Opportunistic AI tool predicts CVD risk using routine mammograms
Experts have developed an artificial intelligence tool that they believe can help identify more women at risk of cardiovascular disease using routine mammograms.
New research published in the European Heart Journal details the utility of a transformer-based neural network that can quantify CVD risk based on the presence of breast arterial calcifications (BACs) on screening mammograms. Since breast cancer screening exams are routinely completed by many women annually or bi-annually, experts believe the tool could help spot signs of developing CVD prior to it progressing. This would give providers an opportunity to proactively treat and manage CVD to mitigate the risk of adverse events.
“Heart disease is the leading cause of death in women worldwide, yet women are consistently underdiagnosed and undertreated compared to men,” noted research leader Hari Trivedi, MD, from the department of radiology at Emory University, Atlanta. “Mammograms, which women already attend for breast cancer screening, can also reveal calcium deposits in breast arteries and this is linked to heart disease. We wanted to test whether AI could use this to identify women at risk of cardiovascular disease at no extra cost or inconvenience.”
For their study, researchers retrospectively tested the tool on nearly 124,000 women who had completed screening mammograms at two healthcare systems. BACs were quantified and later categorized as either zero (0 mm2), mild (>0–10 mm2), moderate (>10–25 mm2) or severe (>25 mm2) with respect to CVD risk. The team compared the tool’s metrics to patients’ medical records to determine whether there were associations between its findings and the likelihood of a woman later experiencing a major adverse cardiovascular event.
Researchers determined women who were in the mild category were around 30% more likely to suffer a major event compared to those without BACs. These risks increased in women with moderate BACs, while those in the severe category were found to have double, sometimes triple, the risk of major events compared to the women without calcifications.
“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,” Trivedi said. “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.”
The group suggested that their findings highlight a considerable opportunity to leverage routine breast cancer screenings to identify women with undiagnosed CVD.
“Policymakers could consider integrating this into existing mammography programs, potentially reaching tens of millions of women annually without any additional infrastructure,” Trivedi suggested.
Read more here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.