An AI algorithm trained on routine mammogram images and a woman’s age can predict her risk of major cardiovascular disease with accuracy comparable to established clinical risk assessment tools, according to new research published in the journal Heart. The findings suggest that mammography could serve as a cost-effective "two-for-one" screening opportunity—detecting both breast cancer and cardiovascular risk during a single, widely available procedure. Cardiovascular disease remains a leading cause of death among women worldwide, yet it is often underrecognized and undertreated. Traditional risk prediction models have historically performed less accurately in women than in men, partly because they rely on extensive medical data that isn’t always accessible. While newer tools like the New Zealand PREDICT and American Heart Association PREVENT calculators offer improved performance, they still require detailed clinical information such as cholesterol levels, blood pressure, and medication use. The researchers explored whether internal breast characteristics visible on mammograms—such as tissue density and calcifications—could offer a more accessible alternative. Previous studies have linked breast arterial calcification (BAC) to cardiovascular risk, though BAC is influenced by factors like obesity and smoking, limiting its reliability. To test their hypothesis, the team analyzed data from 49,196 women, with an average age of 59, enrolled in the Lifepool cohort in Victoria, Australia between 2009 and 2020. At enrollment, participants provided information on age, smoking, alcohol use, BMI, diabetes history, and use of medications for high blood pressure, high cholesterol, and blood thinners. Additional data included menopause status, reproductive history, hormone therapy use, and prior breast treatments. Over an average follow-up of nearly nine years, 3,392 women experienced a first major cardiovascular event—2,383 cases of coronary artery disease, 656 heart attacks, 434 strokes, and 731 cases of heart failure. Using deep learning, the researchers developed an AI model that analyzed the full range of internal breast features visible in mammograms, combined with age, to predict 10-year cardiovascular risk. The model performed as well as or slightly better than standard clinical risk scores, even when those scores included additional clinical data. The researchers note several limitations: variations in mammography equipment across facilities, reliance on self-reported health data, and the dependence of AI models on the quality and scope of their training data. However, they highlight a major advantage—no additional patient history or medical records were needed. The model leveraged an existing screening process already used by millions of women. In a related editorial, Professors Gemma Figtree and Stuart Grieve from the University of Sydney emphasize that heart disease kills far more women globally than breast cancer—yet awareness remains low. They suggest mammography could serve as a critical "touch point" to raise awareness about cardiovascular risk. They caution, however, that while promising, the real challenge lies in implementing such tools effectively within healthcare systems. Still, the potential for a dual-purpose screening method using widely available imaging data represents a significant step forward in preventive care for women.