AI tool detects heart disease from ECG in under two seconds

Doctors have developed a “superhuman” AI tool that can spot heart disease in less than two seconds.

The groundbreaking technology has been trained on millions of patients and works by extracting more information from a routine electrocardiogram (ECG) than the human eye can typically see.

The traditional ECG, which records electrical activity in the heart, including the rate and rhythm, has been a vital medical tool in diagnosing heart attacks and abnormal heart rhythms for a century.

But it cannot detect heart disease. That requires an echocardiogram, a type of ultrasound scan, which patients often have to wait months for.

Now a team have developed an AI tool that can spot signs of heart failure and heart valve disease – two of the most common forms of heart disease – from ECG results in “the blink of an eye”.

Details of the breakthrough, which could boost early diagnosis of heart disease, were presented to thousands of delegates at the European Society of Cardiology annual congress in Munich, the world’s largest heart conference.

Early diagnosis is vital for heart failure and heart valve disease, enabling those who need lifesaving medicines to be spotted sooner, before they become dangerously unwell.

The development is being seen as potentially significant, because ECGs are one of the most common tests in medicine, with about a billion performed worldwide each year.

In a trial involving 67,000 patients in the US, the AI tool was able to identify up to 81% of those who had heart failure, and up to 90% of those with heart valve disease.

Dr Sonya Babu-Narayan, a consultant cardiologist and clinical director of the British Heart Foundation (BHF), which funded the trial, said: “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.

“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.”

The tech cannot be used on its own to definitively diagnose or rule out heart failure or heart valve disease, but it gives a very strong indication someone may have them.

Someone judged as highly likely to have either could be sent rapidly for an echocardiogram, rather than wait months on the standard waiting lists. That could mean a quicker diagnosis which would enable them to start treatment earlier.

Prof Fu Siong Ng, a professor of cardiology at Imperial College London, said: “Patients can often wait several months for a heart ultrasound scan after being referred for one by their doctor.

“This makes it exciting that our technology could identify patients most at risk of heart failure and heart valve disease, so they could be prioritised for scans faster and more urgently.”

The aim of the tool is to speed up diagnosis of those suspected to have heart failure or heart valve disease. But Ng said it could also prove lifesaving by spotting signs of the conditions in people who may have undergone an ECG for different reasons.

“Another potential application of this AI model is to opportunistically diagnose heart failure and heart valve disease in whom these conditions are not suspected,” he said. “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.”

Dr Ahmed El-Medany, a BHF clinical research fellow, who led the Imperial College London analysis, described the tool as a “superhuman AI” and said the next challenge would be to design handheld AI-led ECG readers for healthcare professionals to use.

Delegates in Munich also heard how AI-based analysis of facial videos could rapidly and accurately detect undiagnosed high blood pressure and type 2 diabetes.

Researchers at the University of Tokyo and the Institute of Science Tokyo said AI analysis of five-second facial videos could hep improve diagnosis. Millions of people who have high blood pressure or type 2 diabetes do not know they have the conditions.