Our paper, Using a convolutional neural network to predict readers’ estimates of mammographic density for breast cancer risk assessment, as been published to IWBI 2018.

We use a convolutional neural network to predict breast density from mammograms based on readers’ estimates. Density estimates are then used to predict breast cancer risk. In a matched case-control dataset of screen detected cancers, the odds ratio between the highest and lowest quintiles is 3.07 (95% CI: 1.97 - 4.77).

Full text available here: Using a convolutional neural network to predict readers’ estimates of mammographic density for breast cancer risk assessment