Abstract
PURPOSE: This study aims to show the viability of conducting three-dimensional (3D) myocardial perfusion quantification covering the entire heart using both GRE and bSSFP sequences with hyperpolarized HP001. METHODS: A GRE sequence and a bSSFP sequence, both with a stack-of-spirals readout, were designed and applied to three pigs. The images were reconstructed using 13 C coil sensitivity maps measured in a phantom experiment. Perfusion was quantified using a constrained decomposition method, and the estimated rest/stress perfusion values from 13 C GRE/bSSFP and Dynamic contrast-enhanced MRI (DCE-MRI) were individually analyzed through histograms and the mean perfusion values were compared with reference values obtained from PET( 15 O-water). The Myocardial Perfusion Reserve Index (MPRI) was estimated for 13 C GRE/bSSFP and DCE-MRI and compared with the reference values. RESULTS: Perfusion values, estimated by both DCE and 13 C MRI, were found to be lower than reference values. However, DCE-MRI's estimated perfusion values were closer to the reference values than those obtained from 13 C MRI. In the case of MPRI estimation, the 13 C estimated MPRI values (GRE/bSSFP: 2.3/2.0) more closely align with the literature value (around 3) than the DCE estimated MPRI value (1.6). CONCLUSION: This study demonstrated the feasibility of 3D whole-heart myocardial perfusion quantification using hyperpolarized HP001 with both GRE and bSSFP sequences. The 13 C perfusion measurements underestimated perfusion values compared to the 15 O PET literature value, while the 13 C estimated MPRI value aligned better with the literature. This preliminary result indicates 13 C imaging may more accurately estimate MPRI values compared to DCE-MRI.