Glasser16-type parcellation on fsaverage

I am new to neuroscience and freesurfer software. Please forgive me if my words and knowledge are incorrect.

[Dataset] I already have a fMRI data preprocessed by others. The data includes surface reconstructed images in fsaverage. That cortical surface is the only thing I want to focus on.

[Goal] My goal is to make a cortical ROI masks so that from the fsaverage space, I can only obtain signals from ROIs. For example, my ROIs are on the visual cortex [v1, v2, v3, mt…]. Therefore, I want to make a ROI mask that only collects signals from each of those regions.

[What I have tried] For another data, the surface reconstruction is in fs-LR (I preprocessed it via fMRIprep with proper flags) and I could use Glasser-16 parcellation. And I could make a mask from ROIs.

So I thought that fsaverage should have similar things as Glasser-16. I did my study to find out what are the atlas for fsaverage. And it seems there are many, but I tried two atlas. I have tried to read atlas for fsaverage using read_annotation function in malab.


But what I found is that their labels are not including v1, v2, v3, MT that I am interested in. I think there are many demands to make a ROI mask for those regions v1, v2 … But The atlas I have tried does not include them. So I am suspecting that I am totally in a wrong direction.

I think my goal is simple and general enough so that there should be already established methods to do that. I have searched for this for many days but it was not successful. If Glasser-16 pacellation is defined on fsaverage, that would be great. But other parcellations on fsaverage including finer regions such as v1, v2, v3, v4, MT should work for me.

Could you help me solving this issue?