We hypothesized in vivo MR spectra can be sufficiently described by spectra representing predominant cell/tissue types rather than individual metabolites. We sought to extract two putative basis spectra of grey matter and white matter from 584 single-voxel 3T MR spectra using the principal spectra analysis by linear modeling (PSALM) custom algorithm. Two extracted PSALM bases explained >95% of fit variance; principal component analysis required 6-8 components to achieve the same. We found the algorithm produced high signal-to-noise, low linewidth basis spectra that robustly fit in vivo spectra from normal brain.
This abstract and the presentation materials are available to members only; a login is required.