes gnuplot to error out when creating the tsplots. > > Looks something like this: > >> /Volumes/local/fsl-testing/bin/smoothest -d 13 -m mask -r stats/ >> res4d > stats/smoothness >> WARNING: Extreme smoothness detected in X - possibly biased global >> estimate. >> WARNING: Extreme smoothness detected in Y - possibly biased global >> estimate. >> WARNING: Extreme smoothness detected in Z - possibly biased global >> estimate. >> >> rm -f stats/res4d* >> >> /Volumes/local/fsl-testing/bin/avwmaths stats/zstat1 -mas mask >> thresh_zstat1 >> >> echo 0 > thresh_zstat1.vol >> zstat1: DLH=nan VOLUME=0 RESELS=nan >> >> /Volumes/local/fsl-testing/bin/avwmaths stats/zstat2 -mas mask >> thresh_zstat2 >> >> echo 0 > thresh_zstat2.vol >> zstat2: DLH=nan VOLUME=0 RESELS=nan >> .......... >> >> mkdir tsplot >> >> /Volumes/local/fsl-testing/bin/tsplot . -f filtered_func_data -o >> tsplot >> >> gnuplot> plot 'tsplot/tsplot_zstat1.txt' using 1 title 'data' with >> lines , 'tsplot/tsplot_zstat1.txt' using 2 title 'cope partial >> model fit' with lines , 'tsplot/tsplot_zstat1.txt' using 3 title >> 'full model fit' with lines >> >> >> >> ^ >> line 0: Can't plot with an empty y range! >> >> convertb: Not a GIF image file (./magicsOMaZK). >> convertb: Missing an image file name. > > Is this indicative of a problem with the design or could it be a > result of our mixed input formats? > > -james ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Thu, 31 May 2007 07:41:57 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: TBSS: Overlaying mean_FA_skeleton on target FA RGB Map In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Indeed that could help, as well as looking at the Mori white matter atlas web page or book (lots more detail in the book). On 31 May 2007, at 02:55, Marc Dubin wrote: > Dear FSL Users, > > TO aid in the identification of specific tracts in the > mean_FA_skeleton I > was thinking of overlaying it on the RGB map of the registration > target FA > image. Have others found that to be a useful approach or are there > other > approaches people have tried? > > Best, > Marc ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Thu, 31 May 2007 07:45:15 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: Retrieving pixelwise R-Squared in TBSS In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit You can use avwmeants to extract the "timeseries" (values across subjects) for a given voxel in all_FA_skeletonised using the -c option. Randomise uses the GLM, which is a multiple regression not a correlation - so the raw value you start with from the stats is the tstat that randomise outputs. Steve. On 31 May 2007, at 02:05, Marc Dubin wrote: > Dear FSL Users, > > I'm using TBSS/randomise/Glm to do a voxelwise correlation > analysis with a cognitive measure and was wondering how to > retrieve the FA's of a given voxel that go into the correlation > for that voxel as well as the R-squared and p-value. > > Best, > Marc ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Thu, 31 May 2007 13:44:50 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christina Hugenschmidt <[log in to unmask]> Subject: y-axis of timecourse output in MELODIC Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="windows-1252" Hi! Thank you so much Christian for your answers to my last questions-they we= re=20 most helpful and have given me a lot to think about. I am almost embarassed to ask this one, but I want to be sure that I am=20= interpreting what I am seeing correctly. What are the units on the y-axis= =20 of the timecourse output in MELODIC on the report page (the top plot that= =20 says Timecourse (in seconds); TR=3Dx)? Thank you! Christina ========================================================================= Date: Thu, 31 May 2007 14:06:20 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: SIENA v-2.4 vs v-2.3 In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Content-Transfer-Encoding: 7bit Hi - there are no major conceptual changes - just a series of minor tweaks in different programs used. On linux the two versions take pretty much the same time. On a G5 Mac here I'm getting 1765s for FSL3.2 and 2543s for FSL3.3 actual compute time on an example dataset. I think this is mostly explained by extra robustness checking in FLIRT and in the conversion of siena_diff from C to C++ for compatibility with the rest of FSL (we've nearly finished killing off all the old C code). Both versions give almost the same answer (-3.708 and -3.658). Hence I think all is well, and would just recommend using purely FSL3.3 for these analyses. Cheers. On 30 May 2007, at 23:38, Zografos Caramanos wrote: > Hello again! > > Can anyone please describe/summarize the differences between > running SIENA > v-2.3 (FSL v-3.2beta, FLIRT version 5.2, FAST Version 3.5) and the > current > SIENA v-2.4 (FSL v-3.3.11, FLIRT version 5.4.2, FAST Version 3.53) > -- or > point me to such a description/summary. > > My Scan-Rescan results with v-2.4 seem to suggest that it is more > robust > than v-2.3, but each run of SIENA v-2.4 takes about 60 minutes as > opposed to > about 20 minutes for SIENA v-2.3 (running on a Dual 2GHz PowerPC G5 > with > 2-GB DDR SDRAM and Mac OS X v-10.4.8) and I am curious as to what is > responsible for this. > > Cheers! > Aki > > Zografos Caramanos, M.A. > Magnetic Resonance Studies Unit > Montreal Neurological Institute, McGill University > (e-mail) [log in to unmask] ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- ========================================================================= Date: Thu, 31 May 2007 08:46:33 -0500 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Peter Fried <[log in to unmask]> Subject: Re: ICA Meta-analysis In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit Thanks for the feedback. I was using avwmerge -t, I'm not sure why I wrote -z. I will try letting melodic run for a few hours on the concatenated data and see if it eventually finishes or times out with an error. Also, I will try the transform you mention. Cheers, Pete > Hi > > if you want to concatenate the 4t dimension with avwmerge (or ++) > you'll need to use the -t option. The final results should always be > 4D, not 5D. If melodic does not finish when using option1 what > exactly is the error message? Wrt both options note that when > transforming all images (or IC maps) to a common space you don't need > to upsample to 2mm resolution. If you have all the transformation > matrices computed then you can transform a 4D data set into standard > space while keeping it at e.g. 4x4x4mm resolution using flirt > > flirt -in 4D -ref ${FSLDIR}/etc/standard/avg152 -out 4Dstd - > appyisoxfm 4 transformation.mat > > hope this helps > Christian > > > > > On 30 May 2007, at 21:56, Peter Fried wrote: > >> Hi, >> >> I am having some frustrations while trying to conduct a meta- >> analysis of >> resting-state FMRI data. Hopefully someone can help. >> >> So far, I have processed the raw data using pre-stats in FEAT and run >> MELODIC on each subject in their native space. >> >> Following the history of posts on this issue, I have tried both of the >> following using a test case of 4 subjects: >> >> 1) Transformed each subject's filtered_func_data (feat output) file to >> standard space; concatenated the data together across time using >> avwmerge++ -z; run melodic on the new 5D file. >> >> Problem: melodic finishes in < 5 sec w/ no results. >> >> 2) Transformed each subject's melodic_IC file (melodic output - 30 >> components) to standard space; smoothed the data using a 5 mm Gaussian >> kernel (avwmaths++ -fmean -kernel gauss 5); concatenated the data >> together >> across time (avwmerge++ -z); run melodic on the new 5D file. >> >> Problem: melodic hangs in the command window while eating up most >> available CPU speed and RAM (after 30 minutes of 0 progress I >> killed the >> process). >> >> It certainly seems possible that method (2) may eventually work, >> but if it >> takes all available CPU and most RAM over an extended period of >> time for >> only 4 subjects, I'd hate to see what happens when I run it on 60+. >> >> I'm not sure what's going on with method (1) since I have no problem >> running melodic on an individual subject's filtered_func_data. >> >> Any suggestions on what I'm doing wrong or alternative methods to >> attempt >> would be much appreciated. >> >> Cheers, >> >> Peter Fried >> Center for Magnetic Resonance Research >> University of Minnesota > > ____ > Christian F. Beckmann > University Research Lecturer > Oxford University Centre for Functional MRI of the Brain (FMRIB) > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. > [log in to unmask] http://www.fmrib.ox.ac.uk/~beckmann > tel: +44 1865 222551 fax: +44 1865 222717 > ========================================================================= Date: Thu, 31 May 2007 09:40:34 -0500 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Paul Geha <[log in to unmask]> Subject: Re: AW: [FSL] Motion correction for DTI In-Reply-To: <[log in to unmask] -wuerzburg.de> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="=====================_2655734==.ALT" --=====================_2655734==.ALT Content-Type: text/plain; charset="us-ascii"; format=flowed Thank you Andreas for the reply, is there any specific way to combine the two FA maps in the end using FSL only, or do I have to use CATNAP? Paul At 01:31 AM 5/31/2007, you wrote: >Hi, >I guess it will depend on what you wanna do with the data. It won't >be easy to combine the two runs - because of voxel intensity AND >diffusion gradient issues. So I would probably analyze the two runs >separately and combine them (e.g. by using the FA image or whatever) >later on. Having said this, you may wanna try the CATNAP toolbox >from the BIRN (which uses flirt, of course;). >Cheers- >Andreas > > >---------- >Von: FSL - FMRIB's Software Library [mailto:[log in to unmask]] Im >Auftrag von Paul Geha >Gesendet: Donnerstag, 31. Mai 2007 02:06 >An: [log in to unmask] >Betreff: [FSL] Motion correction for DTI > >Hello, > >I have DTI data collected with 68 ( 8 no diffusion scans) different >directions, with two consecutive scans, that are interleaved to >cover the whole brain. In some of them the subject moved from one >scan to the other so that when I reconstruct the image there is an >obvious shifts between the 2. How can I correct for that. Can I >use Flirt with 6 DOF only to register one to the other and then >scramble them into one image? Is this transformation going to >change the voxel values due to interpolation? >thanks for the help >Paul > > > > >Paul Geha M.D. >Northwestern University >The Feinberg School of Medicine >Department of Physiology M211 >303 E. Chicago Ave. >Chicago, IL 60611 >Tel:312-503 2886 >Fax: 312-503-5101 Paul Geha M.D. Northwestern University The Feinberg School of Medicine Department of Physiology M211 303 E. Chicago Ave. Chicago, IL 60611 Tel:312-503 2886 Fax: 312-503-5101 --=====================_2655734==.ALT Content-Type: text/html; charset="us-ascii" Thank you Andreas for the reply,
is there any specific way to combine the two FA maps in the end using FSL only, or do I have to use CATNAP?
Paul



At 01:31 AM 5/31/2007, you wrote:
Hi,
I guess it will depend on what you wanna do with the data. It won't be easy to combine the two runs - because of voxel intensity AND diffusion gradient issues. So I would probably analyze the two runs separately and combine them (e.g. by using the FA image or whatever) later on. Having said this, you may wanna try the CATNAP toolbox from the BIRN (which uses flirt, of course;).
Cheers-
Andreas


Von: FSL - FMRIB's Software Library [ mailto:[log in to unmask]] Im Auftrag von Paul Geha
Gesendet: Donnerstag, 31. Mai 2007 02:06
An: [log in to unmask]
Betreff: [FSL] Motion correction for DTI

Hello,

I have DTI data collected with 68  ( 8 no diffusion scans) different directions, with two consecutive  scans, that are interleaved to cover the whole brain.  In some of them the subject moved from one scan to the other so that when I reconstruct the image there is an obvious shifts between the 2.  How can I correct for that.  Can I use Flirt  with 6 DOF only to register one to the other and then scramble them into one image?   Is this transformation going to change the voxel values due to interpolation?
thanks for the help
Paul




Paul Geha M.D.
Northwestern University
The Feinberg School of Medicine
Department of Physiology M211
303 E. Chicago Ave.
Chicago, IL 60611
Tel:312-503 2886
Fax: 312-503-5101

Paul Geha M.D.
Northwestern University
The Feinberg School of Medicine
Department of Physiology M211
303 E. Chicago Ave.
Chicago, IL 60611
Tel:312-503 2886
Fax: 312-503-5101

--=====================_2655734==.ALT-- ========================================================================= Date: Thu, 31 May 2007 16:44:44 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Andreas Bartsch <[log in to unmask]> Subject: AW: [FSL] AW: [FSL] Motion correction for DTI MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C7A392.3A5DD470" This is a multi-part message in MIME format. ------_=_NextPart_001_01C7A392.3A5DD470 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Well, you could average them using avwmaths(++) or supply both to TBSS depending on what you want. Cheers- Andreas ________________________________ Von: FSL - FMRIB's Software Library [mailto:[log in to unmask]] Im Auftrag von Paul Geha Gesendet: Donnerstag, 31. Mai 2007 16:41 An: [log in to unmask] Betreff: Re: [FSL] AW: [FSL] Motion correction for DTI Thank you Andreas for the reply,=20 is there any specific way to combine the two FA maps in the end using FSL only, or do I have to use CATNAP? Paul At 01:31 AM 5/31/2007, you wrote: Hi, I guess it will depend on what you wanna do with the data. It won't be easy to combine the two runs - because of voxel intensity AND diffusion gradient issues. So I would probably analyze the two runs separately and combine them (e.g. by using the FA image or whatever) later on. Having said this, you may wanna try the CATNAP toolbox from the BIRN (which uses flirt, of course;). Cheers- Andreas =09 =09 ________________________________ Von: FSL - FMRIB's Software Library [ mailto:[log in to unmask] ] Im Auftrag von Paul Geha Gesendet: Donnerstag, 31. Mai 2007 02:06 An: [log in to unmask] Betreff: [FSL] Motion correction for DTI =09 Hello,=20 =09 I have DTI data collected with 68 ( 8 no diffusion scans) different directions, with two consecutive scans, that are interleaved to cover the whole brain. In some of them the subject moved from one scan to the other so that when I reconstruct the image there is an obvious shifts between the 2. How can I correct for that. Can I use Flirt with 6 DOF only to register one to the other and then scramble them into one image? Is this transformation going to change the voxel values due to interpolation?=20 thanks for the help Paul =09 =09 =09 =09 Paul Geha M.D.=20 Northwestern University=20 The Feinberg School of Medicine=20 Department of Physiology M211=20 303 E. Chicago Ave.=20 Chicago, IL 60611=20 Tel:312-503 2886 Fax: 312-503-5101=20 =09 Paul Geha M.D.=20 Northwestern University=20 The Feinberg School of Medicine=20 Department of Physiology M211=20 303 E. Chicago Ave.=20 Chicago, IL 60611=20 Tel:312-503 2886 Fax: 312-503-5101=20 ------_=_NextPart_001_01C7A392.3A5DD470 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Well, you could average them using avwmaths(++) = or supply=20 both to TBSS depending on what you want.
Cheers-
Andreas


Von: FSL - FMRIB's Software Library=20 [mailto:[log in to unmask]] Im Auftrag von Paul = Geha
Gesendet:=20 Donnerstag, 31. Mai 2007 16:41
An:=20 [log in to unmask]
Betreff: Re: [FSL] AW: [FSL] Motion = correction for=20 DTI

Thank you Andreas for the reply,
is there any specific = way to=20 combine the two FA maps in the end using FSL only, or do I have to use=20 CATNAP?
Paul



At 01:31 AM 5/31/2007, you wrote:
Hi,
I=20 guess it will depend on what you wanna do with the data. It won't be = easy to=20 combine the two runs - because of voxel intensity AND diffusion = gradient=20 issues. So I would probably analyze the two runs separately and = combine them=20 (e.g. by using the FA image or whatever) later on. Having said this, = you may=20 wanna try the CATNAP toolbox from the BIRN (which uses flirt, of=20 course;).
Cheers-
Andreas


Von: FSL - FMRIB's Software = Library [=20 mailto:[log in to unmask]] Im Auftrag von Paul=20 Geha
Gesendet: Donnerstag, 31. Mai 2007 02:06
An:=20 [log in to unmask]
Betreff: [FSL] Motion correction for=20 DTI

Hello,

I have DTI data collected with = 68  ( 8=20 no diffusion scans) different directions, with two consecutive  = scans,=20 that are interleaved to cover the whole brain.  In some of them = the=20 subject moved from one scan to the other so that when I reconstruct = the image=20 there is an obvious shifts between the 2.  How can I correct for=20 that.  Can I use Flirt  with 6 DOF only to register one to = the other=20 and then scramble them into one image?   Is this = transformation=20 going to change the voxel values due to interpolation?
thanks for = the=20 help
Paul




Paul Geha M.D.
Northwestern = University=20
The Feinberg School of Medicine
Department of Physiology M211 =
303=20 E. Chicago Ave.
Chicago, IL 60611
Tel:312-503 2886
Fax:=20 312-503-5101

Paul Geha M.D.=20
Northwestern University
The Feinberg School of Medicine =
Department=20 of Physiology M211
303 E. Chicago Ave.
Chicago, IL 60611 =
Tel:312-503=20 2886
Fax: 312-503-5101

------_=_NextPart_001_01C7A392.3A5DD470-- ========================================================================= Date: Thu, 31 May 2007 16:47:18 +0200 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Andreas Bartsch <[log in to unmask]> Subject: AW: [FSL] y-axis of timecourse output in MELODIC MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi, It is standard deviation units which is in fact a very clever way of = expressing the signal fluctuation (much better than percentages of = change, I would say). So it is centered around zero with a SD of 1 = across the entire time-series. Cheers- Andreas=20 -----Urspr=FCngliche Nachricht----- Von: FSL - FMRIB's Software Library [mailto:[log in to unmask]] Im = Auftrag von Christina Hugenschmidt Gesendet: Donnerstag, 31. Mai 2007 14:45 An: [log in to unmask] Betreff: [FSL] y-axis of timecourse output in MELODIC Hi! Thank you so much Christian for your answers to my last questions-they = were most helpful and have given me a lot to think about. I am almost embarassed to ask this one, but I want to be sure that I am = interpreting what I am seeing correctly. What are the units on the = y-axis of the timecourse output in MELODIC on the report page (the top = plot that says Timecourse (in seconds); TR=3Dx)? Thank you! Christina ========================================================================= Date: Thu, 31 May 2007 10:51:16 -0400 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Christina Hugenschmidt <[log in to unmask]> Subject: Re: AW: [FSL] y-axis of timecourse output in MELODIC In-Reply-To: A<[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Thank you very much Andreas! Regards, Christina Christina Hugenschmidt Graduate Student, Neuroscience Program Wake Forest University School of Medicine [log in to unmask] (336) 716-0972 =20 -----Original Message----- From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On = Behalf Of Andreas Bartsch Sent: Thursday, May 31, 2007 10:47 AM To: [log in to unmask] Subject: [FSL] AW: [FSL] y-axis of timecourse output in MELODIC Hi, It is standard deviation units which is in fact a very clever way of = expressing the signal fluctuation (much better than percentages of = change, I would say). So it is centered around zero with a SD of 1 = across the entire time-series. Cheers- Andreas=20 -----Urspr=FCngliche Nachricht----- Von: FSL - FMRIB's Software Library [mailto:[log in to unmask]] Im = Auftrag von Christina Hugenschmidt Gesendet: Donnerstag, 31. Mai 2007 14:45 An: [log in to unmask] Betreff: [FSL] y-axis of timecourse output in MELODIC Hi! Thank you so much Christian for your answers to my last questions-they = were most helpful and have given me a lot to think about. I am almost embarassed to ask this one, but I want to be sure that I am = interpreting what I am seeing correctly. What are the units on the = y-axis of the timecourse output in MELODIC on the report page (the top = plot that says Timecourse (in seconds); TR=3Dx)? Thank you! Christina ========================================================================= Date: Thu, 31 May 2007 16:04:52 +0100 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Steve Smith <[log in to unmask]> Subject: Re: bug when using SIENAX v-2.4 with -2 flag In-Reply-To: <[log in to unmask]> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: multipart/mixed; boundary=Apple-Mail-9-453229316 --Apple-Mail-9-453229316 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Indeed - I guess no-one's using the 2-class-only option! (It was only really needed for crappy old data with very poor grey-white contrast.) This is a bug which we'll fix in the next release - corrected script attached. Cheers. --Apple-Mail-9-453229316 Content-Transfer-Encoding: 7bit Content-Type: application/octet-stream; x-unix-mode=0755; name=sienax Content-Disposition: attachment; filename=sienax #!/bin/sh # sienax - Structural Image Evaluation, including Normalisation, of Atrophy (X-sectional) # # Stephen Smith, FMRIB Image Analysis Group # # Copyright (C) 1999-2006 University of Oxford # # SHCOPYRIGHT Usage() { echo "" echo "Usage: sienax [-d] [-B \"bet options\"] [-2] [-t2] [-t ] [-b ] [-r] [-lm ] [-S \"segmentation options\"]" echo "" echo "-d : debug (don't delete intermediate files)" echo "-B \"betopts\" : options to pass to BET brain extraction (inside double-quotes), e.g. -B \"-f 0.3\"" echo "-2 : two-class segmentation (don't segment grey and white matter separately)" echo "-t2 : T2-weighted input image (default T1-weighted)" echo "-t : ignore from t (mm) upwards in MNI152/Talairach space" echo "-b : ignore from b (mm) downwards in MNI152/Talairach space. b should probably be -ve" echo "-r : regional - use standard-space masks to give peripheral cortex GM volume (3-class segmentation only) and ventricular CSF volume" echo "-lm : use lesion (or lesion+CSF) mask to remove incorrectly labelled \"grey matter\" voxels" echo "-S \"segopts\" : options to pass to FAST segmentation (inside double-quotes), e.g. -S \"-i 20\"" echo "" exit } if [ _$FSLDIR = _ ] ; then FSLDIR=/usr/local/fsl export FSLDIR fi [ _$1 = _ ] && Usage I=`${FSLDIR}/bin/remove_ext $1`; [ `${FSLDIR}/bin/imtest $I` = 0 ] && Usage echo "-----------------------------------------------------------------------" > ${I}.sienax echo "" >> ${I}.sienax echo " SIENA - Structural Image Evaluation, using Normalisation, of Atrophy" >> ${I}.sienax echo " part of FSL www.fmrib.ox.ac.uk/fsl" >> ${I}.sienax echo " running cross-sectional atrophy measurement: sienax version 2.4" >> ${I}.sienax echo " sienax $@" >> ${I}.sienax echo "" >> ${I}.sienax shift debug=0 regional=0 betopts="" segopts="" nseg=3 stdroi="" origin3=37 # `avwval ${FSLDIR}/etc/standard/avg152T1 origin3` pixdim3=2 # `avwval ${FSLDIR}/etc/standard/avg152T1 pixdim3` imtype=-t1 while [ _$1 != _ ] ; do if [ $1 = -d ] ; then debug=1 shift elif [ $1 = -r ] ; then regional=1 shift elif [ $1 = -B ] ; then betopts=$2 shift 2 elif [ $1 = -S ] ; then segopts=$2 shift 2 elif [ $1 = -2 ] ; then nseg=2 shift elif [ $1 = -t2 ] ; then imtype=-t2 shift elif [ $1 = -t ] ; then stdt=`echo $2 | sed 's/-/_/g'` stdt=`echo "10 k $stdt $pixdim3 / $origin3 + p" | dc -` stdroi="$stdroi -roi 0 1000000 0 1000000 0 $stdt 0 1" shift 2 elif [ $1 = -b ] ; then stdb=`echo $2 | sed 's/-/_/g'` stdb=`echo "10 k $stdb $pixdim3 / $origin3 + p" | dc -` stdroi="$stdroi -roi 0 1000000 0 1000000 $stdb 1000000 0 1" shift 2 elif [ $1 = -lm ] ; then lm=$2 shift 2 else Usage fi done if [ $regional = 1 ] ; then if [ $nseg = 2 ] ; then echo "Can't do regional analysis with 2-class segmentation" exit fi fi echo "---------- extract brain --------------------------------------------" >> ${I}.sienax ${FSLDIR}/bin/bet $I ${I}_brain -s $betopts >> ${I}.sienax echo "" >> ${I}.sienax echo "---------- register to stdairach space using brain and skull --------" >> ${I}.sienax echo "(do not worry about histogram warnings)" >> ${I}.sienax ${FSLDIR}/bin/pairreg ${FSLDIR}/etc/standard/avg152T1_brain ${I}_brain ${FSLDIR}/etc/standard/avg152T1_skull ${I}_brain_skull ${I}2std.mat >> ${I}.sienax 2>&1 ${FSLDIR}/bin/avscale ${I}2std.mat ${FSLDIR}/etc/standard/avg152T1 > ${I}2std.avscale xscale=`grep Scales ${I}2std.avscale | awk '{print $4}'` yscale=`grep Scales ${I}2std.avscale | awk '{print $5}'` zscale=`grep Scales ${I}2std.avscale | awk '{print $6}'` vscale=`echo "10 k $xscale $yscale * $zscale * p"|dc -` echo "VSCALING $vscale" >> ${I}.sienax echo "" >> ${I}.sienax echo "---------- mask with std mask ---------------------------------------" >> ${I}.sienax ${FSLDIR}/bin/convert_xfm -inverse -omat ${I}2std_inv.mat ${I}2std.mat MASK=${FSLDIR}/etc/standard/avg152T1_brain_mask_dil if [ "$stdroi" != "" ] ; then ${FSLDIR}/bin/avwmaths $MASK $stdroi ${I}_stdmaskroi MASK=${I}_stdmaskroi fi ${FSLDIR}/bin/flirt -in $MASK -ref ${I}_brain -out ${I}_stdmask -applyxfm -init ${I}2std_inv.mat ${FSLDIR}/bin/avwmaths ${I}_stdmask -thr 0.5 -bin ${I}_stdmask ${FSLDIR}/bin/avwmaths ${I}_brain -mask ${I}_stdmask ${I}_stdmaskbrain if [ $regional = 1 ] ; then ${FSLDIR}/bin/flirt -in ${FSLDIR}/etc/standard/avg152T1_strucseg_periph -ref ${I}_brain -out ${I}_stdmask_segperiph -applyxfm -init ${I}2std_inv.mat ${FSLDIR}/bin/avwmaths ${I}_stdmask_segperiph -thr 0.5 -bin ${I}_stdmask_segperiph ${FSLDIR}/bin/avwmaths ${FSLDIR}/etc/standard/avg152T1_strucseg -thr 4.5 -bin ${I}_tmpmask ${FSLDIR}/bin/flirt -in ${I}_tmpmask -ref ${I}_brain -out ${I}_stdmask_segvent -applyxfm -init ${I}2std_inv.mat ${FSLDIR}/bin/avwmaths ${I}_stdmask_segvent -thr 0.5 -bin ${I}_stdmask_segvent /bin/rm ${I}_tmpmask* fi echo "" >> ${I}.sienax echo "---------- segment tissue into types --------------------------------" >> ${I}.sienax if [ $nseg = 2 ] ; then ${FSLDIR}/bin/fast -c 2 $imtype -e -ov $segopts ${I}_stdmaskbrain >> ${I}.sienax 2>&1 echo "" >> ${I}.sienax echo "---------- convert brain volume into normalised volume --------------" >> ${I}.sienax S=`${FSLDIR}/bin/avwstats ${I}_stdmaskbrain_pve_1 -m -v` xa=`echo $S | awk '{print $1}'` xb=`echo $S | awk '{print $3}'` ubrain=`echo "2 k $xa $xb * 1 / p" | dc -` nbrain=`echo "2 k $xa $xb * $vscale * 1 / p" | dc -` else if [ _$lm != _ ] ; then ${FSLDIR}/bin/avwmaths_32R $lm -bin -mul -1 -add 1 -mul ${I}_stdmaskbrain ${I}_stdmaskbrain fi ${FSLDIR}/bin/fast $imtype -e -ov $segopts ${I}_stdmaskbrain >> ${I}.sienax 2>&1 if [ _$lm != _ ] ; then ${FSLDIR}/bin/avwmaths_32R $lm -bin -max ${I}_stdmaskbrain_pve_2 ${I}_stdmaskbrain_pve_2 ${FSLDIR}/bin/avwmaths_32SI $lm -bin -mul 3 -max ${I}_stdmaskbrain_seg ${I}_stdmaskbrain_seg fi echo "" >> ${I}.sienax echo "---------- convert brain volume into normalised volume --------------" >> ${I}.sienax echo "" >> ${I}.sienax echo "tissue volume unnormalised-volume" >> ${I}.sienax if [ $regional = 1 ] ; then ${FSLDIR}/bin/avwmaths_32R ${I}_stdmaskbrain_pve_1 -mas ${I}_stdmask_segperiph ${I}_stdmaskbrain_pve_1_segperiph S=`${FSLDIR}/bin/avwstats ${I}_stdmaskbrain_pve_1_segperiph -m -v` xa=`echo $S | awk '{print $1}'` xb=`echo $S | awk '{print $3}'` uxg=`echo "2 k $xa $xb * 1 / p" | dc -` xg=`echo "2 k $xa $xb * $vscale * 1 / p" | dc -` echo "pgrey $xg $uxg (peripheral grey)" >> ${I}.sienax ${FSLDIR}/bin/avwmaths_32R ${I}_stdmaskbrain_pve_0 -mas ${I}_stdmask_segvent ${I}_stdmaskbrain_pve_0_segvent S=`${FSLDIR}/bin/avwstats ${I}_stdmaskbrain_pve_0_segvent -m -v` xa=`echo $S | awk '{print $1}'` xb=`echo $S | awk '{print $3}'` uxg=`echo "2 k $xa $xb * 1 / p" | dc -` xg=`echo "2 k $xa $xb * $vscale * 1 / p" | dc -` echo "vcsf $xg $uxg (ventricular CSF)" >> ${I}.sienax fi S=`${FSLDIR}/bin/avwstats ${I}_stdmaskbrain_pve_1 -m -v` xa=`echo $S | awk '{print $1}'` xb=`echo $S | awk '{print $3}'` ugrey=`echo "2 k $xa $xb * 1 / p" | dc -` ngrey=`echo "2 k $xa $xb * $vscale * 1 / p" | dc -` echo "GREY $ngrey $ugrey" >> ${I}.sienax S=`${FSLDIR}/bin/avwstats ${I}_stdmaskbrain_pve_2 -m -v` xa=`echo $S | awk '{print $1}'` xb=`echo $S | awk '{print $3}'` uwhite=`echo "2 k $xa $xb * 1 / p" | dc -` nwhite=`echo "2 k $xa $xb * $vscale * 1 / p" | dc -` echo "WHITE $nwhite $uwhite" >> ${I}.sienax ubrain=`echo "2 k $uwhite $ugrey + 1 / p" | dc -` nbrain=`echo "2 k $nwhite $ngrey + 1 / p" | dc -` fi echo "BRAIN $nbrain $ubrain" >> ${I}.sienax ${FSLDIR}/bin/overlay 1 1 -c ${I} -a ${I}_stdmaskbrain_seg 1.9 5 ${I}_render if [ $regional = 1 ] ; then ${FSLDIR}/bin/overlay 1 1 -c ${I} -a ${I}_stdmaskbrain_pve_1_segperiph 0.3 0.7 ${I}_periph_render ${FSLDIR}/bin/overlay 1 1 -c ${I} -a ${I}_stdmaskbrain_pve_0_segvent 0.3 0.7 ${I}_vent_render fi if [ $debug = 0 ] ; then /bin/rm -f `$FSLDIR/bin/imglob ${I}_brain* ${I}_stdmask*` /bin/rm -f ${I}2std.avscale ${I}2std_inv.mat fi echo "$nbrain" --Apple-Mail-9-453229316 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed On 30 May 2007, at 23:26, Zografos Caramanos wrote: > Hello! > > I think that I may have found a bug in SIENAX v-2.4 (FSL v-3.3.11). > > When I run a 3-mm, axial, T1-weighted scan with a 3-tissue > segmentation, > SIENAX v-2.4 gives you normalized GREY, WHITE, and BRAIN volumes: > > (SIENAX v-2.4, 3-tissue segmentation) > > ====================================================================== > === > > Class: CSF tissue 1 tissue 2 brain % > Volumes: 205237.1 566241.4 665635.7 0.857188 > > ---------- convert brain volume into normalised volume > -------------- > > tissue volume unnormalised-volume > GREY 760365.64 566240.07 > WHITE 893834.69 665633.72 > BRAIN 1654200.33 1231873.79 > > ====================================================================== > === > (...excerpted from below) > > > BUT, if I run the same scan with the -2 flag for a 2-tissue > segmentation, > SIENAX v-2.4 does not seem to properly calculate the normalized > BRAIN volume: > > (SIENAX v-2.4, 2-tissue segmentation) > > ====================================================================== > === > Class: CSF tissue 1 brain percentage > Volumes: 205237.1 1231946.5 0.857195 > > ---------- convert brain volume into normalised volume > -------------- > BRAIN 1 1 > > ====================================================================== > === > (...excerpted from further below) > > > > Please note that this is not the case when I run the same scan using > SIENAX v-2.3 (FSL v-3.2beta, FAST v-3.5), which gives slightly > different > values: > > (SIENAX v-2.3, 3-tissue segmentation) > > ====================================================================== > === > Class: CSF tissue 1 tissue 2 brain % > Volumes: 195260.0 562279.6 662799.2 0.862526 > > ---------- convert brain volume into normalised volume > -------------- > > GREY 757578.8396899881 > WHITE 893006.5301093496 > > BRAIN 1650585.3697993377 > > ====================================================================== > === > (...excerpted from further below) > > > (SIENAX v-2.3, 2-tissue segmentation) > > ====================================================================== > === > Class: CSF tissue 1 brain percentage > Volumes: 194299.3 1226099.5 0.863208 > > ---------- convert brain volume into normalised volume > -------------- > > BRAIN 1651960.9049744302 > > ====================================================================== > === > (...excerpted from further below) > > > Please let me know if I am doing something wrong, or if there is > some fix > to this problem. > > Cheers! > Aki > > Zografos Caramanos, M.A. > Magnetic Resonance Studies Unit > Montreal Neurological Institute, McGill University > (e-mail) [log in to unmask] > > ====================================================================== > = > > > > +++++++++++++++++++++++++++++++++++++++++++++++ > SIENAX Version 2.4 with 3-tissue segmentation > +++++++++++++++++++++++++++++++++++++++++++++++ > > ---------------------------------------------------------------------- > - > > SIENA - Structural Image Evaluation, using Normalisation, of > Atrophy > part of FSL www.fmrib.ox.ac.uk/fsl > running cross-sectional atrophy measurement: sienax version 2.4 > sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d > > ---------- extract brain > -------------------------------------------- > > ---------- register to stdairach space using brain and skull > -------- > (do not worry about histogram warnings) > VSCALING 1.3428326384 > > ---------- mask with std mask > --------------------------------------- > > ---------- segment tissue into types > -------------------------------- > > FAST - FMRIB's Automated Segmentation Tool Version 3.53 > > Image: /tmp/ > fsl_z5FSyo_mc_carz6809.2007-02-08.Scan_nt1p_none_stdmaskbrain > > T1-weighted image > Imagesize : 192 x 256 x 60 > Pixelsize : 0.976562 x 0.976562 x 3 > > Initial K-means segmentation... > 8 main iterations ... > Estimating PVE... > > Segmentation done successfully! > > Calculation time 234 seconds! > > Write segmentation image /tmp/fsl_z5FSyo.fastout/QQQQQ_seg.hdr > Successfully! > Write PVE image of tissue 0 Successfully! > Write PVE image of tissue 1 Successfully! > Write PVE image of tissue 2 Successfully! > > Class: CSF tissue 1 tissue 2 brain percentage > Volumes: 205237.1 566241.4 665635.7 0.857188 > > ---------- convert brain volume into normalised volume > -------------- > > tissue volume unnormalised-volume > GREY 760365.64 566240.07 > WHITE 893834.69 665633.72 > BRAIN 1654200.33 1231873.79 > > > > > +++++++++++++++++++++++++++++++++++++++++++++++ > SIENAX Version 2.4 with 2-tissue segmentation > +++++++++++++++++++++++++++++++++++++++++++++++ > > ---------------------------------------------------------------------- > - > > SIENA - Structural Image Evaluation, using Normalisation, of > Atrophy > part of FSL www.fmrib.ox.ac.uk/fsl > running cross-sectional atrophy measurement: sienax version 2.4 > sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d -2 > > ---------- extract brain > -------------------------------------------- > > ---------- register to stdairach space using brain and skull > -------- > (do not worry about histogram warnings) > VSCALING 1.3428326384 > > ---------- mask with std mask > --------------------------------------- > > ---------- segment tissue into types > -------------------------------- > > FAST - FMRIB's Automated Segmentation Tool Version 3.53 > > Image: /tmp/ > fsl_69X29e_mc_carz6809.2007-02-08.Scan_nt1p_none_stdmaskbrain > > T1-weighted image > Imagesize : 192 x 256 x 60 > Pixelsize : 0.976562 x 0.976562 x 3 > > Initial K-means segmentation... > 8 main iterations ... > Estimating PVE... > > Segmentation done successfully! > > Calculation time 234 seconds! > > Write segmentation image /tmp/fsl_69X29e.fastout/QQQQQ_seg.hdr > Successfully! > Write PVE image of tissue 0 Successfully! > Write PVE image of tissue 1 Successfully! > > Class: CSF tissue 1 brain percentage > Volumes: 205237.1 1231946.5 0.857195 > > ---------- convert brain volume into normalised volume > -------------- > BRAIN 1 1 > > > > +++++++++++++++++++++++++++++++++++++++++++++++ > SIENAX Version 2.3 with 3-tissue segmentation > +++++++++++++++++++++++++++++++++++++++++++++++ > > ---------------------------------------------------------------------- > - > > SIENA - Structural Image Evaluation, using Normalisation, of > Atrophy > part of FSL www.fmrib.ox.ac.uk/fsl > running cross-sectional atrophy measurement: sienax version 2.3 > sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d > > ---------- extract brain > -------------------------------------------- > > ---------- register to talairach space using brain and skull > -------- > (do not worry about histogram warnings) > WARNING: Both reference and input images have an sform matrix set > The output image will use the sform from the reference image > WARNING: Both reference and input images have an sform matrix set > The output image will use the sform from the reference image > WARNING: Both reference and input images have an sform matrix set > The output image will use the sform from the reference image > VSCALING 1.3473268622 > > ---------- mask with talairach mask > --------------------------------- > > ---------- segment tissue into types > -------------------------------- > > FAST - FMRIB's Automated Segmentation Tool Version 3.5 > > Image: /tmp/ > fsl_mEUvrB_mc_carz6809.2007-02-08.Scan_nt1p_none_talmaskbrain > > T1-weighted image > Imagesize : 192 x 256 x 60 > Pixelsize : 0.976562 x 0.976562 x 3 > > Initial K-means segmentation... > 8 main iterations ... > Estimating PVE... > > Segmentation done successfully! > > Calculation time 234 seconds! > > Write segmentation image /tmp/fsl_mEUvrB.fastout/QQQQQ_seg.hdr > Successfully! > Write PVE image of tissue 0 Successfully! > Write PVE image of tissue 1 Successfully! > Write PVE image of tissue 2 Successfully! > > Class: CSF tissue 1 tissue 2 brain percentage > Volumes: 195260.0 562279.6 662799.2 0.862526 > > ---------- convert brain volume into normalised volume > -------------- > > GREY 757578.8396899881 > WHITE 893006.5301093496 > > BRAIN 1650585.3697993377 > > > +++++++++++++++++++++++++++++++++++++++++++++++ > SIENAX Version 2.3 with 2-tissue segmentation > +++++++++++++++++++++++++++++++++++++++++++++++ > > ---------------------------------------------------------------------- > - > > SIENA - Structural Image Evaluation, using Normalisation, of > Atrophy > part of FSL www.fmrib.ox.ac.uk/fsl > running cross-sectional atrophy measurement: sienax version 2.3 > sienax mc_carz6809.2007-02-08.Scan_nt1p_none -d -2 > > ---------- extract brain > -------------------------------------------- > > ---------- register to talairach space using brain and skull > -------- > (do not worry about histogram warnings) > WARNING: Both reference and input images have an sform matrix set > The output image will use the sform from the reference image > WARNING: Both reference and input images have an sform matrix set > The output image will use the sform from the reference image > WARNING: Both reference and input images have an sform matrix set > The output image will use the sform from the reference image > VSCALING 1.3473268622 > > ---------- mask with talairach mask > --------------------------------- > > ---------- segment tissue into types > -------------------------------- > > FAST - FMRIB's Automated Segmentation Tool Version 3.5 > > Image: /tmp/ > fsl_AlcxrX_mc_carz6809.2007-02-08.Scan_nt1p_none_talmaskbrain > > T1-weighted image > Imagesize : 192 x 256 x 60 > Pixelsize : 0.976562 x 0.976562 x 3 > > Initial K-means segmentation... > 4 main iterations ... > Estimating PVE... > > Segmentation done successfully! > > Calculation time 89 seconds! > > Write segmentation image /tmp/fsl_AlcxrX.fastout/QQQQQ_seg.hdr > Successfully! > Write PVE image of tissue 0 Successfully! > Write PVE image of tissue 1 Successfully! > > Class: CSF tissue 1 brain percentage > Volumes: 194299.3 1226099.5 0.863208 > > ---------- convert brain volume into normalised volume > -------------- > > BRAIN 1651960.9049744302 ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ --- --Apple-Mail-9-453229316-- ========================================================================= Date: Thu, 31 May 2007 13:45:28 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Dianne Patterson <[log in to unmask]> Subject: AAL library of anatomical rois MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_13912_27121722.1180644328603" ------=_Part_13912_27121722.1180644328603 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Dear Group, Any recommendations? Has anyone successfully used a set of rois (I am familiar with the AAL library) to display in fslview as overlays? I gather I would need to reslice them to display them as a layer on one of my anatomical images...but right now, I can't even view one by itself (see below) ============================================================= I am also having trouble displaying the simple binary mask file (img/hdr) in fslview. -I'm using a mask of the right frontal lobe...so it is nice and big...not just lost in the image somewhere. All I see is black. -I used avwstats++ -r -R to get the min max voxel intensities and it reports 0 and 255. -I used avwstats++ -x to locate a max intensity voxel...but when I went to that location I still only saw black, (although fslview reports the voxel as having the value 255, it displays only the background black. -Shifting the transparency slider makes no difference. -If I display the image in ImageJ, I see the white mask of the R_frontal lobe as well as the black background. -SPM5 also displays the image with no problem. -FSLview displays other *nii and *nii.gz files with no issue, both the binary masks bet2 created and all of my other MR images. The little eye is checked on the lower right of fslview indicating that the image should be viewable. toggling it makes no difference. -I presume the images are the old style img/hdr file...but I thought nifti viewers, and especially fslview, could handle that... so I'm a little perplexed... Any thoughts on the source of my troubles would be greatly appreciated. Thanks very much, -Dianne -- Dianne Patterson, Ph.D. [log in to unmask] ERP Lab University of Arizona 621-3256 (Office) ------=_Part_13912_27121722.1180644328603 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Dear Group,
Any recommendations?
Has anyone successfully used a set of rois (I am familiar with the AAL library) to display in fslview as overlays?
I gather I would need to reslice them to display them as a layer on one of my anatomical images...but right now, I can't even view one by itself (see below)
=============================================================
I am also having trouble displaying the simple binary mask file (img/hdr) in fslview.
-I'm using a mask of the right frontal lobe...so it is nice and big...not just lost in the image somewhere.
All I see is black. 
-I used avwstats++ -r -R to get the min max voxel intensities and it reports 0 and 255.
-I used avwstats++ -x to locate a max intensity voxel...but when I went to that location I still only saw black, (although fslview reports the voxel as having the value 255, it displays only the background black.
-Shifting the transparency slider makes no difference.
-If I display the image in ImageJ, I see the white mask of the R_frontal lobe as well as the black background. 
-SPM5 also displays the image with no problem.
-FSLview displays other *nii and *nii.gz files with no issue, both the binary masks bet2 created and all of my other MR images.
The little eye is checked on the lower right of fslview indicating that the image should be viewable.  toggling it makes no difference.
-I presume the images are the old style img/hdr file...but I thought nifti viewers, and especially fslview, could handle that...
so I'm a little perplexed...

Any thoughts on the source of my troubles would be greatly appreciated.

Thanks very much,

-Dianne
--
Dianne Patterson, Ph.D.
[log in to unmask]
ERP Lab
University of Arizona
621-3256 (Office) ------=_Part_13912_27121722.1180644328603-- ========================================================================= Date: Thu, 31 May 2007 16:10:43 -0500 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Peter Fried <[log in to unmask]> Subject: Re: AAL library of anatomical rois In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi Diane, It sounds like you might just need to change your window/leveling. Are the Min and Max values near the top of the screen (in fslview) at 0 and 0? If so change the Max value to 255 (the max value from avwstats++ -R). Hope this helps. -Pete > Dear Group, > Any recommendations? > Has anyone successfully used a set of rois (I am familiar with the AAL > library) to display in fslview as overlays? > I gather I would need to reslice them to display them as a layer on one of > my anatomical images...but right now, I can't even view one by itself (see > below) > ============================================================= > I am also having trouble displaying the simple binary mask file (img/hdr) > in > fslview. > -I'm using a mask of the right frontal lobe...so it is nice and big...not > just lost in the image somewhere. > All I see is black. > -I used avwstats++ -r -R to get the min max voxel intensities and it > reports > 0 and 255. > -I used avwstats++ -x to locate a max intensity voxel...but when I went to > that location I still only saw black, (although fslview reports the voxel > as > having the value 255, it displays only the background black. > -Shifting the transparency slider makes no difference. > -If I display the image in ImageJ, I see the white mask of the R_frontal > lobe as well as the black background. > -SPM5 also displays the image with no problem. > -FSLview displays other *nii and *nii.gz files with no issue, both the > binary masks bet2 created and all of my other MR images. > The little eye is checked on the lower right of fslview indicating that > the > image should be viewable. toggling it makes no difference. > -I presume the images are the old style img/hdr file...but I thought nifti > viewers, and especially fslview, could handle that... > so I'm a little perplexed... > > Any thoughts on the source of my troubles would be greatly appreciated. > > Thanks very much, > > -Dianne > -- > Dianne Patterson, Ph.D. > [log in to unmask] > ERP Lab > University of Arizona > 621-3256 (Office) > ========================================================================= Date: Thu, 31 May 2007 15:39:02 -0700 Reply-To: FSL - FMRIB's Software Library <[log in to unmask]> Sender: FSL - FMRIB's Software Library <[log in to unmask]> From: Dianne Patterson <[log in to unmask]> Subject: Re: AAL library of anatomical rois In-Reply-To: <[log in to unmask]> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_14599_26616881.1180651142238" ------=_Part_14599_26616881.1180651142238 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Wow...thankyou so much...that did the trick...I hadn't seen those tiny little boxes. -Dianne On 5/31/07, Peter Fried <[log in to unmask]> wrote: > > Hi Diane, > > It sounds like you might just need to change your window/leveling. Are the > Min and Max values near the top of the screen (in fslview) at 0 and 0? If > so change the Max value to 255 (the max value from avwstats++ -R). > > Hope this helps. > > -Pete > > > Dear Group, > > Any recommendations? > > Has anyone successfully used a set of rois (I am familiar with the AAL > > library) to display in fslview as overlays? > > I gather I would need to reslice them to display them as a layer on one > of > > my anatomical images...but right now, I can't even view one by itself > (see > > below) > > ============================================================= > > I am also having trouble displaying the simple binary mask file > (img/hdr) > > in > > fslview. > > -I'm using a mask of the right frontal lobe...so it is nice and > big...not > > just lost in the image somewhere. > > All I see is black. > > -I used avwstats++ -r -R to get the min max voxel intensities and it > > reports > > 0 and 255. > > -I used avwstats++ -x to locate a max intensity voxel...but when I went > to > > that location I still only saw black, (although fslview reports the > voxel > > as > > having the value 255, it displays only the background black. > > -Shifting the transparency slider makes no difference. > > -If I display the image in ImageJ, I see the white mask of the R_frontal > > lobe as well as the black background. > > -SPM5 also displays the image with no problem. > > -FSLview displays other *nii and *nii.gz files with no issue, both the > > binary masks bet2 created and all of my other MR images. > > The little eye is checked on the lower right of fslview indicating that > > the > > image should be viewable. toggling it makes no difference. > > -I presume the images are the old style img/hdr file...but I thought > nifti > > viewers, and especially fslview, could handle that... > > so I'm a little perplexed... > > > > Any thoughts on the source of my troubles would be greatly appreciated. > > > > Thanks very much, > > > > -Dianne > > -- > > Dianne Patterson, Ph.D. > > [log in to unmask] > > ERP Lab > > University of Arizona > > 621-3256 (Office) > > > -- Dianne Patterson, Ph.D. [log in to unmask] ERP Lab University of Arizona 621-3256 (Office) ------=_Part_14599_26616881.1180651142238 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Wow...thankyou so much...that did the trick...I hadn't seen those tiny little boxes.

-Dianne

On 5/31/07, Peter Fried < [log in to unmask]> wrote:
Hi Diane,

It sounds like you might just need to change your window/leveling. Are the
Min and Max values near the top of the screen (in fslview) at 0 and 0? If
so change the Max value to 255 (the max value from avwstats++ -R).

Hope this helps.

-Pete

> Dear Group,
> Any recommendations?
> Has anyone successfully used a set of rois (I am familiar with the AAL
> library) to display in fslview as overlays?
> I gather I would need to reslice them to display them as a layer on one of
> my anatomical images...but right now, I can't even view one by itself (see
> below)
> =============================================================
> I am also having trouble displaying the simple binary mask file (img/hdr)
> in
> fslview.
> -I'm using a mask of the right frontal lobe...so it is nice and big...not
> just lost in the image somewhere.
> All I see is black.
> -I used avwstats++ -r -R to get the min max voxel intensities and it
> reports
> 0 and 255.
> -I used avwstats++ -x to locate a max intensity voxel...but when I went to
> that location I still only saw black, (although fslview reports the voxel
> as
> having the value 255, it displays only the background black.
> -Shifting the transparency slider makes no difference.
> -If I display the image in ImageJ, I see the white mask of the R_frontal
> lobe as well as the black background.
> -SPM5 also displays the image with no problem.
> -FSLview displays other *nii and * nii.gz files with no issue, both the
> binary masks bet2 created and all of my other MR images.
> The little eye is checked on the lower right of fslview indicating that
> the
> image should be viewable.  toggling it makes no difference.
> -I presume the images are the old style img/hdr file...but I thought nifti
> viewers, and especially fslview, could handle that...
> so I'm a little perplexed...
>
> Any thoughts on the source of my troubles would be greatly appreciated.
>
> Thanks very much,
>
> -Dianne
> --
> Dianne Patterson, Ph.D.
> [log in to unmask]
> ERP Lab
> University of Arizona
> 621-3256 (Office)
>



--
Dianne Patterson, Ph.D.
[log in to unmask]
ERP Lab
University of Arizona
621-3256 (Office) ------=_Part_14599_26616881.1180651142238--