Hi,
You can extract the values that are used in the statistical testing (after segmentation, registration, modulation and smoothing) by using fslmeants with the same input as used for the randomise call in VBM (e.g. GM_mod_merg_s3). You can specify a mask with fslmeants, and this should be a binary mask formed from the significant cluster that you are interested in (see the documentation on "cluster" for how to extract clusters from the output of randomise).
All the best,
Mark
On 21 Feb 2013, at 21:45, Lara Foland-Ross <[log in to unmask]> wrote:
> Dear FSLers,
>
> I have run a VBM analysis of patients and controls and localized an area of significant GM reduction in my patient group. I'd next like to extract single subject GM values from this significant cluster to use in post-hoc correlation analyses. How can I go about doing this?
>
> Thanks so much,
> Lara
|