Deep learning analysis in coronary computed tomographic angiography imaging for the assessment of patients with coronary artery stenosis

•CCTA reduces time to reconstruct images by 80% comparing with traditional image reconstruction and time to diagnose coronary artery disease by 80% comparing with traditional CCTA.•In identifying ≥50% stenotic vessels, the diagnostic accuracy of CCTA-AI is better than traditional CCTA with DSA as the reference method (AUCAI = 0.870, AUCCCTA= 0.781, p < 0.001).•In the aspect of identifying plaque classification, accuracy of CCTA-AI is moderate compared to traditional CCTA (AUC = 0.750, p < 0.001) with traditional CCTA regarded as the reference method.