MR fingerprinting with implicit neural representation (FINR) for free-breathing 3D whole-liver water T1, water T2, fat fraction, and R2* mapping

Motivation: Water T1, water T2, R2* and PDFF are quantitative measures that have shown utility in the diagnosis of various liver diseases but currently require separate acquisitions, typically using many breath-holds. Goal(s): To develop a single MR acquisition for free-breathing 3D whole-liver quantification of water T1, water T2, PDFF, R2*. Approach: We propose a neural network using implicit neural representation (INR) which simultaneously learns the motion deformation fields and the static reference frame MRI subspace images. Results: Our results showed small bias and narrow 95% limits of agreement on T1, T2, R2* and PDFF values compared to conventional breath-holding scans. Impact: Our work enables 3D whole-liver quantification of water T1, water T2, PDFF, and R2* in a single free-breathing MR acquisition. It also provides a novel solution to reconstruct 5D MRI images (3D spatial + contrast + motion dimension) using INR.