Fig. 3

Results of A, B the signal to noise ratio (SNR) and C, D contrast to noise ratio (CNR) between different image reconstruction methods. The mean SNR and CNR were significantly higher in both the proximal and distal segments of the coronary arteries for images reconstructed with super-resolution (SR) deep learning image reconstruction (DLR) compared to those associated with other image reconstruction methods, including filtered-back projection (FBP), hybrid iterative reconstruction (IR), and DLR. ^***P < 0.001, ^**P < 0.002, ^*P < 0.033