External Email - Use Caution
Dear all,
I have a subject whose Freesurfer processing crashes during topology fixing step. The last lines in `recon-all.log` are:
After retessellation of defect 2 (v0=22890), euler #=-15 (110444,329616,219157) : difference with theory (-18) = -3 CORRECTING DEFECT 3 (vertices=8808, convex hull=2020, v0=31369) Segmentation fault (core dumped) Linux compute-0-2.local 2.6.32-504.16.2.el6.x86_64 #1 SMP Wed Apr 22 06:48:29 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux recon-all -s 81489f2bb-M12_T1_01 exited with ERRORS at Tue Apr 27 21:23:59 CEST 2021
For the full `recon-all.log` and `recon-all.error` files please see the attached files.
I tried to follow the manual topology fixing instructions from https://secure-web.cisco.com/1CVb3gwDr9GTwil5kr7FQYCZpGmZhLu9ovxhkC6nEa1QhFJ..., but *.white and *.pial surface files are missing from `surf` directory. It only contains these files:
lh.defect_borders lh.defect_labels lh.orig.nofix lh.smoothwm.nofix rh.orig.nofix rh.smoothwm.nofix lh.defect_chull lh.inflated.nofix lh.qsphere.nofix rh.inflated.nofix rh.qsphere.nofix
Which files can I use in place of white and pial surface files in order to look for WM segmentation defects?
Some additional information about the data:
- I am running Freesurfer 7.1
- overall, there are no gross errors in WM segmentation, although the subjects does have significant temporal lobe atrophy
- there are two images acquired at different times, and processing fails for both images
- interestingly, processing didnt crash with Freesurfer 6.
Thank you and have a great weekend,
Renat.