Computation and Informatics in Biology and Medicine (CIBM) Seminar:
There is growing evidence that white matter structural systems in the brain have consistent local and global intrinsic geometries. Additionally, the eigenvectors derived from Diffusion Tensor Imaging data are good indicators of local intrinsic geometry of white matter tissue. This talk describes a novel method for extracting a nonlinear manifold model of the intrinsic geometry of white matter systems of the brain. The Semiparametric Geometric Model takes Diffusion Magnetic Resonance Images as input. Then, using a combination of semi-supervised learning and semiparametric modeling, automatically segments white matter structures and outputs a global nonlinear model of the intrinsic geometry of white matter volumes.

The manifold model allows the data to be organized and sampled in a way that reflects the physical organization of white matter. For example: Geodesic curves are a robust estimate of long range fiber organization; manifold surfaces allow better estimation of cross sectional areas and distances; and manifold volumes are a new, robust way to estimate white matter connectivity between regions of the brain. Finally, new methods of statistical and geometric analysis on the white matter structures are presented.