Meeting Banner
Abstract #0859

Fully Automated Probabilistic White-Matter Tractography with Anatomical Priors: Application to Huntington's Disease

Anastasia Yendiki1, Allison Stevens1, Jean Augustinack1, David Salat1, Lilla Zollei1, Ruopeng Wang1, Diana Rosas2, Bruce Fischl1,3

1HMS/MGH/MIT Martinos Center for Biomedical Imaging, Charlestown, MA, USA; 2Department of Neurology, Massachusetts General Hospital, Boston, MA, USA; 3Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA, USA

We illustrate the application of a robust and fully automated method for probabilistic white-matter tractography to analyze diffusion-weighted MR images from a large cohort of Huntington's disease patients and matched healthy controls. The method uses manually labeled paths from a set of training subjects to construct priors on these paths. The priors are then incorporated into a probabilistic tractography framework to trace the paths automatically in the test subjects (Huntington's disease patients and controls). Preliminary results show significant decreases of the fractional anisotropy in several parts of the corticospinal tract and superior longitudinal fasciculus of the patient population.