DENOISING OF DYNAMIC MAGNETIC RESONANCE SPECTROSCOPIC IMAGING USING LOW RANK APPROXIMATIONS IN THE KINETIC DOMAIN

摘要

Kinetic monitoring of in vivo metabolism of labelled tracers is based on singular value decomposition or Tucker Decomposition of magnetic resonance spectral image data. Data decomposition is used in conjunction with rank reduction to improve signal-to-noise ratio. Rank reduction can be applied in one or more of a spectral, spatial, or temporal dimension. Rank is generally reduced based on a number of expected analytes/metabolites or fit of measured data to a model.