Hi Gwenaelle and Jeff
I was actually in the process of quantifying GM volume differences too, and I hope it's ok that I ask a couple of questions. Say for example I had 2 significant clusters in the statistical map of interest, so when I generate a mask using fslmaths mysign_results -thr 0.95 -bin mask_mysign_results, I can see there are 2 non-zero clusters. When I come around to do the fslmeants -i GM_mod_merg_s3 -m mask_mysign_results -o mean_GM_volume.txt command, the output text file only contains 1 GM density value per subject. Is this value a mean of the 2 significant ROIs? I would've thought there would be 2 columns in this case, one for each area?
Also to get a volume estimate, I used the command fslstats mask_mysign_results -V, I got 2 numbers e.g. 10 80.000000. This might be a really silly question but how do I know which value belongs to which ROI?
Thanks for your advice
Cheers
Reem
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Gwenaëlle DOUAUD
Sent: Wednesday, December 01, 2010 2:47 AM
To: [log in to unmask]
Subject: [FSL] Re : [FSL] VBM results volume or focal density?
Hi Jeff,
> Dear FSL_VBM experts,
> Im a bit confused about what the VBM analysis shows. It
> looks to me
> like it investigates focal density differences in GM/WM
> however the
> website http://www.fmrib.ox.ac.uk/fsl/fslvbm/index.html says
> that it
> "investigates voxel-wise changes in the grey matter
> volume/topography". If someone could clarify what VBM
> results mean and
> what the difference between my interpretation of the
> results and the
> websites.
According to Good et al., 2001, the optimised protocol including the Jacobian modulation step leads to results in grey matter "absolute amount" or "volume", while the standard protocol with no modulation leads to (not interpretable) results of "concentration" or "density" (NOT recommended anyway).
> Also is there a way to use VBM to get specific regional
> volume
> differences with actual numbers (say in mm^3)?
Yes, just multiply the volume of the ROI (e.g., whole brain, your significant results etc.) by the mean of the values in the smoothed registered GM inside the ROI.
For instance:
fslmaths mysign_results -thr 0.95 -bin mask_mysign_results
then fslstats mask_mysign_results -V will give you the volume (take the one in mm2) of the ROI, and finally:
fslmeants -i GM_mod_merg_s3 -m mask_mysign_results -o mean_GM_volume.txt will give you the mean GM values for each subject that you'll have to multiply by the volume of the ROI.
Cheers,
Gwenaelle
--------------------------------------------------------------------
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
--------------------------------------------------------------------
__________ Information from ESET NOD32 Antivirus, version of virus signature database 5662 (20101130) __________
The message was checked by ESET NOD32 Antivirus.
http://www.eset.com
__________ Information from ESET NOD32 Antivirus, version of virus signature database 5662 (20101130) __________
The message was checked by ESET NOD32 Antivirus.
http://www.eset.com
