Deep learning for x‐ray scatter correction in dedicated breast CT

This repository accompanies the publication Deep learning for x-ray scatter correction in dedicated breast CT (Pautasso et al., 2023, Med Phys). It includes training and inference scripts, model weights, and analysis files. The method corrects scatter in breast CT projections using a deep convolutional neural network trained on simulated data and validated with both simulated and real patient scans. This dataset was created as part of the EU-funded project BREAST4D (Grant Agreement ID: 864929, DOI: 10.3030/864929), which aimed to develop and validate a 4D breast imaging modality for personalized breast cancer treatment.