AI spots heart disease risk from routine ECGs

AI is being used to turn routine electrocardiogram tests into a faster screening tool for serious heart disease. ECGs are performed roughly a billion times a year worldwide and already help doctors detect rhythm problems and some heart attack signs, but they have not reliably identified structural problems such as heart failure or leaky or narrowed valves, which usually require echocardiograms.

At the European Society of Cardiology’s annual congress in Munich at the end of August, Dr. Ahmed El-Medany of Imperial College London presented models trained on millions of hospital ECG recordings and tested on tens of thousands of patients in the United States. The AI correctly flagged up to four in five people with heart failure and up to 90 per cent of those with significant valve disease.

The system is intended to prioritize patients for ultrasound, not replace it. High-risk patients could move ahead in echocardiogram queues, while hospitals could also run the models across ECGs ordered for other reasons. Imperial is testing the approach in a smaller prospective NHS study in London and Bristol, with future work focused on portable ECG devices and broader clinical deployment through Cardiovolt.ai.