It seems that mri_robust_register crashed. Martin, any ideas? The algorithm worked because the rough registration from the previous step was apparently enough to yield a decent segmentation. Maybe we should consider stopping the whole process if mri_robust_register dies...
Juan Eugenio Iglesias Postdoctoral researcher BCBL www.jeiglesias.com www.bcbl.eu
Legal disclaimer/Aviso legal/Lege-oharra: www.bcbl.eu/legal-disclaimer
From: "Vincent Beliveau" vincent.beliveau@nru.dk To: "Freesurfer support list" freesurfer@nmr.mgh.harvard.edu Sent: Friday, May 13, 2016 2:40:11 PM Subject: [Freesurfer] segfault using brainstem module from dev version
Hi list (and Eugenio),
I've ran the brainstem module from the dev version on some data processed with FS v5.3. The module created some great segmentations of brainstem but when taking a closer look at the log file, there is 2 seg faults which occur when calls to mri_robust_register are made, with the following output (full log attached, see line 319 and 371):
/data1/vbeliveau/software/freesurfer-Linux-centos4_x86_64-dev-110516/bin//mri_robust_register: line 3: 36835 Segmentation fault mri_robust_register.bin "$@" /data1/vbeliveau/software/freesurfer-Linux-centos4_x86_64-dev-110516/bin//mri_robust_register --mov imageDump.mgz --dst /indirect/data1/vbeliveau/atlas/proc/MR/recon_final/f5249_GD/tmp/BS_T1based//BS-DE-binaryMask_autoCropped.mgz -lta trash.lta --mapmovhdr imageDump_coregistered.mgz --sat 50: Segmentation fault mv: cannot stat ?imageDump_coregistered.mgz?: No such file or directory
I'm curious about how is this affecting the overall process, as otherwise everything appears to be working smoothly.
Best,
Vincent.