The use of radiomics in breast cancer screening for high risk patients

It is well known that the risk to develop breast cancer is different for different groups of women. In this project, we will focus on a specific sub group, namely the women with higher than normal risk to develop cancer. One factor that determines the risk to develop cancer is the amount of glandular tissue in the breast, known as breast glandularity. The breasts are typically categorized into BIRADS categorized. However, many other factors determine the risk, such as the frequency of breast cancers in the family, the number of pregnancies, etc. Risk prediction is however far from perfect, and we aim to further refine the risk by using radiomics techniques. ‘Radiomics’ is the science that extracts mathematical features from a radiological image. In this project, radiomics will be applied to enhance the risk prediction in the higher risk group population. In our department, we have a unique data base of 500 patients that have been imaged with 2D mammography, breast tomosynthesis and mostly also breast MRI. The department has recently installed an automated 3D breast ultrasound devices (ABUS) that will be applied in the same subgroup of patients. The higher risk population is even screened yearly. In an extra project, also radiomics of photon counting breast CT can be calculated. In a parallel subproject of current project, radiomics will be applied on the normal screening population, and in particular also their two-yearly subsequent screenings.