'Superhuman' AI Spots Heart Disease in Under 2 Seconds From a Routine ECG

Doctors have developed a "superhuman" AI tool that can spot signs of heart disease in less than two seconds, extracting more information from a routine ECG than the human eye can typically see, The Guardian reports.

The technology, trained on millions of patients, identifies indications of heart failure and heart valve disease — two of the most common forms of heart disease — directly from electrocardiogram results. Details of the breakthrough, reported by The Guardian amid the latest AI developments, were presented to thousands of delegates at the European Society of Cardiology annual congress in Munich, the world's largest heart conference.

Why a Century-Old Test Needed Help

The electrocardiogram has been a vital medical tool for a century, recording the heart's electrical activity — its rate and rhythm — and helping diagnose heart attacks and abnormal heart rhythms. But the traditional ECG cannot detect heart disease itself. That requires an echocardiogram, an ultrasound scan of the heart, for which patients often wait months.

That gap is what the AI tool is designed to close. It reads the subtle patterns in an ECG that correlate with structural heart problems and flags patients who are most likely to have them, in what cardiologists involved described as "the blink of an eye."

The potential is enormous simply because of scale: roughly a billion ECGs are performed worldwide every year, making the test one of the most common in all of medicine. If the software can squeeze diagnostic information out of a test that is already being done anyway, it could reach patients who would never otherwise get scanned in time.

What the 67,000-Patient Trial Found

In a trial involving 67,000 patients in the United States, funded by the British Heart Foundation (BHF), the AI tool identified up to 81 percent of those who had heart failure, and up to 90 percent of those with heart valve disease.

Importantly, the tool is not a standalone diagnostic. It cannot definitively confirm or rule out either condition on its own. What it does is give a very strong indication — enough to fast-track the patients most likely to have a heart abnormality directly to an echocardiogram, rather than leaving them on standard waiting lists for months.

"It is exciting to see that AI can now deliver a read-out from an ECG in what feels like the blink of an eye," said Dr Sonya Babu-Narayan, a consultant cardiologist and clinical director at the British Heart Foundation. "Technology like the AI ECG in this research, which has the potential to identify high-risk patients early, will not detect everyone with a heart condition. But it could be a solution to help fast-track the patients who are most likely to have a heart abnormality. When it comes to the heart, earlier diagnosis and treatment saves and improves lives."

Catching the Diseases Nobody Was Looking For

The most intriguing application may be the one researchers call opportunistic screening. Because ECGs are performed constantly — for palpitations, before surgery, as part of routine checkups — the model could run silently on every ECG a hospital produces, flagging patients whose heart failure or valve disease was never suspected.

"Another potential application of this AI model is to opportunistically diagnose heart failure and heart valve disease in whom these conditions are not suspected," said Prof Fu Siong Ng, a professor of cardiology at Imperial College London. "The AI model could be run on all ECGs done in a hospital to flag those at highest risk of these diseases, so that they can be diagnosed earlier."

Ng noted that patients can often wait several months for a heart ultrasound scan after being referred by their doctor, and said the technology's real value is prioritization: identifying who genuinely needs that scan urgently.

Dr Ahmed El-Medany, a BHF clinical research fellow at Imperial College London who led the analysis, described the tool as "superhuman AI" — and set out the next challenge: designing handheld AI-led ECG readers that healthcare professionals could use outside the hospital, widening access far beyond cardiology departments.

Early diagnosis matters because both conditions become dramatically more dangerous with delay. Heart failure and valve disease can be managed with lifesaving medicines and procedures — but only if patients are identified before they become dangerously unwell.

AI Read Faces Next

The Munich congress also heard how AI could extract diagnostic signals from something even simpler than an ECG: a short video of a person's face. Researchers at the University of Tokyo and the Institute of Science Tokyo presented an AI analysis of five-second facial videos that could rapidly and accurately detect undiagnosed high blood pressure and type 2 diabetes — conditions that millions of people have without knowing it.

Together, the two lines of research point to a common theme gathering pace across medical AI: turning trivial, low-cost measurements into screening tools that catch serious disease earlier. Follow the latest AI developments as health systems weigh how, and how fast, to deploy them.

Stay Ahead of AI

Medical AI is advancing quickly — and so is everything else in the field. Read more AI news on AI Buzz Wire, your source for independent coverage of artificial intelligence.

Get the latest AI research news and see how machine learning is transforming medicine, science, and beyond.