depends on how you want to do it. You can create a contrast to test the interaction (probably 0 0 1 -1) then look for significant clusters. Or you can extract the thicknesses out of the clustter for each subject (already there with the y.ocn file), then do a post hoc test in SPSS, matlab, etc
doug
On 12/23/2013 03:39 AM, jh kim wrote:
Dear Doug and experts
Using DODS model in QDEC, I've found 2 significant clusters in whole-brain cortical thickness analysis between controls and patients (2 group, 1 factor), controlling for the effect of age (one covariate). My question is how can I obtain p value for interaction term (group*age) in each cluster? Is it possible to do that analysis in FreeSurfer pipeline? Or should I extract individual cortical thickness of each cluster and test this interaction term using statistics software (e.g., SPSS)?
Thank you in advance for your help. Kim
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer