John M. Ollinger1, Samuel A. Hurley2, Andrew L. Alexander3, Gerard Riedy1
1NICoE, Walter Reed National Military Medical Center, Bethesda, MD, United States; 2Medical Physics, University of Wisconsin-Madison, Madison, WI, United States; 3Medical Physics, University of Wisconsin, Madison, WI, United States
The parameters of the two-compartment mcDESPOT model for myelin water fraction are estimated with a non-linear least-squares algorithm constrained by a uniform prior. Combined with improved flip angle calibration, flip angle selection, and phase-cycling angles, the method yields estimates of myelin water fraction of visually high quality. However, estimates of myelin T1 and myelin residence time were strongly dependent on the choice of constraint and starting values. Quantitative values of myelin water fraction depended on these values as well as the parameterization of the model. This suggests that myelin water fraction images may represent a combination of physiological parameters.