Thanks for posting back, Bruce.

 

Mri_label2label –projabs N does work but only dilates normal to a surface.

What I’m trying to do is the following.

For cortical region X, I want to identify the sub-volume of each adjacent cortical region which is adjacent to region X. The idea is to dilate region X, say 5 mm, and then use sort and uniq to identify the xyz coordinates in the dilated X which are coincident with xyz coordinates in each adjacent region. It doesn’t matter if the dilated region X has lots of points in common with region X and I can see a brute force way to do what I want and can write it and contribute it to you if it would be helpful. I was just looking for a freesurfer tool or tool combination which does it. I think the idea of adjacent subregions is a useful one.

 

 

Hi Don

 

This is more Doug’s domain than mine, but won’t mri_label2label do the trick?

 

Cheers

Bruce

 

 

        External Email - Use Caution        

Hi everyone,

 

I’m reposting this from November 29 as it hasn’t gotten a reply and it hasn’t shown up in the archive either so maybe it got lost in the sauce.

 

I wish to use the “-dilate n” capability of mri_extract_label, mris_label or other tool to dilate a ctx region and end up with an ascii list of xyz coordinates for the expanded points.

mri_extract_label appears to only be capable of writing output to .mgz files and I do not find a tool for extracting labels from a .mgz into an ascii .label file.

Your direction would be welcome.

 

Thanks -  Don

 

Don Krieger, Ph.D.

Research Scientist

Department of Neurological Surgery

University of Pittsburgh