Hi Brendan
the time to correct goes with the square of the size of the defect (actually the convex hull of the defect) so it can take a very long time for big ones. More importantly, if the defect is that large it may not be corrected in the anatomically desired manner. You should take a look at it and see what is causing it. Is skull left over? Cerebellum attached? Hemispheres connected? Usually the big ones are due to something dramatic. Look at the ?h.orig.nofix and the ?h.inflated.nofix. YOu can load the ?h.defect_labels on top of the surfaces and brain.mgz and the wm.mgz to see which one is 29 (note that they defect_labels file is probably 1-based so you would need to look for 30 in it).
If you can't sort it out gzip and tar your subject and put it on our ftp site and we will take a look for you cheers Bruce
On Fri, 9 Aug 2019, Balken, Brendan wrote:
External Email - Use Caution
Hello!
So I am attempting to use freesurfer’s recon-all script but for one of our subjects I keep getting stuck at this point: “CORRECTING DEFECT 29 (vertices=19321, convex hull=4029, v0=32423) XL defect detected...” It has already corrected 28 defects, but they were much smaller. It has been running for about 48 hours on a very powerful machine and I’m wondering how long I should wait. Will it eventually move on? If no, what is my contingency plan for processing this? I’m using the output anatomy folder in brainstorm for source analysis. Any guidance or information would be greatly appreciated.
Thanks a million!
Brendan
Here’s what the lh and rh.inflated.nofix look like (is that the defect?):
[IMAGE]