---

White matter parcellation from Diffusion Magnetic Resonance Images reveals detailed 3D white matter structure and connectivity of the brain in vivo and thus is a viable tool for investigating neural differences. Measurements derived from these neural tracts or fascicles from Diffusion Tensor Images (DTI) exhibit potential as biomarkers for neurologic and neuropsychiatric disorders. - - - - - - - - - - - - - - - - - - - - We develop and implement a parcellation method that aims for lower sensitivity to initialization and local orientation error by directly segmenting full white matter tracts, rather than reconstructing individual curves, from diffusion tensor fields. A fast, robust volumetric and intrinsically normalized solution is achieved by efficient sampling and noise-filtering using a generic parametrized tract model. Interactive tract modification schemes are also developed to resolve ambiguity in branching tracts or to perform regional investigation for long fascicles.

---

Hon Pong (Gary) Ho has been working on medical and computer vision problems since 1999. He received his PhD of Biomedical Engineering from Yale University in 2011. He has been a senior software engineer at Intel Perceptual Computing in California, USA since 2011, leading R&D of real time interactive computer vision solutions.

---