A study by teams at Imperial College London and Imperial College Healthcare, and funded by the British Heart Foundation, has developed an artificial intelligence-enhanced electrocardiography (AI-ECG) model with the potential to identify female patients “who could benefit from enhanced risk factor modification or surveillance” in relation to elevated cardiovascular risk.
Published in The Lancet Digital Health, the project involved a team of researchers using AI to analyse more than one million ECGs from 180,000 patients, including 98,000 females. The team then developed a score measuring “how closely an individual’s ECG matches ‘typical’ patterns of ECGs for men and women, and which showed a range of risk for each sex”.
Findings included that those women whose ECGs were a closer match with the typical “male” pattern, “such as having an increased size of the electrical signal”, tended to have “larger heart chambers and more muscle mass”. Amongst these women, researchers identified a “significantly higher risk” of cardiovascular disease, heart failure, and heart attacks.
Arunashis Sau of the Imperial College London National Heart and Lung Institute led on the research, observing that it highlighted how “cardiovascular disease in females is far more complex than previously thought”. Simply grouping patients by sex “doesn’t take into account their individual physiology”, Sau reported, whereas the AI-ECG offers “a more nuanced understanding of female heart health” that could be used to “improve outcomes for women at risk of heart disease”.
An earlier study conducted by the research group developed an AI-ECG risk estimation model (AIRE) designed to use ECGs to predict risk of developing and worsening disease, with plans to trial the model with patients from hospitals at Imperial College Healthcare NHS Trust and Chelsea and Westminster Hospital NHS Foundation Trust, scheduled for “late 2025”.
Citation: Sau, A., Sieliwonczyk, E., Patlatzoglou, K., Pastika, L., McGurk, K., Ribeiro, A., Ribeiro, A. L., Ho, J., Peters, N., Ware, J., Tayal, U., Kramer, D., Waks, J., Ng, F.. Artificial intelligence-enhanced electrocardiography for the identification of a sex-related cardiovascular risk continuum: a retrospective cohort study (2025) 7(3) The Lancet Digital Health, E184 – E194. DOI: 10.1016/j.landig.2024.12.003
Innovation at Imperial College Healthcare and Imperial College London
In December, Imperial College Healthcare NHS Trust published its framework and approach to future-proofing the trust’s adoption and utilisation of AI. The framework outlines the aim of improving care for patients “from diagnosis to treatment to the administrative functions that keep our hospitals moving”; adding that it will be “vital” to ensure that AI technologies and usage remains “accessible to all”.
A research team at Imperial College Healthcare and staff from Imperial College London also recently published findings from an AI analysis, noting that AI could help guide the optimum time to collect eggs for IVF success.
