Journal article

Cirrus: An Automated Mammography-Based Measure of Breast Cancer Risk Based on Textural Features

Daniel F Schmidt, Enes Makalic, Benjamin Goudey, Gillian S Dite, Jennifer Stone, Tuong L Nguyen, James G Dowty, Laura Baglietto, Melissa C Southey, Gertraud Maskarinec, Graham G Giles, John L Hopper

JNCI Cancer Spectrum | OXFORD UNIV PRESS | Published : 2018


Background: We applied machine learning to find a novel breast cancer predictor based on information in a mammogram. Methods: Using image-processing techniques, we automatically processed 46 158 analog mammograms for 1345 cases and 4235 controls from a cohort and case-control study of Australian women, and a cohort study of Japanese American women, extracting 20 textural features not based on pixel brightness threshold. We used Bayesian lasso regression to create individual- and mammogram-specific measures of breast cancer risk, Cirrus. We trained and tested measures across studies. We fitted Cirrus with conventional mammographic density measures using logistic regression, and computed odds ..

View full abstract