A new artificial intelligence (AI) model designed specifically for women could improve the early detection of heart disease by analyzing electrocardiograms (ECGs). Researchers behind the project say this innovation could lead to earlier interventions, better treatment, and improved outcomes for female patients. The findings were published today in Lancet Digital Health.
The Gender Gap in Heart Disease Diagnosis
Heart disease, or cardiovascular disease (CVD), has long been considered a greater risk for men. However, women are twice as likely to die from coronary heart disease—the leading cause of heart attacks—than from breast cancer in the UK. Despite this, many women remain underdiagnosed and undertreated due to outdated assumptions about their risk levels.
To address this issue, researchers from Imperial College London analyzed over one million ECGs from 180,000 patients, including 98,000 women, using AI. Their goal was to develop a model that could better assess heart disease risk specifically in female patients.
AI-Enhanced ECGs: A More Personalized Approach for Heart Disease
ECGs measure the electrical activity of the heart and are one of the most common diagnostic tools in medicine. However, traditional ECG interpretations often group patients by sex without considering individual physiological differences.
In this study, the AI model developed a score based on how closely a woman’s ECG matched a typical “male” or “female” pattern. Women whose ECGs resembled a male pattern—characterized by increased signal size—were found to have larger heart chambers, more muscle mass, and, critically, a significantly higher risk of heart disease, heart failure, and heart attacks.
“Our work has underlined that cardiovascular disease in females is far more complex than previously thought,” said Dr. Arunashis Sau, the study’s lead researcher. “The AI-enhanced ECGs give us a more nuanced understanding of female heart health and could be used to improve outcomes for women at risk of heart disease.”
A Step Toward Reducing Gender Disparities in Cardiac Care
One of the study’s most striking findings was that many of the high-risk women identified by the AI model were actually at greater risk than the “average” man.
“If this AI model is widely implemented, it could help close the gender gap in cardiac care and improve outcomes for women,” said Dr. Fu Siong Ng, senior author of the study.
The research team is also behind another AI-ECG risk estimation model, AIRE, which predicts the likelihood of developing or worsening heart disease. Trials for AIRE in the NHS are planned for late 2025, and researchers hope to integrate this new female-specific model alongside it.
The Future of AI in Women’s Heart Health
Experts believe this study could be a turning point for women’s heart care. Dr. Sonya Babu-Narayan, Clinical Director at the British Heart Foundation, emphasized the importance of these findings:
“Far too often, women are misdiagnosed or dismissed by healthcare professionals due to the myth that heart disease is ‘only a male issue.’ This study applies powerful AI technology to a routine, inexpensive heart test, which could help better identify women at highest risk and reduce the gender gap in heart care outcomes.”
While AI tools like this one show promise, experts caution that broader systemic changes are needed to ensure all patients receive timely and accurate heart care. The study was funded by the British Heart Foundation and supported by the NIHR Imperial Biomedical Research Centre, which continues to develop experimental treatments and diagnostics.
If adopted widely, AI-enhanced ECGs could become a key tool in fighting heart disease in women—providing earlier diagnoses, targeted interventions, and ultimately, saving lives.
Reference: Arunashis Sau, Ewa Sieliwonczyk, Konstantinos Patlatzoglou, Libor Pastika, Kathryn A McGurk, Antônio H Ribeiro, Antonio Luiz P Ribeiro, Jennifer E Ho, Nicholas S Peters, James S Ware, Upasana Tayal, Daniel B Kramer, Jonathan W Waks, Fu Siong Ng. Artificial intelligence-enhanced electrocardiography for the identification of a sex-related cardiovascular risk continuum: a retrospective cohort study. The Lancet Digital Health, 2025.
