Hi Sam, it sounds like what you are doing is correct. Another possible solution would be to use mri_binarize, something like mri_binarize EXAMPLE.nii.gz --match 17 --o left.hippo.nii.gz This matches the number rather than doing a thresholding. It seems like the thresholding should work, but there might be some issue internal to fslmaths that is not handling the numerical precision correctly. doug

On 08/16/2012 09:51 PM, Sam Carpenter wrote:

Dear life savers, headache relievers and brain science dreamers,

PROLOG: I'm trying to separate out some regions from a .nii.gz file. I start with a surface parcellation in .annot (Freesurfer land) that I convert to EXAMPLE.nii.gz using mri_aparc2aseg.

THE GOAL: To get a .nii.gz file for each one of these regions. Using fslmaths, I threshold the EXAMPLE.nii.gz file above& below the intensity of that region& save as REGION_name.nii.gz.

THE PROBLEM: This method is working for the majority of the regions. But many of the output REGION_name.nii.gz files are completely empty. They are there in the original EXAMPLE.nii.gz but are lost in the thresholding some(how/where).

THE SOLUTION: Your kind reply...

Thanks very much. -Sam

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer