Accurate myocardial T1 mapping at 5T using an improved MOLLI method: A validation study

Purpose: To develop Coefficient-based Fitting with Adjustment (COFIA), a 5T myocardial T1 mapping method for MOLLI that corrects inversion inefficiency and readout disturbance via empirically estimated factors, addressing limits from inversion factor, readout perturbations, and imperfect recovery. Methods: COFIA uses a modified 5-(3)-3 MOLLI with ECG gating and GRE readout. Inversion at 5T was improved by redesigning the adiabatic pulse (Tan/Tanh vs HSn). The signal model recursively includes an inversion factor (delta) and a correction factor (C); T1 is estimated by nonlinear optimization. Validation included phantom scans and in vivo studies at 5T (21 healthy volunteers, 9 patients). Results: With peak amplitude 10.6 uT, the Tan/Tanh pulse outperformed HSn, yielding average delta = 0.9014 across Delta B0 = +/-250 Hz and B1 = 225-540 kHz. Phantom experiments showed errors within 5%. In vivo, native myocardial T1 (mean+/-SD) was 1468+/-48 ms (apex), 1514+/-39 ms (middle), and 1545+/-50 ms (base); blood T1 was 2182+/-132 ms, 2124+/-153 ms, and 2131+/-158 ms at the three levels. Conclusion: COFIA achieved accurate phantom T1 and feasible in vivo mapping at 5T. Using the standard 5(3)3 MOLLI acquisition, it supports clinical translation of 5T cardiac imaging.