3D deformable registration of longitudinal abdominopelvic CT images using unsupervised deep learning
This work addresses the need for efficient image registration in medical imaging for cancer patients, but it is incremental as it builds on existing deep learning methods with a novel training strategy.
This study tackled the problem of deformable registration of longitudinal abdominopelvic CT images in patients with bone metastases from breast cancer using an unsupervised deep learning framework, achieving registrations approximately 300 times faster than a benchmark method, though with slightly lower quality.
This study investigates the use of the unsupervised deep learning framework VoxelMorph for deformable registration of longitudinal abdominopelvic CT images acquired in patients with bone metastases from breast cancer. The CT images were refined prior to registration by automatically removing the CT table and all other extra-corporeal components. To improve the learning capabilities of VoxelMorph when only a limited amount of training data is available, a novel incremental training strategy is proposed based on simulated deformations of consecutive CT images. In a 4-fold cross-validation scheme, the incremental training strategy achieved significantly better registration performance compared to training on a single volume. Although our deformable image registration method did not outperform iterative registration using NiftyReg (considered as a benchmark) in terms of registration quality, the registrations were approximately 300 times faster. This study showed the feasibility of deep learning based deformable registration of longitudinal abdominopelvic CT images via a novel incremental training strategy based on simulated deformations.