A recent retrospective study provides the latest evidence that using AI to assess calcium buildup in breast arteries, visible on routine mammograms, may help predict cardiovascular disease risk.

The research, led by Matthew Nudy, assistant professor of medicine and public health sciences at Penn State College of Medicine, found that women with more severe breast arterial calcification (BAC) and those whose BAC progressed over time faced a significantly higher risk of serious cardiovascular events, including heart attack, stroke, heart failure and death.

The findings, which were highlighted last month at the annual RSNA meeting in Chicago, examined 10,348 women in the U.S. who underwent sequential mammograms over an average span of 4.1 years.
Using an experimental version of cmAngio, an AI-based tool developed by CureMetrix, researchers quantified BAC based on the length of calcified arteries relative to breast size. BAC severity was then categorized into age-adjusted percentiles and tracked over time.

The future of cardiovascular risk screening strategies
BAC is not typically reported since it has no established link to breast cancer. However, its association with cardiovascular disease has drawn increasing attention. Previous research involving health systems like Kaiser Permanente and Mayo Clinic has shown that BAC presence and severity independently correlate with elevated cardiovascular risk.

Participants in the Penn State study were grouped into four categories (negative, mild, moderate or severe) based on age-adjusted calcium scores. At the outset, 19.4% of participants showed signs of arterial calcification. The risk of cardiovascular events rose in step with increasing BAC severity. Notably, women who moved from moderate to severe calcification between mammograms had a 93% higher risk of cardiovascular-related outcomes. Those who progressed from mild to any higher category faced a 59% increased risk.

"Because BAC is already visible on routine screening mammograms, I see it evolving into a standardized, AI-enabled 'cardiovascular risk enhancer' that can be reported consistently and used to trigger a focused cardiovascular risk-factor assessment, and subsequently used to intensify guideline-based preventive therapies such as statins and intensive blood pressure control," Nudy told HCB News.

While the study findings are promising, additional research is needed to determine how BAC data might be integrated into cardiovascular risk screening strategies. "The biggest evidence gap is demonstrating that BAC-informed care can improve clinical outcomes, not just prediction of risk," said Nudy.

The reimbursement challenge
CureMetrix, the San Diego-based company behind cmAngio, sees lack of reimbursement as a critical hurdle preventing the clinical implementation of AI tools. There is currently no CPT code for mammography AI software, Dr. Lisa Watanabe, the chief medical officer at CureMetrix, told HCB News. "At some mammography facilities, women are opting in to pay out of pocket for the use of AI," she said, adding that HSA/FSA funds may be applied in those cases.

BAC reporting may be reimbursed for groups that are participating in the Medicare Merit-based incentive payment system (MIPS) bonus program.

"BAC reporting is available as a MIPS quality measure (QMM28)," said Watanabe. "Under the measure, presence or absence of BAC in screening mammogram for females, ages 40 and older, must be reported, and a clinical relevance statement suggesting correlation between BAC and CAD with appropriate follow-up recommendation needs to be included."

Although the version of cmAngio used in the new research is not currently approved for clinical use with scoring, another version of the software was cleared by the FDA in 2023 to detect and localize BAC on mammograms.

The study was supported by CureMetrix and involved collaborators from Monash University, the University of Southern California and the CureMetrix research team.

Artificial Intelligence Homepage