Hi Venkateswaran,
what I suspect is that some of your subjects have white matter hypointensities that get misclassified as grey matter... If I'm right, you could maybe change the threshold for the creation of the mask, by default it is:
$FSLDIR/bin/fslmaths GM_merg -Tmean -thr 0.01 -bin GM_mask -odt char
before re-running randomise.
Also, is it an increase of GM that you see instead of a decrease? If so, you can just not report it if you have *checked carefully* that this is indeed not more GM but rather more WM hypointensities in one group than the other.
Hope this helps,
Gwenaelle
--- En date de : Mer 27.10.10, Rajagopalan, Venkateswaran <[log in to unmask]> a écrit :
> De: Rajagopalan, Venkateswaran <[log in to unmask]>
> Objet: Re: [FSL] VBM results in white matter
> À: [log in to unmask]
> Date: Mercredi 27 octobre 2010, 15h59
> Thanks Steve so in that case do I
> need to interfere manually to correct for any registration
> problems, if so how to take care of this for the rest of the
> batch processing steps i.e. do I need to
> change anything in the log files which are created by
> running batch processing commands so that the following/next
> processing steps will take this changes (that I made
> manually) into effect. Is there anywhere I can find
> guidelines how to do this i.e. if there is a manual
> intereference in one of the processing steps how to
> incorporate those changes into the log files so that next
> batch processing commands can go on .
>
> Thanks
>
> Venkateswaran
>
> ________________________________
>
> From: FSL - FMRIB's Software Library on behalf of Stephen
> Smith
> Sent: Tue 10/26/2010 11:38 PM
> To: [log in to unmask]
> Subject: Re: [FSL] VBM results in white matter
>
>
> Hi - where you see such VBM effects at the edges of average
> GM then yes that suggests that the result is an alignment
> artefact and to be treated with caution.
> Cheers.
>
>
>
> On 26 Oct 2010, at 19:22, Rajagopalan, Venkateswaran
> wrote:
>
>
>
> Dear All,
>
> I did VBM analysis using FSL-VBM I am
> seeing significant differences in white matter, what am i
> doing wrong here.
>
> Screeshot attached.
>
> Thanks
>
> Venkateswaran
>
>
--------------------------------------------------------------------
Gwenaëlle Douaud, PhD
FMRIB Centre, University of Oxford
John Radcliffe Hospital, Headington OX3 9DU Oxford UK
Tel: +44 (0) 1865 222 523 Fax: +44 (0) 1865 222 717
www.fmrib.ox.ac.uk/~douaud
--------------------------------------------------------------------
|