AI-Powered Mammograms May Detect Heart Disease Risk in Women
- Routine screening mammograms can help clinicians identify early signs of heart disease in women by detecting calcium deposits in breast tissue.
- Researchers found that an artificial intelligence model measuring breast arterial calcification could flag patients at a significantly higher risk of cardiovascular events and death.
- Heart disease remains the leading cause of death among women, accounting for one in three deaths.
Routine screening mammograms can help clinicians identify early signs of heart disease in women by detecting calcium deposits in breast tissue. That is the conclusion of a large observational study published in the European Heart Journal and discussed by the Mayo Clinic.
Routine Mammograms Reveal Hidden Cardiovascular Risks
Researchers found that an artificial intelligence model measuring breast arterial calcification could flag patients at a significantly higher risk of cardiovascular events and death. Crucially, this detection operates independently of traditional health metrics like cholesterol and blood pressure.
The Scale of the Problem in Women’s Health
Heart disease remains the leading cause of death among women, accounting for one in three deaths. Yet, medicine lacks a standard for annual heart disease screening tailored specifically to how cardiovascular disease develops in female blood vessels.
According to the Mayo Clinic, routine mammograms are widely utilized. Average-risk women begin annual screenings at age 40, offering a built-in opportunity to check vascular health without requiring extra radiation or testing.
Detecting Subtle Vascular Changes With AI
Breast arterial calcification occurs when calcium builds up in the walls of the arteries inside breast tissue, making blood vessels stiffer and less flexible. While this calcium differs from calcium in the heart, its presence signals a higher risk of heart problems because it alters how blood moves.
Dr. Imon Banerjee, scientific director of the Arizona Advanced AI and Innovation Hub at Mayo Clinic, explains that these calcium deposits appear as very subtle findings on standard two-dimensional mammograms. They cannot be consistently quantified by eye, even by experts.
To overcome this diagnostic hurdle, researchers developed a deep learning model that identifies calcium deposits, measures their extent, and classifies severity. According to Mayo Clinic findings, the algorithm allows clinicians to measure breast arterial calcification with a single click, integrating the data directly into the radiologist report.
Tracking 120,000 Patients Across 12 Institutions
The retrospective cohort study evaluated more than 120,000 women who underwent screening mammograms across two healthcare systems. Researchers tested whether AI-derived breast arterial calcification measurements could predict cardiovascular disease and death, validating the deep learning model across 12 institutions and comparing AI-generated measurements directly against radiologist assessments.
The results demonstrated a striking correlation between severe vascular calcification in breast tissue and future cardiac events. According to Mayo Clinic data, women classified with severe breast arterial calcification faced more than 10 times the risk of developing a cardiovascular event within five years compared to those with no or mild calcification.
Moving Toward Widespread Clinical Adoption
Traditional risk assessment tools typically rely on body weight, blood pressure, and cholesterol levels. These metrics often fail to capture the physiological realities of how vascular disease progresses in women.
By extracting vascular data from standard breast cancer screenings, the new AI tool offers an objective risk metric without adding patient burden. The algorithm is currently undergoing review by the Food and Drug Administration to clear the path for widespread clinical adoption.