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

Doctors have created an "AI superhuman" tool that can detect heart disease in less than two seconds. The innovative technology was trained on millions of ECGs and extracts more information from routine electrocardiograms than the human eye typically sees. While traditional ECGs have been used for a century to diagnose heart attacks and abnormal heart rhythms, they cannot identify heart disease.

The AI tool can detect signs of heart failure and heart valve disease, two common forms of heart disease, from ECG results in mere seconds. This breakthrough was presented at the European Society of Cardiology annual congress in Munich, the world's largest heart conference, and could greatly improve early diagnosis of heart disease.

Early detection allows patients with high-risk heart conditions to receive lifesaving medicines before their symptoms become severe. The AI tool was tested on 67,000 patients, identifying up to 81% with heart failure and up to 90% with heart valve disease. While the AI cannot definitively diagnose heart failure or heart valve disease, it provides a strong indication that further echocardiogram testing may be necessary.

The tool's primary goal is to expedite the diagnosis of patients suspected of having heart failure or heart valve disease, potentially speeding up their treatment. However, the technology could also be used opportunistically to diagnose heart conditions in individuals who may have undergone an ECG for unrelated reasons. The next challenge for researchers is to develop handheld AI-led ECG readers for healthcare professionals to easily incorporate into their practice.