Using Spatial Prior Knowledge in the Spectral Fitting of Magnetic Resonance Spectroscopic Images

We propose a Bayesian smoothness prior in the spectral fitting of magnetic resonance spectroscopic images which can be used in addition to commonly employed prior knowledge. By combining a frequency-domain model for the free induction decay with a Gaussian Markov random field prior, a new optimization objective is derived that encourages smooth parameter maps. Using a particular parametrization of the prior, smooth damping, frequency and phase maps can be obtained while preserving sharp spatial features in the amplitude map. A Monte Carlo study based on two sets of simulated data demonstrates that the variance of the estimated parameter maps can be reduced considerably, even below the Cramér-Rao lower bound, when using spatial prior knowledge. Long echo time H-MRSI at 1.5T of a patient with brain tumor shows that using the spatial prior resolves the overlapping peaks of choline and creatine also when a single voxel method fails to do so. Improved and detailed metabolic maps can be derived from high spatial resolution short echo H-MRSI at 3T. Finally, the evaluation of four series of long echo time brain MRSI data with various signal-to-noise ratios shows the general benefit of the proposed approach. (190 words)