‘Superhuman’ AI tool spots heart disease in less than 2 seconds

THE NEWS

What happened

A new AI tool developed by Imperial College London and the British Heart Foundation can detect signs of heart failure and valve disease from routine ECGs in under two seconds. Tested on 67,000 US patients, it identified up to 81% of heart failure cases and 90% of valve disease cases, potentially bypassing months-long waiting lists for echocardiograms.

CONTEXT

Why it matters

Doctors have developed a 'superhuman' AI that spots heart disease in less than two seconds using standard ECGs. In a US trial with 67,000 patients, it identified up to 81% of heart failure cases and 90% of valve disease cases. This breakthrough could fast-track high-risk patients for urgent echocardiograms, bypassing months-long waiting lists. While not a definitive diagnostic tool on its own, it offers a powerful early warning system, enabling earlier treatment that saves lives. Researchers are now working on handheld devices to bring this technology directly to healthcare professionals.

AT A GLANCE

Key facts

• The AI analyzes routine electrocardiograms (ECGs) to detect heart failure and valve disease, conditions previously invisible on standard ECGs.
• In a trial involving 67,000 patients in the US, the tool identified up to 81% of heart failure cases and up to 90% of heart valve disease cases.
• The technology was presented at the European Society of Cardiology annual congress in Munich.
• It aims to fast-track high-risk patients for echocardiograms, reducing wait times from months to days.
• The tool cannot definitively diagnose conditions on its own but serves as a powerful screening indicator.
• Researchers are developing handheld AI-led ECG readers for healthcare professionals.
• Similar AI models can also opportunistically detect high blood pressure and type 2 diabetes from facial videos.