Maxime Descoteaux1, Jeff Calder2, Cyril Poupon1, Fabrice Poupon1, Rachid Deriche2
1NeuroSpin, IFR 49, CEA Saclay, Gif-sur-Yvette, France; 2Odyssee Projet, INRIA, Sophia Antipolis, France
We propose an optimal real-time system for q-ball imaging using a regularized Kalman filtering framework and incremental orientation sets. We show that the ODF reconstruction for every incoming diffusion measurements. We also show that the orientation set can be computed on the fly with an efficient recursive algorithm.