Onderzoek van geavanceerde röntgenbeeldvormingstechnieken van de borst met behulp van fysieke metingen en simulaties

This thesis focuses on enhancing virtual imaging trials (VITs) as a tool for evaluating and optimizing advanced breast x-ray imaging modalities, such as digital breast tomosynthesis (DBT), synthetic mammography (SM) and contrast-enhanced mammography (CEM). VITs relies on computational models that accurately represent human anatomy, pathology, and imaging systems. This work contributes to the validation of existing computational breast phantoms, explores a deep learning method to enhance the efficiency of mass model generation, and improves the realism of simulated mass lesions by developing an automated procedure for insertion into patient images. Additionally, it enables simulations for the latest Siemens mammography system, and demonstrates the added value of VITs in field of breast x-ray imaging.