UMedNeRF: Uncertainty-aware Single View Volumetric Rendering for Medical Neural Radiance Fields (2311.05836v7)
Abstract: In the field of clinical medicine, computed tomography (CT) is an effective medical imaging modality for the diagnosis of various pathologies. Compared with X-ray images, CT images can provide more information, including multi-planar slices and three-dimensional structures for clinical diagnosis. However, CT imaging requires patients to be exposed to large doses of ionizing radiation for a long time, which may cause irreversible physical harm. In this paper, we propose an Uncertainty-aware MedNeRF (UMedNeRF) network based on generated radiation fields. The network can learn a continuous representation of CT projections from 2D X-ray images by obtaining the internal structure and depth information and using adaptive loss weights to ensure the quality of the generated images. Our model is trained on publicly available knee and chest datasets, and we show the results of CT projection rendering with a single X-ray and compare our method with other methods based on generated radiation fields.
- “Extraction of airways from ct (exact’09),” IEEE Transactions on Medical Imaging, vol. 31, no. 11, 2012.
- “Artificial intelligence in image-based cardiovascular disease analysis: A comprehensive survey and future outlook,” arXiv:2402.03394, 2024.
- “2d-to-3d: A review for computational 3d image reconstruction from x-ray images,” Archives of Computational Methods in Engineering, 2023.
- “Deep reinforcement learning for image-to-image translation,” arXiv:2309.13672, 2023.
- “Nerf: Representing scenes as neural radiance fields for view synthesis,” 2021, vol. 65, ACM New York, NY, USA.
- “Neural radiance fields in medical imaging: Challenges and next steps,” 2024.
- “Mednerf: Medical neural radiance fields for reconstructing 3d-aware ct-projections from a single x-ray,” in EMBC. IEEE, 2022.
- “Multi-task learning using uncertainty to weigh losses for scene geometry and semantics,” in CVPR, 2018.
- “Uncertainty aware semi-supervised learning on graph data,” NeurIPS, vol. 33, pp. 12827–12836, 2020.
- “Learning by minimizing the sum of ranked range,” NeurIPS, vol. 33, pp. 21013–21023, 2020.
- E. B. Tsai et al., “Medical imaging data resource center (midrc) - rsna international covid open research database (ricord) release 1b - chest ct covid,” 2021.
- “The cancer imaging archive (tcia): Maintaining and operating a public information repository,” 2013, vol. 26.
- “A combined experimental and computational approach to subject-specific analysis of knee joint laxity,” 2016, vol. 138.
- “Validation of predicted patellofemoral mechanics in a finite element model of the healthy and cruciate-deficient knee,” 2016.
- “Graf: Generative radiance fields for 3d-aware image synthesis,” 2020, vol. 33.
- “pixelnerf: Neural radiance fields from one or few images,” in CVPR, 2021.
- “Giraffe: Representing scenes as compositional generative neural feature fields,” in CVPR, 2021.
- “Rank-based decomposable losses in machine learning: A survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.