Numerical approaches to solve magnetic resonance fingerprinting inverse problem

Magnetic resonance fingerprinting (MRF) is an emerging medical imaging technology based on compressed sensing for measuring multitudes of quantitative NMR properties from tissues in short scan-times. As the underlying physical model of tissues' magnetisation responses used by MRF takes nonlinear and high-dimensional forms, it opens various numerical challenges to solve this inverse imaging problem. In my talk I will introduce you to the MRF paradigm, discuss about these challenges and ways to address them, particularly, in the context of MRF image reconstruction from compressive measurements as well as the MRF multi-compartment tissue separation.