Hi Rachel,
yes, there is a difference if you compute SPC of the average thickness in an ROI or if you average the SPC computed on a vertex level.
For the second, you can use long_mris_surf to map your third group to fsaverage and compute SPC, there is an option to write out the stack, then using the label and that stack you can average the values inside the label. I would do that in Matlab (read the stack, read the label, select the vertices with that label and average). Maybe there are other tools for that.
Best, Martin
On Sep 16, 2016, at 5:05 PM, Rachel Romeo rromeo@mit.edu wrote:
Hi,
I have 3 participant groups in my data set, all with 2 longitudinal time points approximately evenly spaced. I've run everyone through the longitudinal pipeline, and used the Monte Carlo simulation to find a cluster that differs in SPC between groups 1 and 2. Now I want to calculate the average SPC in that cluster for group 3, simply for qualitative reference.
I've extracted the cluster to a label using mri_surfcluster, mapped that label to each of the participants' surfaces using mri_label2label, and used mris_anatomical_stats to extract the average thickness in the ROI, and then I manually calculated SPC per subject. But, no matter whether I use participants' CROSS or LONG surfaces, I can't replicate the average SPC values in the "y.ocn.dat" file generated by the simulation.
My guess is that the difference arises because of the averaging -- I'm averaging over thickness, and then calculating SPC, whereas I'm guessing that the GLM calculates SPC per voxel, and then averages over the cluster.
Is there a way I can extract the average SPC for group 3 over this ROI?
Thanks, Rachel
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer