We aim to model diffusion MRI signals to achieve accurate reconstruction of the orientation density function (ODF). Current work is mostly based on Q-ball imaging, using sparse fits of different classes of kernel functions to robustly model the signal, reject noise, and detect fiber crossings.
The software in this repository builds on DiPy.