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

A team of doctors and researchers have created a groundbreaking "superhuman" AI tool capable of detecting heart disease in less than two seconds. This innovative technology was trained using data from millions of ECGs and works by extracting additional information that the human eye typically misses. While the traditional ECG, which has been a vital medical tool for diagnosing heart issues for a century, cannot identify heart disease, this AI tool can spot signs of heart failure and heart valve disease.

The findings were presented at the European Society of Cardiology annual congress in Munich, the world's largest heart conference.

Early diagnosis of heart failure and heart valve disease is crucial, as it allows those in need of lifesaving medications to receive treatment sooner, preventing dangerous health complications. The AI tool has shown promising results in a trial involving 67,000 US patients, correctly identifying up to 81% of those with heart failure and up to 90% of those with heart valve disease.

However, it is important to note that this AI tool cannot definitively diagnose or rule out heart failure or heart valve disease. Instead, it provides a strong indication that someone may have the condition, which could lead to a faster referral for an echocardiogram, the standard ultrasound scan used to diagnose these heart conditions. This could lead to quicker diagnoses and earlier treatment, potentially saving and improving lives.

Prof Fu Siong Ng, a professor of cardiology at Imperial College London, expressed enthusiasm about the potential of this technology to fast-track high-risk patients for treatment. He also mentioned that the AI model could be used opportunistically to diagnose heart failure and heart valve disease in patients who may not have initially been suspected of having these conditions. The next step for the developers is to create handheld AI-led ECG readers for healthcare professionals to use.