Deep Learning Reconstruction (DLR) in Liver MRI: Comparison of Image Quality and diagnostic efficiency of LAVA sequence

Motivation: The potential of DLR to improve the quality of liver acquisition with volume acceleration (LAVA) sequence images and its impact on lesion diagnosis has not been extensively reported. Goal(s): This study aims to investigate the utility of DLR in liver MR by comparing the image quality and diagnostic efficacy of the original LAVA sequence in the venous phase with the DL-LAVA sequence. Approach: The image quality and diagnostic performance of the DL-LAVA sequence were compared with the original LAVA sequence. Results: The results revealed that the DL-LAVA sequence significantly improved the image quality of the LAVA sequence, and its diagnostic performance was superior. Impact: The results show that DLR can significantly improve the image quality of LAVA sequence, which may improve the detection and diagnosis of liver-related lesions, providing a powerful imaging basis for prevention and treatment.