Dear Pierre,

I'm reporting some clinical results on fMRI connectivity analyses, and we are handling a small volume in the brain, the hippocampus, across our group of patients with temporal lobe epilepsy. The reason behind this concern is actually that our resampled voxel size was quite large, 4 mm x 4 mm x 4mm isotropic voxels. 

So I'm wondering how I can best report the quality of the nonlinear warp performed using NIAK, to compare if the same voxel represents the same area (part of hippocampus) across subjects in our study. Any suggestion?

While my connectivity analyses was based on my own development, I preprocessed my T1 and BOLD images from individual patients using the standard pipeline suggested by NIAK.

Thank you for your help.

Best,
Kangjoo