mlsanityData to 14 Oct 2026

DeepSparse: A Foundation Model for Sparse-View CBCT Reconstruction

arXiv 2505.02628 ↗First posted 5 May 2025eess.IV cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

Cone-beam computed tomography (CBCT) is a critical 3D imaging technology in the medical field, while the high radiation exposure required for high-quality imaging raises significant concerns, particularly for vulnerable populations. Sparse-view reconstruction reduces radiation by using fewer X-ray projections while maintaining image quality, yet existing methods face challenges such as high computational demands and poor generalizability to different datasets. To overcome these limitations, we propose DeepSparse, the first foundation model for sparse-view CBCT reconstruction, featuring DiCE (Dual-Dimensional Cross-Scale Embedding), a novel network that integrates multi-view 2D features and multi-scale 3D features. Additionally, we introduce the HyViP (Hybrid View Sampling Pretraining) framework, which pretrains the model on large datasets with both sparse-view and dense-view projections, and a two-step finetuning strategy to adapt and refine the model for new datasets. Extensive experiments and ablation studies demonstrate that our proposed DeepSparse achieves superior reconstruction quality compared to state-of-the-art methods, paving the way for safer and more efficient CBCT imaging.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 33 of its 65 references that are in the map. First posted on arXiv 5 May 2025.

Builds on

33 of 65 references.

References · 33

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2016Densely Connected Convolutional Networkscited by 6,720
  3. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
  4. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  5. 2021Solving Inverse Problems in Medical Imaging with Score-Based Generative Modelscited by 521
  6. 2016Deep Convolutional Neural Network for Inverse Problems in Imagingcited by 433
  7. 2022BBDM: Image-to-image Translation with Brownian Bridge Diffusion Modelscited by 225
  8. 2016Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challengecited by 201
  9. 2021NeRP: Implicit Neural Representation Learning with Prior Embedding for Sparsely Sampled Image Reconstructioncited by 146
  10. 2022Solving 3D Inverse Problems using Pre-trained 2D Diffusion Modelscited by 132
  11. 2022NAF: Neural Attenuation Fields for Sparse-View CBCT Reconstructioncited by 108
  12. 2019X2CT-GAN: Reconstructing CT from Biplanar X-Rays with Generative Adversarial Networkscited by 85
21 more
  1. 2024A Vision-Language Foundation Model to Enhance Efficiency of Chest X-ray Interpretationcited by 81
  2. 2019DuDoNet: Dual Domain Network for CT Metal Artifact Reductioncited by 68
  3. 2024R2-Gaussian: Rectifying Radiative Gaussian Splatting for Tomographic Reconstructioncited by 67
  4. 2023AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weekscited by 52
  5. 2023Learning Deep Intensity Field for Extremely Sparse-View CBCT Reconstructioncited by 42
  6. 2018Radon Inversion via Deep Learningcited by 35
  7. 2023Enhancing Representation in Radiography-Reports Foundation Model: A Granular Alignment Algorithm Using Masked Contrastive Learningcited by 34
  8. 2024Towards a Holistic Framework for Multimodal Large Language Models in Three-dimensional Brain CT Report Generationcited by 31
  9. 2024EVA-X: A Foundation Model for General Chest X-ray Analysis with Self-supervised Learningcited by 25
  10. 2024Interpretable Bilingual Multimodal Large Language Model for Diverse Biomedical Taskscited by 24
  11. 2025Triad: Vision Foundation Model for 3D Magnetic Resonance Imagingcited by 24
  12. 2022SNAF: Sparse-view CBCT Reconstruction with Neural Attenuation Fieldscited by 23
  13. 2016Deep Residual Learning for Compressed Sensing CT Reconstruction via Persistent Homology Analysiscited by 22
  14. 2023FreeSeed: Frequency-band-aware and Self-guided Network for Sparse-view CT Reconstructioncited by 21
  15. 2023Perspective Projection-Based 3D CT Reconstruction from Biplanar X-rayscited by 18
  16. 2024Learning 3D Gaussians for Extremely Sparse-View Cone-Beam CT Reconstructioncited by 18
  17. 2024C^2RV: Cross-Regional and Cross-View Learning for Sparse-View CBCT Reconstructioncited by 15
  18. 2021DuDoTrans: Dual-Domain Transformer Provides More Attention for Sinogram Restoration in Sparse-View CT Reconstructioncited by 12
  19. 2024Imaging foundation model for universal enhancement of non-ideal measurement CTcited by 6
  20. 2025Open Foundation Models in Healthcare: Challenges, Paradoxes, and Opportunities with GenAI Driven Personalized Prescriptioncited by 2
  21. 2023Self-supervised Learning for Enhancing Geometrical Modeling in 3D-Aware Generative Adversarial Networkcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 3.

5 papers.

  1. 2025X-GRM: Large Gaussian Reconstruction Model for Sparse-view X-rays to Computed Tomographycited by 6
  2. 2025Cross-view Generalized Diffusion Model for Sparse-view CT Reconstructioncited by 6
  3. 2026ILV: Iterative Latent Volumes for Fast and Accurate Sparse-View CT Reconstructioncited by 0
  4. 2026LUCID: Learned Undersampling-Adaptive Consistency-Guided Inference with Deterministic Flow Matching for Sparse-View CT Reconstructioncited by 0
  5. 2026RePrior: Measurement-Reconciled Prior Correction for Sparse-View CBCT Reconstructioncited by 0