Hi Cate,
there is a section on the wiki about troubleshooting the skullstripping. Try some of those recommendations first (mostly changing the preflooding height). If the cerebellum is messed up, then the aseg probably won't work well which can hurt the whole processing stream.
cheers, Bruce
On Mon, 5 Feb 2007, Catherine Hartley wrote:
Hi,
I ran autorecon1 on a group of data consisting of a single set of mprage DICOMs per subject, acquired on a Siemens Allegra 3T scanner. Several of the brainmask.mgz images had large portions of the brainstem and cerebellum removed. A couple had portions of the neck which were not removed. The talairach transforms looked fine.
A more experiened Freesurfer user suggested that adding control points and re-running the intensity normalization might fix this. I assume that I can follow the instructions here,
http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial_2fControlPoints
but that I would need to edit the orig.mgz file and then run autorecon1-cp. Is this correct?
My intended analysis (cortical thickness in PFC regions) does not depend upon these flawed regions, but I want to make sure these errors aren't indicative of some larger problem. Also, I know there have been issues with gradient warping with the Allegra. Do these problems sound like gradient warping symptoms?
Thanks, Cate
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer