EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

EpiC-NeRF: Epistemic Uncertainty-Guided Neural Radiance Fields for Sparse-View CT Reconstruction

Donghyuk Choo, Haill An, Younhyun Jung

Mathematics · 2026-08-04 · 14 (15) · 2802

DOI: 10.3390/math14152802 ↗

At a glance

Journal article in Mathematics, published online on 2026-08-04. Written by Donghyuk Choo and 2 co-authors. 36 references. The full text is freely available (open access, CC BY 4.0). Funded by: Korean government, Gachon University.

AstrattDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Sparse-view computed tomography (CT) reconstruction aims to recover high-quality CT volumes from a limited number of X-ray projection images, thereby reducing radiation exposure during image acquisition. However, this problem is inherently ill-posed because each projection provides only indirect line-integral supervision, and different attenuation distributions can explain similar sparse measurements. Existing analytic and iterative methods often suffer from streak artifacts and unstable solutions, while supervised learning-based methods require paired training data and may generalize poorly across anatomical regions or acquisition settings. Neural Radiance Field (NeRF)-based methods have recently shown promise by representing the attenuation field as a continuous coordinate-based function optimized directly from projection images. Nevertheless, these methods mainly enforce projection consistency and do not explicitly use volume-domain uncertainty to guide subsequent reconstruction. In this work, we propose EpiC-NeRF, a CT-specific closed-loop framework that actively feeds estimated epistemic uncertainty back into sparse-view reconstruction. EpiC-NeRF adapts evidential uncertainty estimation and aggregation to the X-ray CT line-integral formulation and maintains the resulting spatial uncertainty in a persistent three-dimensional Epistemic Grid Map. The accumulated uncertainty is used by Epistemic-Adaptive Layer Normalization to modulate intermediate features and by dual active sampling to guide ray- and point-level sample allocation. The newly estimated uncertainty then updates the grid map and guides subsequent optimization iterations, forming a unified feedback loop between uncertainty estimation and CT reconstruction. Experiments on four CT volume datasets demonstrate that EpiC-NeRF achieves improved reconstruction fidelity over existing analytic, iterative, and neural implicit reconstruction methods.

Funding

  • Korean government
  • Gachon University

Cite

APA

Choo, D., An, H., & Jung, Y. (2026). EpiC-NeRF: Epistemic Uncertainty-Guided Neural Radiance Fields for Sparse-View CT Reconstruction. Mathematics, 14(15), 2802. https://doi.org/10.3390/math14152802

Vancouver

Choo D, An H, Jung Y. EpiC-NeRF: Epistemic Uncertainty-Guided Neural Radiance Fields for Sparse-View CT Reconstruction. Mathematics. 2026;14(15):2802. doi:10.3390/math14152802

BibTeX
@article{Choo2026,
  author = {Choo, Donghyuk and An, Haill and Jung, Younhyun},
  title = {EpiC-NeRF: Epistemic Uncertainty-Guided Neural Radiance Fields for Sparse-View CT Reconstruction},
  journal = {Mathematics},
  year = {2026},
  volume = {14},
  number = {15},
  pages = {2802},
  doi = {10.3390/math14152802}
}
Start for free →