Key takeaways
- Cardiovascular diseases are often underdiagnosed or diagnosed late in women.
- An AI model was able to identify hypertension, ischaemic heart disease and stroke from mammograms in a retrospective study.
- In the future, routine mammograms could help flag women who may benefit from a cardiovascular assessment.
Munich, Germany – 27 August 2026: Artificial intelligence (AI) analysis of mammograms could be used to detect different common cardiovascular diseases (CVDs), according to a study that will be presented at ESC Congress 2026.[1]
Presenter, Doctor Viana Copeland from Chaim Sheba Medical Center, Tel Aviv University, Ramat Gan, Israel, explained why new detection methods are needed for CVD in women: “Despite being the leading cause of death in women worldwide, CVD is consistently underdiagnosed and undertreated. A common finding in our medical centre, and around the world, is that when women do seek medical help, their CVD is already advanced. On the other hand, many women do attend routine breast cancer screening, even when they haven't sought care for cardiovascular symptoms. We investigated whether AI could help mammography serve an additional purpose in this group – the early detection of CVD – enabling preventive strategies to be implemented.”
This retrospective cohort study involved data from 29,921 women who underwent 97,364 mammography examinations. The cohort had a median age of 54 years. Clinical information on the presence of three common CVDs – hypertension, ischaemic heart disease (also known as coronary artery disease) and stroke – was extracted from various sources including electronic medical records, medication prescriptions, and procedural and imaging findings. The prevalence was 16% for hypertension, 2.5% for ischaemic heart disease and 2.5% for stroke.
A deep learning model was trained to identify features from the mammograms of women who had hypertension, ischaemic heart disease or stroke. The model’s ability to distinguish between women with and without each cardiovascular condition was evaluated using areas under the receiver operating characteristic curves (AUROC), where values range from 0.5 for random guessing to 1.0 for perfect discrimination.
The initial model performed well, yielding AUROCs of 0.79 for hypertension, 0.78 for ischaemic heart disease and 0.86 for stroke. The results were consistent when considering cancer status and age.
Doctor Copeland noted, “Because mammography is already widely used, analysing the same images for cardiovascular information could potentially offer a scalable approach without requiring an additional imaging examination. Mammography also reaches many women in midlife, an important period for recognising and addressing cardiovascular risk.”
The researchers are now working to improve the model’s accuracy and reduce both false positives and false negatives. They also plan to investigate whether mammograms could help identify other cardiovascular conditions.
Commenting on the findings, Associate Professor Elena Arbelo, Member of the ESC Communication Committee, said: “As both a cardiologist and a woman, I find this concept compelling: a mammogram may one day do more than look for breast cancer − it may also offer a window onto cardiovascular health. That matters because CVD in women is still too often recognised late. It is great to see innovative AI studies being presented at ESC Congress 2026, aiming to address unmet needs. The challenge now is to establish accuracy and reliability − to move from experimentation to clinical implementation.”
ENDS
