You can also run recon-all on the mni152, then run. Load it with your volume-based atlas to make sure that the alignment is good and that the surfaces surround your labels, then run mri_vol2surf (--projfrac 0.5) on your volume-based atlas to create a surface-based atlas, then run mris_seg2annot to create the annotation (run with --help to get more docs)

On 10/11/2021 10:09 AM, Fischl, Bruce wrote:

Hi Paul

mri_vol2label maps the volume into an ascii label structure but has no knowledge of surfaces. You then need to run mri_label2label with options like –sample to sample it onto a surface

cheers

Bruce

*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu *On Behalf Of *Paul Dhami *Sent:* Sunday, October 10, 2021 6:53 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* [Freesurfer] Creating a surface parcellation from a volume atlas

*        External Email - Use Caution *

Dear Freesurfer community,

I have a .nii volume atlas (CAREN), defining 5 networks, in MNI (ICBM152) space.

However, I wish to do my analysis in surface rather than volume space, and for the software I am using, that means I need to load in a .annot file.

For the first step, I used:

mri_vol2label --c CAREN_5networks.nii --id 1 --l 1.label

and did this for values 1 to 5 (for each of the networks). However, looking at the resulting label file, the first column is nothing but -1 values. I believe this should be the vertex number, and trying to combine these labels into an .annot file is not working.

Apologizes for the naive question, but can anyone share as to how I can ultimately create a surface atlas when I have the volume.nii atlas file? Was I wrong with mri_vol2label as my first step?

I am using Freesurfer version 7.2.0.

Thank you, Paul

---

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