How AI-assisted mammograms can identify cardiovascular risk for women

Over the next 25 years, the number of women living with heart disease is expected to grow substantially. The American Heart Association predicts that by 2050, almost 60% of all women will have high blood pressure, and nearly one-third of women 20 to 44 years of age will have some type of cardiovascular disease.

“More than 62 million women in the U.S. are living with some type of cardiovascular disease, and that comes with a price tag of at least $200 billion annually. Our estimates indicate that if we stay on the current path, these numbers will grow substantially over the next 25 to 30 years,” Karen E. Joynt Maddox, M.D., M.P.H., FAHA, said in a news release. She is a professor of medicine and public health and the co-director of the Center for Advancing Health Services, Policy and Economics Research at the Washington University School of Medicine.

Mammograms, however, can help to diagnose cardiovascular disease sooner. In a new study, researchers found that artificial intelligence (AI)-assisted quantification of breast arterial calcification (BAC) can be used to help physicians identify patients at risk for heart disease.

BACs are benign calcium deposits in the arteries of the breast. These deposits are not associated with breast cancer, but research over more than two decades has demonstrated that they are a predictor of cardiovascular disease.

In the study, recently published in the European Heart Journal, researchers wanted to determine whether AI could identify BAC that is likely to lead to cardiovascular disease. Researchers were led by Hari M. Trivedi, M.D., an associate professor in the Department of Radiology and Imaging Sciences and in the Department of Biomedical Informatics at Emory University School of Medicine.

The study included 74,124 women who were in the Emory Breast Imaging Dataset, as well as 49,638 women from Mayo Clinic Enterprise. Women were between the ages of 40 and 79 years, and researchers excluded women who had major adverse cardiovascular events (MACEs) and women who had coronary artery calcification scoring or coronary angiography before the index mammogram. The median follow-up period was seven years.

For all analyses, only a single index mammogram per patient was used. Any BAC was quantified through an automated AI system to determine the severity of the calcification: zero, mild, moderate or severe. The model was trained, validated and tested using a set of 1,000 screening mammograms.

Researchers reviewed electronic health records at Emory for diagnostic codes to identify MACEs and all-cause mortality. Researchers also had follow-up visits with patients not in the Emory health records.

The AI-assisted mammograms were able to detect BAC in 16.1% of the patients at Emory patients and 20.6% of those at Mayo Clinic. In both groups of women, the incidence of MACEs was higher in those with severe BAC. The severity of the calcification was correlated with age and cardiometabolic risk factors, including diabetes, use of antihypertensive medications and statins, higher systolic blood pressure and higher body mass index. Smoking, however, was not found to be associated with BAC.

The researchers said the AI-assisted approach has predictive value that supplements the Predicting Risk of Cardiovascular Disease Events (PREVENT) score that is generally recommended. The PREVENT calculator was developed by the American Heart Association in 2023 to estimate 10-year and 30-year risk for cardiovascular disease. It is a risk tool that combines cardiovascular, kidney and metabolic health measures to guide prevention efforts.

“Automatically quantified BAC is an independent predictor of MACEs and mortality, adding prognostic value to the PREVENT score. This approach may provide an opportunistic cardiovascular risk assessment during routine mammography screening without additional radiation exposure to guide earlier and more effective preventive care for women,” the authors wrote.