Cone Beam
Cone-beam computed tomography (CBCT) is a medical imaging technique producing detailed 3D images, primarily used in dentistry and radiation therapy. Current research focuses on improving CBCT image quality through techniques like deep learning-based image enhancement and synthesis (using models such as GANs, diffusion models, and U-Nets), motion artifact reduction via optimization algorithms and deep learning, and efficient reconstruction methods from sparse-view data using neural networks and 3D Gaussian representations. These advancements aim to reduce radiation exposure, improve diagnostic accuracy, and enable more precise treatment planning, ultimately impacting patient care and clinical workflows.
Papers
Deep learning based projection domain metal segmentation for metal artifact reduction in cone beam computed tomography
Harshit Agrawal, Ari Hietanen, Simo Särkkä
REGAS: REspiratory-GAted Synthesis of Views for Multi-Phase CBCT Reconstruction from a single 3D CBCT Acquisition
Cheng Peng, Haofu Liao, S. Kevin Zhou, Rama Chellappa