Hi Stanley - you just need to do the following to feed your own
standard-space images into high-level FEAT:
make a something.feat directory and make a stats subdirectory in there.
In stats put all your standard-space cope and varcope files numbered
cope1, varcope1, cope2, varcope2 etc.
in the FEAT directory run
touch design.lev
(this tells FEAT it's a higher-level directory)
avwmerge -t mask stats/varcope*hdr
avwmaths mask -Tmin -bin mask
(this creates a sensible binary brain mask)
copy into the FEAT directory the standard-space image you want used as the
background in colour overlays, naming it example_func
Now you should be able to enter the copes into the FEAT GUI and all should
be well :)
ttfn, Steve.
On Fri, 11 Oct 2002, Stanley Colcombe wrote:
> Hi Guys. I probably should have been a lot more explicit in my initial
> e-mail. What I'd like to do is analyze copes that were produced with FEAT,
> but were not registered in the standard path (why I did that is a long and
> irrelevant story), so the directory structure, etc. is not the same as in a
> standard analysis. Ideally, I'd like to have FLAME just believe me when I
> tell it that these images are already in common space, and run the analyses
> on whatever maps I give it. Is there a reasonably simple hack to take the
> safties off (e.g. using a common ANDed mask)? I haven't had a look at the
> code yet.
> Many thanks in advance,
> Stan.
>
>
> -----Original Message-----
> From: Stephen Smith [mailto:[log in to unmask]]
> Sent: Friday, October 11, 2002 4:46 AM
> To: [log in to unmask]
> Subject: Re: FLAME and COPEs
>
> Hi Christian's right when it comes to feeding in data from first-level
> analyses. However, if you are feeding in standard-space copes (eg from a
> second-level anlaysis) into FEAT then what group-FEAT does is:
>
> 1) assumes that the copes are in the "stats" subdirectory in a feat
> directory. this way it knows to look in the feat directory itself for
> things like the mask image etc.
>
> 2) it thinks that the data is already in standard space because it was
> a second-level analysis if the "design.lev" file exists in the FEAT
> directory - so you might want to create this if you have generated this
> data some other way.
>
> 3) now, when you did a second level analysis, the standard space versions
> of the mask image and the stats images get put in a reg_standard
> directory. to keep things easy for higher-than-second-level anlaysis, the
> second-level feat directory has a link in the feat called reg_standard
> pointing simply straight back into the feat directory, so that when it
> looks for standard space stuff "in reg_standard" it finds all the files
> (which are already in standard space). moral of the story is - you might
> need to create that link, ie in the "second-level" feat directory, do
> ln -s . reg_standard
>
>
> sorry this sounds copmlicated - but the whole thing is designed so that
> for users doing standard analyses these kind of details are taken care of
> automatically.
>
> maybe we should prepare a more detailed clear version of the above to add
> to the manual.
>
> please email again if this doesn't cover what you are trying to do!
>
> Thanks, Steve
>
>
> On Thu, 10 Oct 2002, Stanley Colcombe wrote:
>
> > Hi All-
> > This is probably a simple oversight on my end. I'm trying to use FLAME
> to
> > analyze some existing cope images that are already registered to MNI
> space.
> > However, after setting up the parameters in the model, FLAME gives me the
> > following message:
> > "Registration has not been run for all of the FEAT directories that you
> have
> > selected for group analysis. Please turn on and setup registration."
> > I do have "Inputs are cope images" selected in the "Data" tab. What am
> I
> > missing?
> > Thanks in advance.
> > Stan.
> >
>
> Stephen M. Smith
> Head of Image Analysis, FMRIB
>
> Oxford University Centre for Functional MRI of the Brain
> John Radcliffe 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
Head of Image Analysis, FMRIB
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
|