Hi FS group,


My goal is to calculate the % of voxels in a given region that are higher than a region-specific threshold. This is what I've done so far for each subject:



For each subject, for each cortical region (Seg1001-1035 & Seg2001-2035), I now want to count the number of voxels that are above region-specific thresholds as found in thresh.lst, ie.


  ctx-lh-bankssts    Seg1001    1.383697722
  ctx-lh-caudalanteriorcingulate    Seg1002    1.272858762
  ctx-lh-caudalmiddlefrontal    Seg1003    1.167540497
  ...

Do you have any advice on the best way to do this? I think I can use mri_segstats, but am not sure if I'm looking at the right arguments and/or if there is a more straightforward way to do this. This is what I have so far:


  foreach s (`cat s.lst`)   /   foreach seg (`cat thresh.lst | awk '{print $2}'`)

  set thresh = `cat thresh.lst | grep $seg | awk '{print $3}'`

  mri_segstats --i vol.nii --seg aparc+aseg.functionalspace.nii --id $seg --maskthresh $thresh --masksign pos --sum tmp.$seg.stats

  ...and then grep NVoxels from tmp.$seg.stats, grep total NVoxels from vol.aparc+aseg.stats, divide/echo into summary stats file?



Thank you in advance for any advice!


Best,

Sara