hi all ... I was wondering if anyone has suggestions for particular options to
tweak to improve BET in FSL-VBM in the following scenario:
BET is not including portions of the occipital lobe and cerebellar grey matter. I
know there are a variety of parameters that can be played with - any thoughts
for this particular case in FSL-VBM would be most appreciated. I am obviously
new to FSL. This issue is affecting my entire sample.
Another probably very basic question: Some of my BETs are showing a lot of
"speckling" in the white matter. Is this a problem? My subjects are MS patients
and I will eventually have lesion masks to apply which will certainly be essential
to exclude lesions inappropriately identified as GM.
Finally, forgive me, another question: I have a large sample of subjects and,
before trying FSL, I used an in-house brain extraction procedure that includes
FSL BET plus manual editing and other algorithms as I understand. The brain
extractions came out very nice using the in-house procedure. Is there any way
I could use those images in FSL-VBM (and other fsl tools) in lieu of fsl-bet so
as to keep the most brain and lose the most non-brain tissue for other fsl
tools?
Many thanks for taking the time to read my many questions ...
Omar
________________________________
Omar Ghaffar
PhD Candidate
Institute of Medical Science
University of Toronto
|