Magnetic Resonance Spectroscopic Imaging (MRSI) is a valuable tool to characterize metabolic concentrations and changes in several spatial locations in a single recording. However, the long acquisition time of conventional three-dimensional (3D) MRSI limits its practical application. Non-uniformly sampled 3D echo planar spectroscopic imaging (EPSI) has been proposed to accelerate the scan time, combined with compressed sensing (CS) to retain reconstruction fidelity. We apply the novel approach of reconstructing 3D EPSI data by applying TV, Wavelet-$$$\ell_{1}$$$, and TV + Wavelet-$$$\ell_{1}$$$ CS-based regularization on both the combined spectral and two undersampled spatial dimensions. These three reconstruction methods were evaluated in both simulated and in retrospectively undersampled data of a brain phantom.
This abstract and the presentation materials are available to members only; a login is required.