By Billy Candra 
A groundbreaking study presented at the American College of Cardiology’s Annual Scientific Session (ACC.25) reveals that mammograms, when enhanced with sophisticated artificial intelligence (AI) models, can offer far more than just breast cancer detection. These routine cancer screening tools hold the potential to simultaneously assess cardiovascular health by quantifying calcium buildup in the arteries within breast tissue, a known indicator of heart disease risk. This innovative application of existing imaging technology promises to transform preventive healthcare, particularly for women, who often face unique challenges in early cardiovascular disease diagnosis.
The Untapped Potential of Routine Mammography
Mammography, an X-ray of the breast, is a cornerstone of women’s health, with the U.S. Centers for Disease Control and Prevention recommending biennial or annual screenings for middle-aged and older women. Approximately 40 million mammograms are performed in the United States each year, making it one of the most common diagnostic procedures. While breast artery calcifications (BAC) are frequently visible on these images, radiologists have historically not quantified or routinely reported this information to patients or their referring clinicians. The new research, leveraging a novel AI image analysis technique, aims to bridge this critical gap by automating the analysis of BAC and translating these findings into a tangible cardiovascular risk score.
Dr. Theo Dapamede, MD, PhD, a postdoctoral fellow at Emory University in Atlanta and the study’s lead author, highlighted the significant opportunity this presents. "We see an opportunity for women to get screened for cancer and also additionally get a cardiovascular screen from their mammograms," Dr. Dapamede stated, emphasizing the dual-purpose potential. "Our study showed that breast arterial calcification is a good predictor for cardiovascular disease, especially in patients younger than age 60. If we are able to screen and identify these patients early, we can refer them to a cardiologist for further risk assessment." This proactive approach could lead to earlier interventions and improved outcomes for a population segment often underserved in cardiovascular prevention.
Cardiovascular Disease: A Silent Threat to Women
Heart disease remains the leading cause of death globally and in the United States, yet it is notoriously underdiagnosed in women, with a persistent lack of awareness regarding its specific manifestations and risk factors in the female population. Traditional risk assessments, while valuable, may not always capture the full spectrum of risk for women, who can present with atypical symptoms or have different underlying pathologies compared to men. The ability to leverage an existing, widely adopted screening tool like mammography for cardiovascular risk assessment could therefore represent a monumental leap forward in addressing this disparity. By identifying women with early signs of cardiovascular disease through AI-enabled mammogram screening, healthcare providers can initiate targeted preventive strategies and lifestyle modifications, potentially averting serious cardiac events.
A buildup of calcium in blood vessels, known as atherosclerosis, is a definitive sign of cardiovascular damage, intricately linked to early-stage heart disease and the natural aging process. Prior research has robustly demonstrated that women exhibiting calcium buildup in their arteries face a substantially elevated risk—specifically, a 51% higher risk—of experiencing heart disease and stroke. This underscores the critical importance of detecting and monitoring BAC, which can serve as an early warning signal even before overt symptoms manifest.
The Genesis of the AI Tool: Deep Learning and Segmentation
The development of this innovative screening tool involved training a sophisticated deep-learning AI model. Researchers specifically taught the AI to "segment" calcified vessels within mammogram images. These calcifications appear as bright pixels on X-rays, and the AI’s task was to precisely delineate and quantify these areas. This segmentation approach is a key differentiator from previous AI models developed for analyzing breast artery calcifications, allowing for a more granular and accurate assessment. Following segmentation, the AI model was trained to calculate the future risk of cardiovascular events by correlating these imaging findings with comprehensive electronic health record data.
The strength of this model is further augmented by the expansive dataset utilized for its training and testing. The study incorporated images and health records from over 56,000 patients who underwent mammography at Emory Healthcare between 2013 and 2020. Crucially, each patient in the dataset had at least five years of follow-up electronic health records data, providing a robust longitudinal perspective on their health outcomes. This extensive data pool enabled the AI to learn and generalize patterns of cardiovascular risk with remarkable precision, moving beyond theoretical models to practical, clinically relevant applications.
Dr. Dapamede elaborated on the technological underpinnings, noting, "Advances in deep learning and AI have made it much more feasible to extract and use more information from images to inform opportunistic screening." This concept of "opportunistic screening" is central to the study’s potential impact, suggesting that existing medical procedures can be repurposed to gain additional, highly valuable health insights without requiring separate appointments or additional patient burden.
Key Findings: Predictive Power Across Age Groups
The overall findings of the study demonstrate the new AI model’s impressive performance in characterizing patients’ cardiovascular risk. Based solely on mammogram images, the model effectively categorized risk levels as low, moderate, or severe. Researchers then calculated the risk of major cardiovascular events—such as acute heart attack, stroke, heart failure, or death from any cause—at both two-year and five-year intervals.
A significant revelation from the analysis was the correlation between breast arterial calcification levels and the incidence of these serious cardiovascular events. The model showed a clear increase in event rates corresponding to higher BAC levels in two crucial age categories: women younger than age 60 and those between age 60-80. Interestingly, this predictive power was not observed in women over age 80, suggesting that in very elderly populations, other age-related factors might overshadow BAC as a primary cardiovascular risk indicator. This makes the AI tool particularly valuable for providing early warning of heart disease risk in younger women, a demographic that stands to benefit most from early interventions and aggressive risk factor modification.
The study provided compelling statistical evidence of this predictive power. Women with the highest level of breast arterial calcification (exceeding 40 mm²) exhibited a significantly lower five-year rate of event-free survival compared to those with the lowest level (below 10 mm²). Specifically, only 86.4% of women with severe BAC survived for five years without a major cardiovascular event, in stark contrast to 95.3% of those with minimal or no calcification. This translates to an approximately 2.8 times higher risk of death within five years for patients with severe breast arterial calcification compared to those with little to no BAC. Such a dramatic difference underscores the profound clinical implications of these findings.
Collaboration, Validation, and Future Horizons
The AI model represents a collaborative effort between two prominent medical institutions: Emory Healthcare and Mayo Clinic. While the model has shown promising results in this initial study, it is not yet available for clinical use. The path to widespread adoption involves several critical steps, including external validation to confirm its efficacy across diverse patient populations and healthcare settings. Following successful validation, the tool would need to secure approval from the U.S. Food and Drug Administration (FDA) to ensure its safety and effectiveness for commercial availability. If these hurdles are cleared, researchers envision the tool being seamlessly integrated into routine mammogram processing and follow-up care within other healthcare systems.
Beyond its immediate application in cardiovascular risk assessment, the research team is already exploring broader horizons for similar AI models. They plan to investigate how AI could be utilized to extract biomarkers for other significant conditions, such as peripheral artery disease and kidney disease, from existing mammogram images. This vision points towards a future where routine medical imaging, once focused on a single diagnostic purpose, becomes a rich source of comprehensive health data, revolutionizing opportunistic screening across a spectrum of diseases.
Broader Implications for Public Health and Healthcare Systems
The implications of this research extend far beyond individual patient care. From a public health perspective, the ability to identify a large cohort of women at elevated cardiovascular risk through an existing, routine screening program could significantly improve population health outcomes. Early detection enables earlier intervention, potentially reducing the incidence of costly and debilitating cardiovascular events like heart attacks and strokes. This shift from reactive treatment to proactive prevention could lead to substantial savings in healthcare costs over time, alleviating the burden on healthcare systems.
For radiologists, this development presents both an opportunity and a challenge. While the AI automates the quantification of BAC, it may necessitate new protocols for reporting these findings and collaborating with cardiologists. Training for radiologists and other healthcare professionals will be crucial to ensure effective communication of risk scores to patients and appropriate referral pathways. Cardiologists, in turn, could see an influx of referrals for early-stage risk assessment, requiring adjustments in practice patterns and increased focus on preventive cardiology.
Patient communication will also be paramount. Explaining a cardiovascular risk score derived from a mammogram, traditionally associated solely with cancer, will require clear, sensitive, and informative dialogue to avoid confusion or anxiety. Educational campaigns may be necessary to raise public awareness about the dual screening potential of mammograms and the significance of breast arterial calcification.
In essence, this study heralds a new era of "precision prevention" in women’s health. By transforming a long-standing cancer screening into a powerful dual diagnostic tool, AI-enhanced mammography offers a practical, scalable, and non-invasive method to identify women at risk for heart disease earlier than ever before. This innovative application of artificial intelligence underscores its transformative potential in medical diagnostics, promising a future where healthcare is not only more efficient but also more comprehensive and ultimately, more life-saving. The journey from research to widespread clinical implementation is complex, but the promise of improved cardiovascular outcomes for millions of women makes this a frontier well worth exploring.