AI Detects Heart Disease Risk in Mammograms: New Screening Potential

The routine breast cancer screening many women undergo could soon offer a second, unexpected benefit: an assessment of their heart disease risk. New research published today in the European Heart Journal demonstrates that artificial intelligence can analyze mammograms to quantify calcium deposits in breast arteries – a marker strongly linked to the potential for serious cardiovascular events like heart attack and stroke.

The study, led by Dr. Hari Trivedi of Emory University in Atlanta, USA, analyzed mammograms from over 123,000 women aged 40 to 79 who had no prior history of cardiovascular disease. Researchers found a clear correlation between the amount of calcium detected in breast arteries and the likelihood of experiencing a major cardiovascular event within approximately seven years. Women with moderate calcification faced a risk increase of over 70%, while those with severe calcification saw their risk climb two to three times higher compared to women with no detectable calcification. Even younger women, under 50, showed a statistically significant increase in risk with higher levels of arterial calcium.

Beyond Cancer Detection: A New Role for Mammography?

Cardiovascular disease remains the leading cause of death for women globally, yet it’s often underdiagnosed and undertreated in this population. This research suggests a potential pathway to address this disparity. Mammograms, already a widely utilized screening tool, could offer a ‘two-for-one’ opportunity to assess both breast and heart health. The AI-driven analysis focuses on breast arterial calcification (BAC), a process where calcium builds up in the walls of the arteries, indicating hardening and increasing the risk of blockages. Inside Precision Medicine highlights that this method could calculate cardiovascular risk based on the amount of calcium present.

“We found that the more calcium visible in the breast arteries on a mammogram, the higher a woman’s risk of a serious heart event such as a heart attack, stroke or heart failure,” explained Dr. Trivedi. “This was true even in younger women under 50—a group often considered low-risk—and held up after accounting for other risk factors like diabetes and smoking.”

How the Study Worked and What It Doesn’t Share Us

The study utilized a retrospective analysis of mammograms from two large US health systems. The AI algorithm measured the area of calcium in the breast arteries, categorizing women into four groups: no BAC, mild BAC (0–10 mm²), moderate BAC (10–25 mm²), and severe BAC (over 25 mm²). Researchers then tracked the women for an average of seven years, monitoring for the occurrence of acute myocardial infarction (heart attack), stroke, heart failure, and all-cause death.

It’s important to note that this study demonstrates a strong correlation between BAC and cardiovascular risk, but it does not prove causation. While the presence of calcium in breast arteries is a marker of arterial health, it doesn’t directly cause heart attacks or strokes. Other factors, such as genetics, lifestyle, and underlying health conditions, similarly play significant roles. The study population was limited to women who had undergone routine breast cancer screening, which may not be fully representative of all women. The Medical Xpress report notes this is the largest study of its kind, covering multiple races across two major US health systems, which strengthens the findings, but doesn’t eliminate the need for further research.

Implications for Prevention and Public Health

The potential implications of this research are substantial. Integrating BAC assessment into routine mammography could identify women at increased cardiovascular risk who might not otherwise be flagged. This could lead to earlier interventions, such as lifestyle modifications, cholesterol management, or medication, potentially preventing serious heart events.

Professor Lori B. Daniels from the University of California, San Diego, USA, in an accompanying editorial, points out that mammography uptake is relatively high – nearly 70% of women aged 45 and older in the US are up to date with screening – while awareness of cholesterol levels is significantly lower (under 40%). Leveraging an existing, widely-used screening program to identify cardiovascular risk could be a highly effective public health strategy. The George Institute for Global Health developed a similar machine learning model, demonstrating comparable accuracy to traditional cardiovascular risk calculators without requiring extensive clinical data.

What’s Next: From Research to Routine Screening?

While the findings are promising, several steps are needed before BAC assessment becomes a standard part of mammography. Dr. Trivedi and his team are planning a clinical trial to test the feasibility and effectiveness of integrating the AI tool into existing imaging workflows and establishing clear guidelines for notifying patients and doctors about their risk levels. This trial will be crucial in determining how best to translate this research into clinical practice.

Further research is also needed to refine the AI algorithm and determine the optimal threshold for BAC levels that trigger further cardiovascular risk assessment. Standardizing the measurement of BAC (currently reported in mm²) will be essential for widespread adoption. Policymakers will need to consider the cost-effectiveness of integrating this new assessment into existing mammography programs and address any potential ethical concerns related to incidental findings.

For women, this research offers a potential new layer of insight into their overall health. It underscores the importance of discussing cardiovascular risk factors with a healthcare provider and staying up-to-date with recommended screenings. The conversation sparked by a routine mammogram could now extend beyond breast cancer detection to encompass a more comprehensive assessment of heart health.