Maximum Likelihood Estimation of T1 Relaxation Parameters Using VARPRO

INTRODUCTION Accurate T1 relaxation time parameter estimation is essential for a wide variety of quantitative MR applications; for example, obtaining accurate T1 values can be crucial in problems such as quantitative contrast enhanced imaging studies, image segmentation and tissue characterization, and absolute metabolite quantification in NMR spectroscopy. Typical experiments designed specifically for T1 estimation often involve acquiring multiple datasets, with the timings within the experiment chosen differently for each dataset. T1 can then be estimated by a parametric fitting of the data with signal intensity formulas derived from the physics of the MR experiment. Despite the wide variety of possible T1 measurement schemes [1], the resulting model signal intensity function often takes the following simplified form (assuming proper phase cycling and/or crushing and that the RF pulse flip angles might be different from their nominal values) [2]: