Application of compressed sensing to multidimensional spectroscopic imaging in human prostate

Abstract The application of compressed sensing is demonstrated in a recently implemented four‐dimensional echo‐planar based J ‐resolved spectroscopic imaging sequence combining two spatial and two spectral dimensions. The echo‐planar readout simultaneously acquires one spectral and one spatial dimension. Therefore, the compressed sensing undersampling is performed along the indirectly acquired spatial and spectral dimensions, and the reconstruction is performed using the split Bregman algorithm, an efficient TV‐minimization solver. The four‐dimensional echo‐planar‐based J‐resolved spectroscopic imaging data acquired in a prostate phantom containing metabolites at physiological concentrations are accurately reconstructed with as little as 20% of the original data. Experimental data acquired in six healthy prostates using the external body matrix “receive” coil on a 3T magnetic resonance imaging scanner are reconstructed with acquisitions using only 25% of the Nyquist–Shannon required amount of data, indicating the potential for a 4‐fold acceleration factor in vivo, bringing the required scan time for multidimensional magnetic resonance spectroscopic imaging within clinical feasibility. Magn Reson Med, 2012. © 2012 Wiley Periodicals, Inc.