Joint estimation of spherical harmonic coefficients from magnitude diffusion-weighted images with sparsity constraints

This paper presents a new method to jointly estimate the spherical harmonic coefficients for all the voxels from noisy magnitude diffusion-weighted images acquired in high angular resolution diffusion imaging. The proposed method uses a penalized maximum likelihood estimation formulation that integrates a noncentral χ distribution based noisy data model, a sparsity promoting penalty on the spherical harmonic coefficients and a joint sparse regularization on the diffusion-weighted image series. An efficient algorithm based on majorize-minimize and alternating direction method of multipliers is proposed to solve the resulting optimization problem. The performance of the proposed method has been evaluated using simulated and experimental data, which demonstrate the improvement over conventional methods in terms of estimation accuracy.