Overall, the study found that both in-depth learning models were better than the standard approach for detecting women at high risk for breast cancer. The last hope is to tailor breast cancer screening to each woman,” said Adam Yala, a principal investigator and doctoral student at the Massachusetts Institute of Technology. From there, the researchers developed two in-depth learning models: one that used only mammography data and the other, a “hybrid” model that included traditional factors such as age and family history, as well as the woman’s breast density. It appears that doctors already take some factors into account when assessing a woman’s breast cancer risk. He estimates a woman’s risk of developing breast cancer based on traditional risk factors and breast density. Nearly one-third of women who developed breast cancer ranked in the top 10% of the risk category.