AI in mammograms may predict women's heart disease risk

A new study suggests that mammograms could help detect heart disease in women, using artificial intelligence (AI). This research is being presented at the American College of Cardiology's Annual Scientific Session. Mammograms are typically used to screen for breast cancer. However, they can also show calcium buildup in arteries, which is important for heart health. Currently, about 40 million mammograms are performed in the U.S. each year. Most doctors do not analyze or report the calcium information found in these images. The study used a new AI image analysis method to evaluate breast arterial calcification. This technique helps derive a cardiovascular risk score for women. The lead author, Dr. Theo Dapamede, emphasized that mammograms could provide additional cardiovascular screenings for women, especially for those under 60 years old. Heart disease is the leading cause of death among women in the U.S., but many cases go undiagnosed. Researchers believe that using AI in mammogram screenings can help identify women at risk of heart disease earlier. The new AI tool was trained on over 56,000 patients' records from Emory Healthcare. It successfully categorized women's cardiovascular risk as low, moderate, or severe based on their mammogram images. The results showed that younger women with more calcium buildup had a higher risk of serious heart issues within five years. Currently, this AI model is not available for use. If it receives further validation and approval from the U.S. Food and Drug Administration, it could be adopted by healthcare systems during routine mammogram screenings. Researchers are also interested in how similar AI technology can be used to assess other health conditions in the future.