Adjacent Surface ROIs from Atlas

Hello NeuroStars!

I would like a measure of edge-ness for each parcel within the Yeo/Schaefer 400 parcel 7 network atlas.

I know that the network likelihood metric is already published for the 17 and 7 network atlases. I can average this metric within parcel to get a value of network likelihood.

However, I would like to know if there is a method for pulling the adjacent ROIs. Each ROI shares sides with 6 ROIs (apart from parcels bordering the medial wall). I would like a list of these adjacent ROIs for each ROI.


I guess that you are implicitly referring to fsaverage7 model;
then you can easily obtain an adjacency graph from this mesh (I can provide code that does that)
This graph is then easily represented by a sparse matrix.
Taking the membership matrix from the parcellation, you can then by simple sparse matrix multiplication create a reduced graph that links regions: non-zero edges of this graph then represent neighboring regions.

Thanks, @bthirion. Messaged you!