2020/05/15 by Maureen van Eijnatten, van Eijnatten, Maureen, Leonardo Rundo +17
Engineering · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Medical Imaging Techniques and Applications #Medical Imaging and Analysis #Radiomics and Machine Learning in Medical Imaging #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2005.07545
openalex publication_date 2020/05/15 · openalex created_date 2025/10/27 · openalex updated_date 2026/07/28
This study investigates the use of the unsupervised deep learning framework\nVoxelMorph for deformable registration of longitudinal abdominopelvic CT images\nacquired in patients with bone metastases from breast cancer. The CT images\nwere refined prior to registration by automatically removing the CT table and\nall other extra-corporeal components. To improve the learning capabilities of\nVoxelMorph when only a limited amount of training data is available, a novel\nincremental training strategy is proposed based on simulated deformations of\nconsecutive CT images. In a 4-fold cross-validation scheme, the incremental\ntraining strategy achieved significantly better registration performance\ncompared to training on a single volume. Although our deformable image\nregistration method did not outperform iterative registration using NiftyReg\n(considered as a benchmark) in terms of registration quality, the registrations\nwere approximately 300 times faster. This study showed the feasibility of deep\nlearning based deformable registration of longitudinal abdominopelvic CT images\nvia a novel incremental training strategy based on simulated deformations.\n