High-Resolution Deep Learning Reconstruction (HR-DLR) with Compressed Sensing for a Highly Accelerated Acquisition

Motivation: While Compressed Sensing (CS) is known as a powerful method to accelerate an acquisition, there still is an inevitable trade-off between image quality and scan time. A recently proposed High-Resolution Deep Learning Reconstruction (HR-DLR) may bring a breakthrough to the limitation. Goal(s): Examine potential of the combination of HR-DLR and CS for highly accelerated acquisitions. Approach: HR-DLR was employed to CS acquisitions with various acceleration factors and regularization strengths. The resulting image quality was examined in terms of SNR and sharpness. Results: HR-DLR improved both SNR and sharpness of CS images, allowing acceleration beyond that which was previously limited by typical trade-offs. Impact: High-Resolution Deep Learning Reconstruction (HR-DLR) re-defined the trade-off between scan time, image SNR and sharpness that previously limited Compressed Sensing (CS) acceleration. Combination of HR-DLR and CS makes yet higher acceleration practical, benefiting throughput and patient comfort.