Abstract 4147159: Feasibility of Cine Magnetic Resonance Fingerprinting for Combined Assessment of Cardiac Structure, Function and Tissue Characteristics in Cardiomyopathy.

Introduction: Cardiac Magnetic Resonance has emerged as the non-invasive standard for assessment of cardiac structure, function and tissue characteristics in patients with suspected cardiomyopathy. However, there are several well-known barriers. Cine magnetic resonance fingerprinting (MRF) is a novel imaging technique that streamlines ‘all-in-one’ assessment of myocardial function and tissue characteristics. It allows for simultaneous quantitative assessment of biventricular size and systolic function, T 1 , T 2 maps and synthetic late gadolinium enhancement (LGE) imaging. This model compensates for various confounding factors, including arrhythmias, slice profile imperfections, inversion pulse efficiency and B1+ inhomogeneities to generate accurate and reproducible measurements. Hypothesis: Cine MRF can be used as an “all-in-one” imaging platform for diagnosis of various cardiomyopathy phenotypes. Goals: We aim to assess the accuracy of cine MRF compared to standard MRI for volumetric analysis and parametric tissue mapping. Further explore diagnostic yield of simulated images with synthetic scar imaging using MRF in comparison to standard LGE imaging. Method: Feasibility study. N= 8; 3 healthy volunteers and 5 patients with reduced ejection fraction (defined as EF <50%) were scanned using 1.5 T MRI. Cine MRF was acquired using retrospective ECG gating (25 cardiac phases, spatial resolution of 1.6x1.6x8.0 mm). Slice duration (10.7s breathhold + 10s pause) of total ~4 minutes for 12 short-axis slices. MRF data were processed using a self-supervised deep learning (Deep Image Prior) reconstruction. Synthetic cine images were derived analytically from the maps. Post-contrast cine MRF maps were used to calculate synthetic PSIR LGE images.Volumetric analysis by 2 individual Level II/III readers. Bland-Altman analysis was used. Further comparative qualitative assessment of standard LGE to cine MRF derived synthetic LGE was performed. Results: Volumetric analysis using standard vs Cine MRF showed high agreement. Cine MRF parametric mapping correlated with standard T1/T2 mapping techniques with expected difference in reference ranges. Qualitative assessment of synthetic LGE images derived from cine MRF allowed for detection or exclusion fibrosis/scar that correlated with standard LGE imaging. Conclusion: Cine MRF holds promise as a single ‘all-in-one’ platform for assessment of myocardial structure, function and tissue characteristics in patients with cardiomyopathy.