AIMS Most existing machine learning methods for coronary angiography analysis are limited to a single aspect. We aimed to achieve an automatic and multimodal analysis to recognize and quantify coronary angiography, integrating multiple aspects, including the identification of coronary artery segments and the recognition of lesion morphology.
METHODS AND RESULTS A dataset of 20,612 angiograms was retrospectively collected: among which 13,373 angiograms were labeled with coronary artery segments, and 7,239 were labeled with special lesion morphology. Trained and optimized by these labeled data, one network recognized 20 different segments of coronary arteries, while the other detected lesion morphology, including measures of lesion diameter stenosis as well as calcification, thrombosis, total occlusion, and dissection detections in an input angiogram. For segment prediction, the recognition accuracy was 98.4%, and the recognition sensitivity was 85.2%. For detecting lesion morphologies including stenotic lesion, total occlusion, calcification, thrombosis, and dissection, the F1-scores were 0.829, 0.810, 0.802, 0.823, and 0.854 respectively. Only 2 seconds were needed for the automatic recognition.
CONCLUSIONS Our deep learning architecture automatically provides a coronary diagnostic map by integrating multiple aspects, which helps cardiologists to flag and diagnose lesion severity and morphology during the intervention..