Optimized pulse parameters for reducing quantitation errors due to saturation factor changes in magnetic resonance spectroscopy.

Abstract

We present an analysis of the effects of chemical exchange and changes in T(1) on metabolite quantitation for heart, skeletal muscle, and brain using the one-pulse experiment for a sample which is subject to temporal variation. We use an optimization algorithm to calculate interpulse delay times, TRs, and flip angles, theta, resulting in maximal root-mean-squared signal-to-noise per unit time (S/N) for all exchanging species under 5 and 10% constraints on quantitation errors. The optimization yields TR and theta pairs giving signal-to-noise per unit time close or superior to typical literature values. Additional simulations were performed to demonstrate explicitly the dependence of the quantitation errors on pulse parameters and variations in the properties of the sample, such as may occur after an intervention. We find that (i) correction for partial saturation in accordance with the usual analysis neglecting variations in metabolite concentrations and rate constants may readily result in quantitation errors of 15% or more; the exact degree of error depends upon the details of the system under consideration; (ii) if T(1)'s vary as well, significantly larger quantitation errors may occur; and (iii) optimal values of pulse parameters may minimize errors in quantitation with minimal S/N loss.