Cedars-Sinai's AI beats specialist models at reading heart scam

"EchoPrime, a video-based vision-language model, analyses echocardiogram footage and generates a written report of cardiac form and function. Its findings were published in Nature (volume 650, pages 970-977) in February 2026, under the title 'Comprehensive echocardiogram evaluation with view primed vision language AI.'"

"The scale of the training is what sets EchoPrime apart. The model was trained on more than 12 million echocardiography videos paired with cardiologists' written interpretations, drawn from 275,442 studies across 108,913 patients at Cedars-Sinai. No previous AI model for echocardiography has been trained on data of that volume."

"Tested across five international health systems, EchoPrime achieved state-of-the-art performance on 23 diverse benchmarks of cardiac structure and function, outperforming both task-specific AI approaches, models trained to do one thing, like measure ejection fraction, and previous foundation models that aimed for broader capability."

EchoPrime is a vision-language AI system developed by researchers at Cedars-Sinai Medical Center and collaborating institutions that analyzes echocardiogram videos and generates written reports describing cardiac form and function. Published in Nature in February 2026, the model was trained on over 12 million echocardiography videos paired with cardiologist interpretations from 275,442 studies across 108,913 patients. This unprecedented training scale enables EchoPrime to achieve state-of-the-art performance across 23 diverse cardiac benchmarks when tested at five international health systems. The model outperforms both task-specific AI tools designed for single measurements and previous foundation models attempting broader capabilities. Designed to assist rather than replace clinicians, EchoPrime produces verbal summaries to support clinical decision-making. The code, weights, and demonstration are publicly available.

Read at TNW | Health-Tech