Demo of clustering pipeline

Data source was a single-subject resting state scan. Blah blah preprocessing parameters, blah. Pairwise Pearson product-moment coefficient computed for all voxels. Then this was turned into:

Eta2 matrix

Note: The eta2 matrix only considers voxels in a sphere around medial parietal cortex, because I was trained by Jody Culham and parietal cortex >>> all other cortices.

Graph modularity clusters (Louvain)

Modularity score: 0.705579 (fraction of edges that fall within clusters minus the expected fraction if edges were distributed at random; scale -0.5-->1.0, so 0.7 is very good!)

Machine learning clusters (k-means)