While going through GroupAnalysis tutorial for freesurfer (MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysis), I was wondering if there was a way to analyze mri_glmfit by regions from an atlas. So instead of looking at vertex-by-vertex or clusters, I wanted to look at areas that are statistically significant by cortical atlas such as Desikan-Killiany or DK-Tourville atlas. 

I understand that mri_glmfit has --label option. Can I just use this to tell mri_glmfit to use atlas label file as a mask then? 

Thank you in advance for your help.

Best,

Julia Shin