Statistical shape analysis of brain arterial networks (BAN)

The arterial networks in the human brain, termed brain or BANs, are complex arrangements of individual arteries, branching patterns, and interconnectivity. BANs play an essential role characterizing understanding physiology, one would like tools for statistically analyzing shapes BANs. These include quantifying shape differences, comparing populations subjects, studying effects covariates on these shapes. This paper mathematically represents analyzes BAN as elastic graphs. Each graph consists nodes, points 3D, connected by 3D curves, edges, with arbitrary We develop a mathematical representation, Riemannian metric other geometrical tools, such computations geodesics, means, covariances, PCA, helping analyze apply this analysis to after dividing them into four components—top, bottom, left, right. framework is then used generate summaries from 92 subjects study age gender components. While require further investigation, we conclude that has clear, quantifiable effect Specifically, find increased variance increases.