REMOVAL OF LIPID ARTIFACTS IN H-1 SPECTROSCOPIC IMAGING BY DATA EXTRAPOLATION

Proton MR spectroscopic imaging (MRSI) of human cerebral cortex is com plicated by the presence of an intense signal from subcutaneous lipids , which, if not suppressed before Fourier reconstruction, causes ringi ng and signal contamination throughout the metabolite images as a resu lt of limited k-space sampling. In this article, an improved reconstru ction of the lipid region is obtained using the Papoulis-Gerchberg alg orithm, This procedure makes use of the narrow-bandlimited nature of t he subcutaneous lipid signal to extrapolate to higher k-space values w ithout alteration of the metabolite signal region, Using computer simu lations and in vivo experimental studies, the implementation and perfo rmance of this algorithm were examined. This method was found to permi t MRSI brain spectra to be obtained without applying any lipid suppres sion during data acquisition, at echo times of 50 ms and longer, When applied together with optimized acquisition methods, this provides an effective procedure for imaging metabolite distributions in cerebral c ortical surface regions.