I run Brainsuite till the end (with registration and labeling and everything) and I get the excel files with the ROI volumes.
I afterwards divide the patients in 2 groups, those with the symptom and those without, and I compare their ROIs (the volumes of them).
I find significance in, for instance, 4 ROIs. That means that by the patients with the symptom these 4 regions are volumetric bigger than in the group without the symptom.
Normally, as a method, Brainsuite makes surface based morphometry, right? Just like freesurfer etc.
But in this case I compare ROIs (voxel numbers). Does it make my analysis a voxel based analysis or is it still a surface based analysis because of the software (Brainsuite)?