Abstract #3236

Song T, Gasparovic C, Devier D, Bockholt J, Caprihan A, Mullins P, Posse S, Morrison L, Jung R

A strategy for using tissue water as a concentration standard in 1H magnetic resonance spectroscopic imaging studies on the brain is presented and the potential errors that may arise due to the partial volume effect and tissue classification are examined. In particular, different estimates of the cerebrospinal fluid contribution to spectroscopic voxels, based on different segmentation methods, are shown to lead to a large variability in estimates of metabolite concentrations in gray matter.