Hi Amelia,
there is a problem with your design.con (see the
attached one instead).
The first contrast is giving you where the FA is
higher for the second group than for the first one
(B>A), and the second contrast is telling you where
FA is smaller for the second group (A>B).
Then, assuming that you've got results only in one
contrast, so let's say smaller FA values in A (so in
FA_tstat1), you should see if it's explained by lower
L1 in A (so in L1_tstat1) and/or higher RD (so in
RD_tstat2). Both contribution can be non significant
but contribute to a significant effect on FA values.
Hope this helps,
Gwenaelle
--- "Versace, Amelia" <[log in to unmask]> a écrit :
> Thank you for the prompt reply!
>
>
>
> I guess! Let me spell out what I did.
>
> I ran randomise with FA data (randomise -i
> all_FA_skeleton -o tbss_FA -m mean_FA_skeleton_mask
> -d design.mat -t design.com -c 3 -n 10000 -v 5)
> using the attached design.mat and design.com.
>
>
>
> Then, I used the same design.com and design.mat to
> run L1 and RD data:
>
> randomise -i all_L1_skeleton -o tbss_L1 -m
> mean_FA_skeleton_mask -d design.mat -t design.com -c
> 3 -n 10000 -v 5;
>
> randomise -i all_RD_skeleton -o tbss_RD -m
> mean_FA_skeleton_mask -d design.mat -t design.com -c
> 3 -n 10000 -v 5.
>
> At the end, I overlapped in fslview FA_tstat1/tstat2
> maps with the corresponding L1 and/or RD maps,
> focusing on the highly significant FA coordinates.
>
>
>
>
>
> Is this what you mean?
>
> Thanks a lot for your help!!
>
> Amelia
>
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: FSL - FMRIB's Software Library
> [mailto:[log in to unmask]] On Behalf Of Steve Smith
> Sent: Thursday, November 22, 2007 1:54 AM
> To: [log in to unmask]
> Subject: Re: [FSL] FA, axial and perpendicular
> diffusivity
>
>
>
> Hi - you found a change in FA and are looking to see
> whether that's
>
> caused more by change in L1 or (L2+L3)/2; did you
> test those for
>
> changes in both directions (ie using +1 and -1
> contrasts separately)?
>
>
>
> Cheers.
>
>
>
>
>
> On 21 Nov 2007, at 20:37, Versace, Amelia wrote:
>
>
>
> > Hi FDT experts,
>
> >
>
> > We did two paired sample t test, comparing group A
> vs group B (using
>
> > the statistic tool, TBSS).
>
> > We defined first and merged then the resulting DTI
> vector tstat_maps
>
> > (FA, axial=L1 and perpendicular
> diffusivity=(L2+L3)/2) in order to
>
> > define which primary components (L1 or L2+L3/2)
> mainly determined
>
> > the FA differences.
>
> > For some region, we got some rois with significant
> FA deference
>
> > between A and B whit a very low L1tstat1 value
> (es. 0.2) and also
>
> > very low (L2+L3)/2 tstat2 value (es. 0.3). Doesn't
> this mean that
>
> > the anisotropy is not mainly due to L1 as well
> (L2+L3)/2??? So,
>
> > where the FA highly significant difference come
> from??
>
> >
>
> > Could anybody explain why this is happening ??? Am
> I doing something
>
> > wrong? Any help would be really appreciated!!
>
> >
>
> > Thank you for your time!!
>
> > Amelia
>
> >
>
> >
>
> >
>
> > Amelia Versace, MD
>
> > Loeffler Building
>
> > 121 Meyran Avenue, room 202
>
> > Pittsburgh, PA 15213
>
> > Phone 412- 383-8202
>
> > Fax 412-383-8336
>
>
>
>
>
>
---------------------------------------------------------------------------
>
> 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
>
>
---------------------------------------------------------------------------
> > /NumWaves 2
> /NumPoints 56
> /PPheights 1 1
> /Matrix
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 1 0
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> 0 1
> > /NumWaves 2
> /NumContrasts 2
> /PPheights 1 1
> /Matrix 1 -1 -1 1
_____________________________________________________________________________
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail
|