Hi Glenn,
part of the output of mris_make_surfaces (in addition to the ?h.white and ?h.pial surfaces) is a measure of the cortical thickness, contained in the file ?h.thickness. You can do group analysis of these files in two ways. For a simple comparison of means or a t-test, you can use mris_twoclass. For a more sophisticated GLM analysis, you can use mris_glm.
cheers, Bruce
On Tue, 30 Nov 2004, Glenn Lawyer wrote:
Hi,
Is there any documentation for the statistical tools available with Freesurfer? We'd like to do a case/control study on cortical thickness and/or curvature in schizophrenia, the type of study mentioned in ^5 Fischl, Bruce, and Dale, A.M., (2000). Measuring the Thickness of the Human Cerebral Cortex from Magnetic Resonance Images. We are able to generate the thickness, curvature, and spherical surfaces from our scans, the question is how to analyze the output.
I understand that there are lots of ways to use Freesurfer with fMRI data, specifically the close link with FSL (FEAT and MELODIC are mentioned once or twice) and all those nice little programs in the $AFNI_DIR (which aren't mentioned in the Freesurfer documentation in any obvious place). Our study, however, is morphological, not functional.
We could consider using FSL/SIENAX to get measures of brain atrophy between case/control, and then map this data onto the Freesurfer cortical surfaces, except then one wonders why we bothered processing all of our scans with Freesurfer (we could display the SIENAX output on Bert's brain, if Freesurfer is only used as a renderer). Further, this doesn't give information on changes in curvature. Nor is it certain that the two measures are equivalent (though in an ideal world they should correspond, both measure changes in white matter in the cortex). Also, I'm not sure if the SIENAX output is directly compatible with Freesurfer in the way that FEAT is.
Thanks,
+glenn