Automated assessment of right ventricular systolic function from coronary angiograms with video-based artificial intelligence algorithms: development, validation, comparison against humans, and prospective deployment

Right ventricular systolic function (RVSF) is a critical determinant of cardiovascular outcomes, yet assessment during coronary angiography remains challenging without prior imaging. We developed and validated DeepRV, a deep learning model predicting RVSF from routine coronary angiograms.

DeepRV, a video-based deep neural network, was developed using 8053 coronary angiography studies from 6923 patients at Montreal Heart Institute (2017–23), with RVSF determined by echocardiography. The model was externally validated at the University of California, San Francisco, and prospectively deployed during primary percutaneous coronary intervention (PCI) for ST-segment elevation myocardial infarction (STEMI). In the internal test set (n = 1586; 10.5% reduced RVSF), DeepRV achieved area under the receiver operating characteristic curve (AUROC) 0.80 [95% confidence interval (CI): 0.76–0.84], sensitivity 70.5%, specificity 78.5%, and negative predictive value 95.8%. External validation demonstrated AUROC 0.75 (95% CI: 0.72–0.77) on the UCSF dataset (n = 2247 studies; 30% reduced RVSF). Prospective deployment of DeepRV during STEMI cases at our institution (n = 82) achieved AUROC 0.83 (95% CI: 0.71–0.93) using post-PCI angiogram with a median 5.1 s inference time. In a human performance evaluation (n = 200), artificial intelligence (AI) assistance improved accuracy of identifying RVSF for cardiologists (72.1–77.6%) and medical students (43.5–64.0%). Artificial intelligence alone achieved the highest accuracy (79.5%) and sensitivity (70.0%), while cardiologists with AI achieved the highest specificity (84.6%).

DeepRV enables automated RVSF assessment from routine coronary angiograms and enhances diagnostic accuracy across experience levels. Real-time inference and open-weight availability support its potential as a point-of-care tool for risk stratification during coronary angiography.