Abstract
<jats:p>Breast cancer is the most common type of cancer in women. Early detection is important for survival. That is why there is a breast cancer screening program, in which X-ray images of the breasts are taken to detect the disease at an early stage. These images are assessed by two radiologists for possible abnormalities. This is a challenging process, because abnormalities are often difficult to see. This thesis shows how the interpretation of these images can be improved. It investigated which radiologists work best together, in what order the images are best read, and how artificial intelligence (AI) can support the interpretation. The research shows that radiologists performed better when they reviewed images from low to high breast density and when they used AI support. These are practical improvements that can be directly applied to improve image interpretation in breast cancer screening, contributing to better early detection of breast cancer.</jats:p>