Accelerating 2D Kidney Magnetic Resonance Fingerprinting using Deep Learning–Based Tissue Quantification

In this pilot study, we developed and optimized a spatially-constrained convolutional neural network to accelerate the scan time for 2D kidney Magnetic Resonance Fingerprinting (MRF). Our results suggest that an acceleration factor of 3 can be achieved with the proposed method, which shortens the 2D breath-hold MRF scan from 15 sec to 5 sec. In addition, the deep learning based approach can be applied for T1 and T2 quantification of both normal renal tissues and pathologies including renal cell carcinoma.