AI echocardiograms flagged up to 90% of valve disease in 67,000-patient trial

Patients suspected of heart disease often wait months for echocardiograms, delaying diagnosis of heart failure and heart valve disease.

A team has developed an AI tool that can spot signs of heart failure and heart valve disease from routine ECGs in less than two seconds. The technology has been trained on millions of patients.

The work was presented to thousands of delegates at the European Society of Cardiology annual congress in Munich. In a trial involving 67,000 patients in the US, the AI identified up to 90% of those with heart valve disease and up to 81% of those with heart failure.

It works by extracting more information from a routine ECG than the human eye can typically see, so someone judged as highly likely could be sent rapidly for an echocardiogram rather than wait months.

The tool cannot be used on its own to definitively diagnose or rule out these conditions and will not detect everyone. It could, however, be used to help fast-track the patients who are most likely to have a heart abnormality.