AI Tool Detects Elevated Heart Disease Risk in Women
A groundbreaking development in health technology has emerged with a new AI model capable of identifying female patients at an elevated risk of heart disease through analysis of electrocardiograms (ECGs). This innovative approach may facilitate earlier detection, improving treatment and care for high-risk women. The research findings have been published in Lancet Digital Health.
ECGs, which monitor the heart’s electrical activity, are among the most frequently performed medical tests globally. The research, funded by the British Heart Foundation, involved the analysis of over one million ECGs from 180,000 patients, with a focus on 98,000 female patients.
The researchers developed a scoring system that compares individual ECG patterns to typical patterns associated with both sexes, revealing the associated health risks. Notably, females whose ECGs resembled ‘typical’ male patterns, characterized by an increased electrical signal size, also exhibited larger heart chambers and muscle mass, correlating with a heightened risk for cardiovascular conditions, heart failure, and heart attacks.
Historically, cardiovascular disease has been perceived as a predominantly male issue, leading to an underestimation of the risks faced by women. Recent data indicates that women are actually twice as likely to die from coronary heart disease compared to breast cancer in the UK. A consensus statement has called attention to cardiovascular disease as the leading cause of female mortality, underscoring the urgent need for enhanced diagnosis, treatment, and increased female representation in clinical research.
Dr. Arunashis Sau, the lead researcher from Imperial College London, emphasized that the complexities of cardiovascular disease in women are often overlooked. He stated, “In the clinic, we tend to categorize patients by sex without sufficient regard for their individual physiological differences. AI-enhanced ECGs provide a more nuanced view of female heart health, which can lead to better health outcomes for women at high risk of heart disease.”
Dr. Fu Siong Ng, a senior author on the study, noted that many women identified through the AI model were found to be at greater risk than the average male. If implemented widely, this model could significantly diminish the gender disparities in cardiac healthcare.
The research team recently published an additional paper on an AI-ECG risk estimation model known as AIRE, which forecasts patients’ risk of disease progression based on ECG data. Trials for AIRE are scheduled to begin in late 2025 across hospitals affiliated with Imperial College Healthcare NHS Trust.
“Far too often, women are misdiagnosed or even dismissed due to the misconception that heart disease only affects men. Even when diagnosed correctly, women often receive less effective treatment compared to men,” said Dr. Sonya Babu-Narayan from the British Heart Foundation.
This study demonstrates the potential of applying advanced AI technology to routine ECGs, which are inexpensive and widely available. By harnessing such technology, healthcare systems can better identify those at risk for future heart complications and work toward closing the gender gap that exists in cardiovascular care. However, achieving this goal calls for reforms across the entire healthcare system to ensure proper heart care for all individuals when needed.
The research received funding from the British Heart Foundation, including a BHF Clinical Research Training Fellowship for Dr. Sau and a BHF Programme Grant for Dr. Fu Siong Ng. Support was also provided by the NIHR Imperial Biomedical Research Centre, emphasizing the partnership between Imperial College Healthcare NHS Trust and Imperial College London, which was granted £95 million in 2022 to advance new treatments and diagnostics for patients.