Diffusion MRI is an in vivo and non invasive imaging technology that uses water diffusion as a proxy to probe architecture of biological tissues. Diffusion MRI technology has been widely used for white matter fiber tracts reconstruction. It also has many clinical applications in neurodegenerative diseases such as Alzheimer's.
In this talk, We discuss various statistical models for analyzing diffusion MRI data. These models aim to elucidate local (voxel-level) neuronal fiber organizations based on D-MRI measurements, which are in turn used as inputs in tracking algorithms to reconstruct white matter fiber tracts. We focus on their capability in resolving crossing fibers -- a major challenge in diffusion MRI data analysis, and their computational scalability. We also discuss spatial smoothing schemes that leverage information from neighboring brain voxels. These methods are applied to both synthetic experiments and to real D-MRI data from large imaging consortium.