AI reading heart disease from routine ECGs

Where it stands as of Sep 22, 2026

An AI model trained on millions of patients read ordinary ECGs from 67,000 people and correctly flagged 81 percent of heart failure cases and 90 percent of heart valve disease cases, catching conditions that normally hide until a cardiac arrest reveals them. The model has been validated on a large retrospective dataset. It has not yet been shown to change patient outcomes in prospective clinical use.