Dear Anders,
What you are doing seems fine - although you should use the -M option to fslstats, not the -m option.
As for the space, the template should be in the same space as the MNI152, so you can use images in the MNI152 space directly. However, as the registrations will have used a different reference image there may be some small differences. Only if very fine detail was a particular concern would I worry about this though. It would be possible to resample your segmented images using the functional analysis registrations, to generate a new GM_mod_merg, and this is probably the easiest way to get them to be matching, but as I said, I'd only worry about this if you think that you are sensitive to very small spatial changes.
All the best,
Mark
On 12 Dec 2012, at 09:10, Anders Hougaard <[log in to unmask]> wrote:
> Dear FSL Experts,
>
>
> I have found some clusters in an fMRI experiment and I would now like to use these in a VBM analysis to see if there are differences in GM between two groups for these specific clusters.
>
> I have done the three steps of FSL-VBM on the data.
> What I plan to do next is to mask the GM_mod_merg with each of the clusters:
>
> fslmaths GM_mod_merg -mas cluster1.nii.gz GM_cluster1.nii.gz
>
> and then extract the mean value:
>
> fslstats cluster1.nii.gz -m
>
> Is this the correct approach?
>
> The problem here is that the clusters are in MNI152 standard space and the VBM output is registered to a study specific template.
> Should I split the GM_mod_merg file into separate volumes, register each of these to MNI space and then perform the procedure above?
>
>
> All the best,
> Anders
>
>
|