Artificial intelligence is making strides in healthcare, with new research revealing its capability to “accurately” predict the risk of heart attack or stroke in women through mammogram scans. This groundbreaking study suggests a dual-purpose screening program that could detect two leading causes of death—heart disease and breast cancer—in women globally.
The study, published in the journal Heart, highlights a machine learning model developed by The George Institute for Global Health. This model can predict heart disease risk in women by analyzing mammograms, marking a significant advancement in medical technology.
Revolutionary Screening Approach
Developed in collaboration with the University of New South Wales and the University of Sydney in Australia, this deep learning algorithm is the first of its kind. It utilizes mammographic features and age to predict major cardiac events with accuracy comparable to traditional cardiovascular risk calculators.
Associate Professor Clare Arnott of The George Institute emphasized the necessity of new methods to identify women at risk of cardiovascular disease (CVD). She noted the common misconception that CVD predominantly affects men, leading to underdiagnosis and undertreatment in women.
“By integrating CV risk screening with breast screening through the use of mammograms—something many women already engage with at a stage in life when their cardiovascular risk increases—we can identify and potentially prevent two major causes of illness and death at the same time,” Arnott explained.
Model Validation and Comparison
The model was validated using routine mammograms from over 49,000 women in Victoria, Australia, linked to individual hospital and death records. Researchers compared the model’s performance to traditional models based on known cardiovascular risk factors such as blood pressure and cholesterol.
Arnott stated, “We found that our model performed just as well without the need for extensive clinical and medical data.” Previous research focused on certain mammographic features like breast arterial calcification (BAC), which is associated with cardiovascular risk in some populations. However, BAC alone has limitations, particularly in predicting CVD risk in older women.
“Our model is the first to use a range of features from mammographic images combined simply with age—a key advantage of this approach being that it doesn’t require additional history taking or medical record data, making it less resource-intensive to implement, but still highly accurate,” Arnott added.
Global Implications and Future Prospects
Cardiovascular disease remains the leading cause of death among women worldwide, accounting for approximately nine million deaths annually, or about one in three of all female deaths. Studies indicate that cardiovascular disease symptoms and risk factors are “under-considered” in women, resulting in fewer diagnostic tests and specialist referrals compared to men.
Meanwhile, mammography-based screening programs have successfully engaged women in several countries, with participation rates exceeding 67% in the UK and the United States. Dr. Jennifer Barraclough of The George Institute highlighted the potential of leveraging an existing risk screening process widely utilized by women to serve as a cardiovascular risk prediction tool globally.
“We hope this technology will one day provide greater and more equitable access to screening in rural areas, as many women already benefit from mobile mammography units free of charge,” Barraclough said.
She further expressed optimism about testing the model in diverse populations and understanding potential barriers to its implementation. The move represents a significant step towards more inclusive and effective healthcare solutions for women, potentially transforming how cardiovascular risks are assessed and managed worldwide.
