Considering low-rank, sparse and gas-inflow effects constraints for accelerated pulmonary dynamic hyperpolarized Xe MRI.

Dynamic hyperpolarized (HP) Xe MRI is able to visualize the process of lung ventilation, which potentially provides unique information about lung physiology and pathophysiology. However, the longitudinal magnetization of HP Xe is nonrenewable, making it difficult to achieve high image quality while maintaining high temporal-spatial resolution in the pulmonary dynamic MRI. In this paper, we propose a new accelerated dynamic HP Xe MRI scheme incorporating the low-rank, sparse and gas-inflow effects (L + S + G) constraints. According to the gas-inflow effects of HP gas during the lung inspiratory process, a variable-flip-angle (VFA) strategy is designed to compensate for the rapid attenuation of the magnetization. After undersampling k-space data, an effective reconstruction algorithm considering the low-rank, sparse and gas-inflow effects constraints is developed to reconstruct dynamic MR images. In this way, the temporal and spatial resolution of dynamic MR images is improved and the artifacts are lessened. Simulation and in vivo experiments implemented on the phantom and healthy volunteers demonstrate that the proposed method is not only feasible and effective to compensate for the decay of the magnetization, but also has a significant improvement compared with the conventional reconstruction algorithms (P-values are less than 0.05). This confirms the superior performance of the proposed designs and their ability to maintain high quality and temporal-spatial resolution.