HYDRA: Hybrid Deep Magnetic Resonance Fingerprinting

Magnetic resonance fingerprinting (MRF) methods typically rely on dictionary matching to map the temporal MRF signals to quantitative tissue parameters. Such approaches suffer from inherent discretization errors, as well as high computational complexity as the dictionary size grows. To alleviate these issues, we propose a HYbrid Deep magnetic ResonAnce fingerprinting approach, referred to as HYDRA, which involves two stages: a model-based signature restoration phase and a learning-based parameter restoration phase. The former phase is implemented using low-rank based de-aliasing techniques and the latter phase is implemented using a deep residual convolutional neural network (MRF-ResNet). The designed network is trained on synthesized MRF data simulated with the Bloch equations and fast imaging with steady state precession (FISP) sequences, and then it takes a temporal MRF signal as input and produces the corresponding tissue parameters. In contrast to conventional dictionary-matching-based MRF approaches, our approach significantly improves the inference speed by eliminating the time-consuming dictionary matching operation, and alleviates discretization errors by outputting continuous-valued parameters. We further avoid the need to store a large dictionary. We validated our MRF-ResNet on both synthetic data and phantom data generated from a healthy subject, demonstrating improved performance over state-of-the-art MRF approaches in terms of inference speed, reconstruction accuracy, and storage requirements.