Hi JP,

I assume you mean FreeSurfer as opposed to FSL. Otherwise, it could be my Beijing Blues, but I'd have no idea how to help out with this one. So let us assume it is FreeSurfer of which we speak.

If you have a cluster of computers somewhere handy, I'd say install the latest FreeSurfer and fire off 220 reconstruction commands like this one: recon-all -subjid Joe -all -i Joe/anat.nii . It take 1-2 days per subject per cpu but if you have a cluster handy that is not a problem. You'd have results by the time we finish sorting out option 2. SUMA works quite well with FreeSurfer's ouput. Under rosy scenarios, you could be doing group comparisons by the end of the week.

Option 2: I have not worked with BrainSuite data but I can help you port things over if we have some help on the other side. The first question is does BrainSuite write out GIFTI surface format? This is the equivalent of NIFTI for surface-based programs. If we have a yes, then that's a start, otherwise we'll need a description of the BrainSuite file format and I'll need some time to write I/O functions for it.
The next step is to understand how node-based correspondence between subjects is established. Do they use the same mesh to generate all subjects' surfaces? Do they do any surface-based registration (if yes then how?) or do they start from volumes that are in standard space and use the same mesh for all subjects?
Once we know the answers to this set of questions we can continue the long march forward.

cheers,
z