What if a mammogram could reveal a woman’s risk of developing breast cancer years before the disease actually appears? Scientists are increasingly using artificial intelligence to make that possibility a reality.
Researchers have developed AI systems capable of analysing mammogram images and identifying patterns associated with a woman’s future breast cancer risk.
One of the best known systems, Mirai, was developed by researchers from MIT and Massachusetts General Hospital. The model was trained using more than 200,000 mammography examinations and was designed to estimate breast cancer risk at different points over the following five years.
Unlike conventional mammography, which primarily looks for signs of cancer that are already present, Mirai analyses subtle patterns in breast tissue that may be linked to future cancer development. MIT researchers reported that the system could identify women at elevated risk several years before a diagnosis.
However, the technology does not mean AI can definitively diagnose cancer five years in advance. It estimates risk, and researchers say such models require continued validation across different populations and healthcare settings before they can be widely integrated into routine care.
Recent research has also continued to examine how Mirai makes its predictions, including the role of subtle differences between the two breasts and features such as calcifications.
The development highlights a growing role for AI in medical imaging, where researchers hope earlier identification of women at higher risk could help doctors personalise screening and preventive care.
