AI Model Can Read ECGs to Identify Female Patients at Higher Risk of Heart Disease

A new AI model can identify female patients who are at higher risk of heart disease based on an electrocardiogram (ECG). The researchers say that the algorithm, developed specifically for female patients, could enable doctors to identify high-risk women earlier, allowing for better treatment and care. Details were published in Lancet Digital Health.

Women Are at Very High Risk of Developing Heart Disease

An ECG records the electrical activity of the heart and is one of the most common medical tests worldwide. In their study, funded by the British Heart Foundation, the researchers used artificial intelligence to analyze over a million ECGs from 180,000 patients, 98,000 of whom were female. In the latest study, the researchers developed a score that measures how closely an individual’s ECG matches the “typical” ECG patterns for men and women, and that indicates a risk range for each gender. Women whose ECGs were more in line with the typical “male” pattern – such as an increased size of the electrical signal – tended to have larger heart chambers and more muscle mass.

Crucially, these women were also found to have significantly higher risks of cardiovascular disease, future heart failure and heart attacks compared to women whose ECGs were more in line with the “typical female” ECG. Previous findings have shown that men tend to have a higher risk of heart disease – more specifically, cardiovascular disease – which may be due to differences in hormone profiles and lifestyle factors. For this reason, healthcare professionals and the public assume that the risk of cardiovascular disease in women is low. However, this is not the case, because women are at high risk as well: in the UK, women are twice as likely to die from coronary heart disease, the leading cause of heart attack, than to die from breast cancer.

Gender Disparities in Cardiac Care Should be Reduced

A recent consensus statement calls cardiovascular disease the “number one killer” of women, calling for better diagnosis and treatment for women, and for women to be better represented in clinical trials. Dr. Arunashis Sau, an academic clinical lecturer at Imperial College London’s National Heart and Lung Institute and a cardiology registrar at Imperial College Healthcare NHS Trust, led the research. He said: “Our work has shown that cardiovascular disease in women is far more complex than previously thought. In the clinic, we use tests like ECGs to provide a snapshot of what’s going on, but this can result in patients being grouped by gender without taking into account their individual physiology.

The AI-enhanced ECGs provide a more sophisticated understanding of women’s heart health – and the researchers believe this could be used to improve outcomes for women at risk of heart disease.” Dr. Fu Siong Ng, a lecturer in cardiac electrophysiology at the National Heart & Lung Institute at Imperial College London and a consultant cardiologist at Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust, was the lead author of the study. “Many of the women identified were at even higher risk than the ‘average’ man. If widely adopted, the AI model could, over time, reduce gender disparities in cardiology care and improve outcomes for women at risk of heart disease,” said Fu Siong Ng.

According to Dr. Sonya Babu-Narayan, Clinical Director at the British Heart Foundation, women are far too often misdiagnosed or even dismissed by healthcare professionals because they believe the myth that heart disease is ‘only a man’s problem’. Even when they do receive the correct diagnosis, women are demonstrably less likely to receive the recommended treatments than men. This study applied powerful AI technology to EKGs, a routine, inexpensive and widely available cardiac test. Realizing the potential of this type of research could help to better identify those patients at highest risk for future heart problems and reduce the gender gap in cardiac care outcomes. However, a test alone will not ensure equal opportunity. To ensure that everyone gets the right cardiac care when they need it, changes are needed across our healthcare system.

According to the experts, many of the women identified were at even higher risk than the ‘average’ man. If widely adopted, the AI model could, over time, reduce gender disparities in cardiac care and improve outcomes for women at risk of heart disease. The research group recently published another article on the related AI ECG risk estimation model AIRE, which can predict the risk of patients developing a disease and worsening based on an ECG. Trials of AIRE in the NHS are already planned for late 2025. These will evaluate the benefits of implementing the model with real patients from hospitals in the Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust. This model will be tested in conjunction with AIRE.