Correction and Normalization of Magnetization Transfer Ratio Maps for Quantitative Analysis

Introduction Magnetization Transfer (MT) imaging is of increasing interest in neuroimaging, yielding important information, e.g. about myelin integrity in the central nervous system [1, 2]. However, a quantitative analysis of the MT ratio (MTR) is affected by several problems, especially at high field strengths: B1 inhomogeneities influence the MT saturation angle, biasing MTR values. High specific absorption rates (SAR) may require a reduction of the saturation angle for individual subjects, impairing comparability of results in longitudinal studies. Purpose of this study was (1) to determine calibration parameters valid at 3 Tesla to correct for B1 bias in MTR maps similar to a method described for 1.5 Tesla [3], and (2) to test if MTR data acquired at 10% and 20% lower MT saturation angles can be normalized to allow for comparison with data acquired at the full angle.