Artificial intelligence model bests previous methods in predicting breast cancer risk

Investigators at MIT and MGH have collaborated to develop a new AI model to predict a woman’s future risk of breast cancer based on her mammogram alone, and performed validation trials in patient populations in Europe and Asia. The system was significantly more accurate than prior methods in predicting cancer risk and identifying high risk groups across the three datasets the team used.

This work, recently reported in Science Translational Medicine, is part of an ongoing collaboration between MIT computer scientist/professor Regina Barzilay, PhD, and Constance Lehman, MD, PhD, professor of radiology at Harvard Medical School and head of Breast Imaging at MGH, who were co-senior authors.

Says Leslie Lamb, MD, co-author, faculty at Harvard Medical School and breast imaging radiologist at MGH: “We are delighted with the promise of the AI tools to better predict a woman’s true risk of breast cancer and eager to evaluate the ability of these tools to improve outcomes for our patients.”

Read the full story from MIT here.