External Email - Use Caution

Dear all,

I have a subject for whom I need to segment the right hippocampal and amygdalar subfields. However, they have a significant white matter lesion and large ventricles in the right temporal lobe, such that skull-stripping consistently cuts away the cortex in that region, and subsequent aseg and surface reconstruction is inaccurate.

My question is, does the hippocampal subfield module (I'm using the old segmentHA_T1.sh for consistency with our other subjects) use the aseg froom recon-all as a basis for its segmentation? If I just manually correct the hippocampus in the aseg and try running the subfields again, would it work? I've tried adjusting the skull-strip and no combination of wsthresh and no-wgcsaatlas has helped. I'm running FS6.0 on Ubuntu 18.

Thank you!

Regards, Linda