Exploiting Spatial Information for Estimating Metabolite Concentration in MRSI

INTRODUCTION Magnetic Resonance Spectroscopic Imaging (MRSI) is a noninvasive monitoring technique that provides significant biochemical information on the molecules of the organism under investigation. Metabolite quantification of MRSI data needs to be accurate to help in brain tumor diagnosis. The quantities of interest are the metabolite concentrations, which can be computed from the weighting coefficients (amplitudes) of the linearly combined in vitro profiles. In this study we propose a new quantification method for MRSI data, which exploits spatial prior knowledge. Previous studies showed that the exploitation of spatial context led to significantly better results for quantification, see [1].