Hi Xiaoyu,

I don't think it makes sense to model the hemisphere each as a different group. Do you mean different time point?

The problem is that left hemispheres are probably anatomically slightly different than right hemispheres (on average in healthy people).

I would run every subject only once, then do a x-hemi analysis (I think there are tools for that, which try to keep stuff unbiased with respect to the hemi) and then perform your statistic on that.

Since not all diseased hemispheres are on one side, it makes sense to attach a sign to the x-hemi difference so that you always have, e.g. healthy-diseased. Then simply do a GLM on that.

Best, Martin

On Tue, 2018-08-28 at 22:18 +0000, Wang, Xiaoyu wrote:

Hello Freesurfer Experts,   I am trying to compare the left and right hemisphere of a group of PD patients with Mass-Univarate LME.   From what I can see in the wiki tutorial, the mris_preproc only assembles data from one hemisphere at a time, which is not very usefull if I am doing a hemisphere comparison where each hemisphere is a different group.   I am thinking of getting around this issue by flipping each subject time point so that the right hemisphere is on the left and adding it as a new timepoint. For example: S01_1                    s01_base S01_1f                  s01_base S01_2                    s01_base S01_2f                  s01_base   However, Can I still create the base file the same way using this?   recon-all -base <templateid> -tp <tp1id> -tp <tp2id> ... -all   also for the fsaverage, would make_average_subject still work?   Xiaoyu _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer