Bo Zhao1, Justin P. Haldar2, Kawin Setsompop1, and Lawrence L. Wald1
1Martinos Center for Biomedical Imaging, Chalestown, MA, United States, 2Signal and Image Processing Institute, University of Southern California, Los Angeles, CA, United States
A principled
framework is proposed to optimize the experiment design for magnetic resonance
fingerprinting (MRF) based on the Cramer-Rao bound. Within this framework, we optimize
the acquisition parameters (flip angle, TR, etc.) to maximize the SNR
efficiency of quantitative parameter estimation. Preliminary results indicate
that the optimized experiments allow for substantially reducing the length of
an MRF acquisition and substantially improving
estimation performance for the T2 map, while maintaining similar accuracy level
for the T1 map. The proposed framework should prove useful for fast quantitative
MR imaging with MRF.
