There are two ways to do a paired-t. The easiest, I think, is to use the --paired-t option to mris_preproc. The other option is to set it up in the fsgd file, but that's more complicated and less flexible. What do you mean by "group"? A class in the fsgd gets it's on mean regressor and possibly slope regressor. Eg, if you have two groups: male and female, and you want to analyze taking into account handedness, then you'd have 4 classes: ML, MR, FL, FR.
doug
Reschke, Cole wrote:
All-
I am looking to run freesurfer group analysis on a longitudinal study, examining the changes between baseline and follow-up. I would like to set up a paired t-test, yet I am still in need of some assistance creating my FSGD file and my design matrix. I have read the GroupAnalysis wiki page, and did not see anything regarding t-tests. Also, is "Class" is equivalent to "Group" in the FSGD file?
I have been using your sample GroupDescriptorFile to create my own, and pasted an excerpt below (my edits are in red and underlined).
GroupDescriptorFile 1
Title MyTitle
Class bsl plus blue
Class flp circle green
Variables Age Gender
Input subjid1 bsl 20 M
Input subjid1 flp 20 M
DefaultVariable Age
Cole Reschke
Data Manager, PET Center
Banner Alzheimer Institute
Banner Good Samaritan Medical Center
Phone: (602) 239-4868
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer