 |
 |
Hi Paul and FSL community,
It has been awhile since I posted this question, but I finally had the time to return to this project. Now I am running into an issue with the previously provided advice.
We are attempting to use the Cerebellar Atlas (FNIRT) to create probabilistic ROI masks for sections of the cerebellum. For this example, I will use the Vermis IX as my ROI. We want to segment this ROI and extract the number of voxels contained within the mask.
I created the mask and thresholded using this script:
fslmaths Vermis_IX_FNIRT_mask.nii.gz -thr 50 -bin Vermis_IX_FNIRT_mask_50%thr.nii.gz
This worked without an issue and now we wanted to extract the number of voxels within this mask for our control and clinical participants.
Originally, I proposed on this forum using:
fslmeants -i ID#_T1_BET_FNIRT.nii.gz -m Vermis_IX_FNIRT_mask_50%thr.nii.gz
It was recommended that I used the following instead: fslstats <mask input> -V
First I started with fslstats and received an output of 4662. I then did this on the next participant and got the same result, which I anticipated since I was not calling in any particular T1 structural image. I then tried the fslmeants approach and the output was 63.89 for one participant, and 62.47 for another. Given the size of this cerebellar structure, 62 and 63 voxels seems incredibly small.
I have not found any other ways to extract the number of voxels within the mask. Any advice would be appreciated!
Thank you!
Best,
Victor