Scan-Rescan Repeatability and Impact of B 0 and B 1 Field Nonuniformity Corrections in Single-Point Whole-Brain Macromolecular Proton Fraction Mapping

Single-point macromolecular proton fraction (MPF) mapping is a recent quantitative MRI method for fast assessment of brain myelination. Information about reproducibility and sensitivity of MPF mapping to magnetic field nonuniformity is important for clinical applications. To assess scan-rescan repeatability and a value of B and B field inhomogeneity corrections in single-point synthetic-reference MPF mapping. Prospective. Eight healthy adult volunteers underwent two scans with 11.5 ± 2.3 months interval. 3T; whole-brain 3D MPF mapping protocol included three spoiled gradient-echo sequences providing T , proton density, and magnetization transfer contrasts with 1.25 × 1.25 × 1.25 mm resolution and B and B mapping sequences. MPF maps were reconstructed with B and B field nonuniformity correction, B - and B -only corrections, and without corrections. Mean MPF values were measured in automatically segmented white matter (WM) and gray matter (GM). Within-subject coefficient of variation (CV), intraclass correlation coefficient (ICC), Bland-Altman plots, and paired t-tests to assess scan-rescan repeatability. Repeated-measures analysis of variance (ANOVA) to compare field corrections. Maximal relative local MPF errors without correction in the areas of largest field nonuniformities were about 5% and 27% for B and B , respectively. The effect of B correction was insignificant for whole-brain WM (P > 0.25) and GM (P > 0.98) MPF. The absence of B correction caused a positive relative bias of 4-5% (P