Hello All,
To begin I'm running 4.0.1 on Mac OSX Tiger (Intel).
I'm having trouble with one subject in particular failing before generating the aseg.auto.mgz or aseg.mgz. I don't see anything amiss in the previously generated files and images, however, this subject has extremely large ventricles, which I think may be the issue. I get the same error whether running autorecon2, recon-all -calabel, etc.
Here's the last some-odd lines of the log file:
gca peak Unknown = 0.94091 ( 0) gca peak Left_Inf_Lat_Vent = 0.15806 (38) gca peak Left_Thalamus_Proper = 0.10842 (113) gca peak CSF = 0.21568 (39) gca peak Left_Accumbens_area = 0.39370 (75) gca peak Left_undetermined = 1.00000 (34) gca peak Left_vessel = 0.36599 (60) gca peak Left_choroid_plexus = 0.07952 (51) gca peak Right_Inf_Lat_Vent = 0.21306 (34) gca peak Right_Thalamus_Proper = 0.09033 (102) gca peak Right_Caudate = 0.20951 (79) gca peak Right_Hippocampus = 0.24233 (70) gca peak Right_Accumbens_area = 0.32477 (77) gca peak Right_vessel = 0.58618 (58) gca peak Right_choroid_plexus = 0.14637 (47) gca peak Fifth_Ventricle = 0.96184 (31) gca peak WM_hypointensities = 0.12793 (82) gca peak non_WM_hypointensities = 0.07548 (80) gca peak Optic_Chiasm = 0.48149 (75) not using caudate to estimate GM means setting label Left_Hippocampus based on Right_Hippocampus = 1.00 x + 0 setting label Left_Caudate based on Right_Caudate = 1.00 x + 0 estimating mean gm scale to be 0.99 x + 0.0 estimating mean wm scale to be 1.00 x + 0.0 estimating mean csf scale to be 0.93 x + 0.0 46135 gm and wm labels changed (%25 to gray, %75 to white out of all changed labels) 401 hippocampal voxels changed. 0 amygdala voxels changed. pass 1: 101186 changed. image ll: -2.438, PF=1.000 pass 2: 16699 changed. image ll: -2.435, PF=1.000 pass 3: 6379 changed. pass 4: 2882 changed. Segmentation fault Darwin hlcpro2.psychology.wustl.edu 8.11.1 Darwin Kernel Version 8.11.1: Wed Oct 10 18:23:28 PDT 2007; root:xnu-792.25.20~1/RELEASE_I386 i386 i386
recon-all exited with ERRORS at Thu Feb 21 23:29:10 CST 2008
I've seen a -nobigventricles flag, is there perhaps a -bigventricles option? Any suggestions are appreciated. Thanks! -Marlisa
Marlisa Isom Research Coordinator Head Research Laboratory Washington University in St. Louis