A cutting-edge AI-ML model is turning routine mammograms into a powerful tool to predict women’s risk of heart disease. Developed by The George Institute for Global Health in collaboration with the University of New South Wales and University of Sydney, the model uses mammographic images and age alone to forecast major cardiac events, matching the accuracy of traditional risk calculators that rely on multiple clinical measurements.
Globally, Cardiovascular disease (CVDs) is the leading cause of death in women, responsible for around 9 million deaths each year. Yet, Cardiovascular risk is quite under-assessed in females.
Trained on 49,000 mammograms from women in metropolitan and rural Victoria, Australia, and linked to hospital and mortality records, the AI model identifies subtle patterns that may escape human observation. Yet, many women are underdiagnosed and undertreated, as heart disease is often mistakenly seen as a male-dominated condition.
The new model could change that. By integrating cardiovascular risk screening into mammography- a test millions of women already undergo, the system offers a “two-for-one” approach, detecting both breast cancer and heart disease risk simultaneously. Unlike earlier studies that relied primarily on breast arterial calcification, this AI tool analyzes a broader range of mammographic features, making it more accurate across all age groups and easier to implement without additional patient data.
This AI model, trained on mammograms and patients age, matched the accuracy of top-tier heart risk calculators like New Zealand’s ‘PREDICT’ tool and the American Heart Association ‘PREVENT’ equations with a c-index of 0.72. All without needing blood tests or clinical records.
Experts say the approach could revolutionize preventive care, especially in rural areas where mobile mammography units already reach underserved communities. Early identification of high-risk women could allow timely interventions, from lifestyle changes to medication, potentially saving thousands of lives. The team plans to test the model in diverse populations worldwide and explore practical ways to integrate it into screening programs. By turning a routine imaging test into a predictive health tool, this innovation may bring women’s cardiovascular care into a new era being proactive, precise, and life-saving.
