Hi Christine
why don't you upload the subject and we'll take a look
cheers Bruce On Tue, 16 Jul 2013, Christine Smith wrote:
I have a subject for whom we have had consistent difficulties within freesurfer. The difficulties existed for a single scan from 2008 and also for a pair of scans (averaged together) that were collected in 2013. The difficulty is that freesurfer will fail to label white matter as white matter. The problem is not small like the one listed in the troubleshooting area of the wiki. In some cases an entire lobe is not labeled. There are also instances where freesurfer will delete areas of brain out of the brainmask. Again, these are large areas of the brain, including regions where we have control points. I have included a picture of what is happening. We are going on 4000+ control points for this person. I should mention that after we address one problem in one part of the brain (by adding hundreds of control points and also drawing in the white matter underneath on the the wm.mgz) it will still fail to capture the brain properly. It will also then decide that another part of the brain (which had been parcellated perfectly before) is now not considered part of the brainmask or has no white matter.
The picture shows these problems perfectly.
Any advice about how to get this to work?
Regards, Christine
-- Christine N. Smith, Ph.D. Department of Psychiatry University of California, San Diego