Improving breast cancer screening through dynamic big data analytics of Quantitative Imaging Biomarkers (QIBs)

In order obtain a risk assessment for a person from a breast cancer screening, Quantitative Imaging Biomarkers (QIB) can be used. They are hand-crafted imaging filters that allow for the detection of specific patterns in an image of a breast scan. Instead of dedicating a huge effort toward improving these hand-crafted filters, the training of Machine-learned filters via a Convolutional Neural Network might allow for an improved screening. The data consists not only of the different images of different patients, but also of different time points for a given patient. To incorporate this time data, a unique Neural Network architecture will be developed with the goal of accurately assessing the risk for breast cancer and therefore improving the screening process.