Reconstruction of HARDI data using a split Bregman optimization approach

Abstract

Among the current tools of diffusion Magnetic Resonance Imaging (dMRI), high-angular resolution diffusion imaging (HARDI) excels in its ability to delineate complex directional patterns of water diffusion at any predefined location within the brain. It is known that HARDI signals present a practical trade-off between their directional resolution and their signal-to-noise ratio (SNR), suggesting the need for effective denoising algorithms for HARDI measurements. The most effective approaches to alleviate this problem have been proven to be those which exploit both the directional and the spatial regularity of HARDI signals. Unfortunately, many of these algorithms entail substantial computational burdens. Accordingly, we propose a formulation of the problem of reconstruction of HARDI signals which leads to a particularly simple numerical implementation. The proposed algorithm allows for the separation of the original problem into procedural steps which can be executed in parallel, which suggests its computational advantage. © 2013 Springer-Verlag.