Hi, all.
I'm a beginner with FSL and having trouble running "run_first_all" for T1 for healthy children in ages 11 to 13. My goal is to get the subcortical areas parcellated for volume measurement and shape comparison.
I'm looking for any advice on how to sort this out. These are the problems I've encountered in running this process. (fyi #2 is my current main concern):
1. I first tried simply running "run_first_all" for the nifti T1 files, and for some reason, a problem would occur in the non-brain structure removal process (BET). There were extensive frontal lobe removal, while non-brain structures were left in the anterior areas of brain stem. So I tried lining the AC-PC subject by subject using
bet <input> <output> -o -c <x y z>.
(This was something I didn't have to do for adult subjects in their 20's so I was bewildered at this point. Is this a common step to be taken?)
2. However while doing so, one of my peers had performed non-brain structure removal in Freesurfer for all of my subjects already, so I decided to use his data for run_first_all instead.
I changed the freesurfer brainmask.mgz files to FSL nifti files using "mri_convert" on Freesurfer.
I reoriented these brainmask.nii.gz to an MNI152 standard template with "fslreorient2std".
Then, I ran these reoriented files with run_first_all using the following command:
run_first_all -i <reoriented brainmask.nii.gz> -o <first_brainmask.nii.gz> -b
While running the command, I found that out of the 70 subjects, 4 subjects failed to go through the command (the command seemed to be stuck on these subjects, so when I checked for the output files, all 4 subjects had only one "first_brainmask.com" file, while other subjects had .com, .com2, .com3 files.)
Of the 66 subjects that did go through the "run_first" command, only 14 subjects yielded the "all_fast_firstseg.nii.gz" files. The other subjects only yielded "all_fast_origsegs.nii.gz" files.
I checked for how the parcellation done for the 14 subjects and found that the right hemispheric region was slanted/squeezed/split, showing results that did not seem right at all.
I really have no clue as of what step I might have missed or have done wrong.
I also did
cat first_brainmask.logs/*.e*
command but it showed me nothing.
Any advice I'll truly appreciate!
Thank you for taking your time to read my concern.
Sincerely, Sooyun Cho.
|