An artificial intelligence (AI) tool to help doctors identify high-risk heart disease patients is soon to be put into trial use in England. A study found that it can accurately predict the risk of death.
The international research team, led by Imperial College London, has trained its artificial intelligence model, known as AI-ECG risk estimation, or AIRE, on millions of results from electrocardiograms (ECGs), a common medical test that records the electrical signals in and between the heart chambers. This method is commonly used to diagnose heart disease and other abnormalities.
The goal is to identify nuanced patterns that may mean someone is at high risk of health problems or death.
In the trial, the model predicted the likelihood of death in the decade after the EKG - and was correct 78% of the time.
"We think this could have big benefits for the NHS and globally," said Dr Fu Siong Ng, a cardiac electrophysiology researcher at Imperial College London who worked on the project.
The system can also predict heart attacks, heart failure and heart rhythm problems, and the researchers said it could be rolled out across Britain's National Health Service (NHS) within the next five years.
Trials with real patients are already planned for several sites in London and are expected to start by mid-2025, Euronews reported.
They will evaluate the benefits of the model, using patients from outpatient clinics and hospital medical wards.
Potential of artificial intelligence to improve heart health
AI-powered EKGs are already used to diagnose heart disease, but they are not part of routine medical care and have not yet been used to determine an individual patient's risk levels.
"This could take the use of ECGs beyond what has been possible by helping to assess the risk of future heart and health problems, as well as the risk of death," explained Brian Williams, chief scientific and medical director of the British Heart Foundation. which funded the study.
The researchers, who published their results in the Lancet Digital Health journal, said the predictions where the AI got it wrong could be due to other unknown factors, such as whether the patient received additional treatment or died unexpectedly. The model can still pick up subtle changes in the heart's structure that could serve as a warning sign of disease or risk of death that doctors might miss.
"We cardiologists use our experience and standard guidelines when looking at ECGs, dividing them into 'normal' and 'abnormal' patterns to help us diagnose disease," said Dr Arunashis Sau, an academic physician at Imperial college in London, who led the new research.
"However, the AI model detects much finer details, so it can 'spot' problems in the EKG that appear normal to us, potentially long before the disease is fully developed," explained Sau.
More research is needed in hospitals and other healthcare settings to determine the model's future role in diagnosis and treatment. Patients with other health problems could probably benefit from it as diseases like diabetes can also affect the heart. | BGNES
.jpeg)
