Life Sciences & Medicine Active Updated Oct 10, 2026
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.
The story so far
-
Oct 10, 2026 Latest
The NIH reported that a deep learning model reading ECGs already recorded during sleep studies, trained on 15,809 patients and tested on 9,810 and 12,576 more, sorted people by long-term risk of atrial fibrillation, heart failure and death.
-
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.
This story moves with the morning letter.
When this question gets a new answer, the page above changes the same morning. Each morning's letter brings the day's progress in science, medicine and beyond. Free, every day.
Free forever, no spam. One click to unsubscribe, and we never sell your email.