Figure 1.

Pipeline for clustering MSNs in the neonatal brain: (a) Cortical regions are defined using Voronoi tessellation of the cortical surface. (b) Each individual region is characterized by an eight-feature vector including averaged normalized values of cortical thickness (CT), mean curvature (MC), myelin index (MI), surface area (SA), fractional anisotropy (FA), mean diffusivity (MD), neurite density index (NDI), and orientation dispersion index (ODI). (c) Each pair of regions is correlated using Pearson’s r, resulting in a subject-specific similarity matrix with size n regions × n regions, which are then averaged to create a group mean similarity matrix, and the resulting group matrix is clustered using affinity propagation algorithm to examine network modularity.