AI can detect heart disease in women using mammograms, study suggests

Researchers in Israel have demonstrated that routine mammograms, traditionally used to screen for breast cancer, can also be leveraged to detect cardiovascular disease in women. By applying artificial‑intelligence algorithms to 97,364 mammographic scans from 29,921 women with an average age of 54, the team identified markers of high blood pressure, coronary heart disease and prior stroke. Cross‑referencing these images with medical records revealed that 16 % of the participants had hypertension, while 2.5 % each suffered from coronary disease or had experienced a stroke. The machine‑learning model correctly flagged stroke cases 86 % of the time and identified high blood pressure and coronary disease with accuracies of 79 % and 78 % respectively, regardless of age or concurrent breast cancer status. The findings were presented at the European Society of Cardiology’s annual congress in Munich, underscoring a potential dual‑purpose use for mammography that could reach millions of women worldwide.

The significance of this breakthrough lies in addressing the chronic underdiagnosis of cardiovascular disease (CVD) among women, the leading cause of death in this population. Dr Viana Copeland of Tel Aviv University highlighted that women often present with advanced CVD only after symptoms become severe, whereas many attend regular breast‑cancer screenings even in the absence of cardiac complaints. By extracting heart‑health information from an already widespread imaging modality, the approach could provide a scalable, cost‑effective means of early risk detection without additional tests, particularly targeting midlife women when cardiovascular risk begins to rise. Experts emphasized that this could shift the paradigm of preventive care, offering a “window on to cardiovascular health” through a familiar, widely accepted procedure.

The research community is now focused on refining the AI model to improve precision, reduce false positives, and expand the range of detectable cardiac conditions. European Society of Cardiology communication committee member Elena Arbelo described the results as “compelling,” noting the need to move from experimental validation to clinical implementation. Similarly, British Heart Foundation cardiologist Dr Sonya Babu‑Narayan praised the potential to combat the myth that heart disease is a predominantly male issue, stressing that a proven dual‑purpose mammogram could enhance early detection and prevention for women. Ongoing collaborations aim to bolster the algorithm’s reliability, paving the way for integration into national breast‑screening programs and ultimately broadening the impact of existing preventive health infrastructure.

⚡ Effects Interpreter