Hi
I am new to the world of FMRI and would really appreciate any help on the
following. I am dealing with lesioned pediatric Brains. I need to use an
effective skull stripping algorithm and then use a segmentation algorithm to
find lesion volumes, CSF, white matter, Grey matter etc. I am using AFNI to
look at the lesions. I did look up BET and FAST. The problem I am facing
with BET is that the Brain stem and cerebellum also comes into the picture.
So then I tried using Freesurfer. The problem I faced there was with very
large lesions extending from the ventricle to the cortex, Freesurfer did get
rid of the brainstem for me but also got rid of a part of the cortex (didnt
give very accurate Results). I then used FAST on these outputs and I got
volumes for CSF, white matter and grey matter. But obviously it is not
accurate. Could anyone please suggest a method for stripping off the skull
for these lesioned brains? I guess after that it is just a matter of using
FAST. Or could anyone suggest an entire new way to look at volumes? Any help
would be appreciated. thank you
NL
|