Awesome Computed Tomography 中文 Star on GitHub

Diffusion & Generative Priors

Diffusion and generative priors use a model trained to generate realistic CT images, such as a score-based or denoising diffusion model, as the prior in tomographic reconstruction. The generative model captures what plausible anatomy or objects look like, and consistency with the measured projections is enforced during sampling or by alternating with a physics-based solver.

They are applied to sparse-view and limited-angle CT, low-dose denoising, metal artifact reduction and CBCT enhancement. A practical attraction is flexibility: one unconditional model trained on images can be reused for different acquisition settings, because the forward operator only enters at reconstruction time.

Approaches include posterior sampling with gradient guidance, plug-and-play diffusion inside ADMM or other splitting schemes, conditional and bridge models trained on paired data, latent diffusion, and 2D models extended to 3D with cross-slice regularisation. Known challenges are sampling speed, hallucinated structure under severe undersampling and the gap between simulated and measured data.

Browse all 68 Diffusion & Generative Priors papers in the searchable directory

Start here

If you are new to this topic, these are good first reads, chosen by the maintainers.

Papers

The 30 most recent of 68 papers, newest first.

PaperVenueYearLinks
Anatomy-preserving unpaired cone-beam CT refinement for image-guided radiotherapy using pseudo-label guided diffusionpreprint2026paper
CBCT-Based Synthetic CT Generation Using Conditional Flow Matching Modelpreprint2026paper
Conditional Diffusion for 3D CT Volume Reconstruction from 2D X-raysBMVC2026paper · code
Conditional Diffusion Posterior Alignment for Sparse-View CT Reconstructionpreprint2026paper
Cone-Beam CT Image Quality Enhancement Using A Latent Diffusion Model Trained with Simulated CBCT Artifactspreprint2026paper
Continuity-driven Synergistic Diffusion with Neural Priors for Ultra-Sparse-View CBCT Reconstructionpreprint2026paper
Cross-Distribution Diffusion Priors-Driven Iterative Reconstruction for Sparse-View CTTMI2026paper
Cross-Modal Guidance for Fast Diffusion-Based Computed TomographyICASSP2026paper
Plug-and-Play Diffusion Meets ADMM: Dual-Variable Coupling for Robust Medical Image ReconstructionICML2026paper · code · project
DiffNR: Diffusion-Enhanced Neural Representation Optimization for Sparse-View 3D Tomographic ReconstructionAAAI2026paper · project
Regularizing INR with Diffusion Prior for Self-Supervised 3D Reconstruction of Neutron Computed Tomography DataICASSP2026paper
DM4CT: Benchmarking Diffusion Models for CT ReconstructionICLR2026paper · code
Dose-Aware Cold Diffusion with Physics Consistency for Generalizable Low-Dose CT ReconstructionIJCNN2026paper
Field-of-View Extension in Dental Cone-Beam CT via Implicit Neural Representations and Diffusion Model-Based RefinementMICAD2026paper · code
Foundation VAEs for 3D CT Reconstruction, Augmentation, and GenerationICML2026paper · code
From Sparse X-rays to 3D CT: Training-Free Reconstruction with Diffusion PriorsDGM4MICCAI2026paper · code
Geometry-Aware Diffusion Approximate Posterior Sampling for Sparse-View and Limited-Angle CTpreprint2026paper
Human-Guided Shade Artifact Suppression in CBCT-to-MDCT Translation via Schrödinger Bridge with Conditional DiffusionTRPMS2026paper
Improving 2D Diffusion Models for 3D Medical Imaging with Inter-Slice Consistent StochasticityICLR2026paper · code
KLIP: Localized Distribution Shift Detection via KL-Divergence with Diffusion Priors in Inverse ProblemsCVPR2026paper · code
LUCID: Learned Undersampling-Adaptive Consistency-Guided Inference with Deterministic Flow Matching for Sparse-View CT Reconstructionpreprint2026paper
Zero-shot CT Super-Resolution using Diffusion-based 2D Projection Priors and Signed 3D GaussiansMICCAI2026paper
PhyDiCT: Plug-and-Play CT Reconstruction from Sparse X-Rays via Differentiable Rendering and Strong PriorsMICCAI2026paper · code
Spectral Consistent Flow for One-Step 3D Medical Image TranslationECCV2026paper
Toward a Foundation Plug-and-Play Prior for Computed Tomography Reconstruction via a Multimodal Diffusion ModelSC Workshops2026paper
Toward CT-Equivalent Image Quality in Low-Dose Radiotherapy Planning: Conditional Diffusion-Based CBCT-to-CT Synthesis and the Impact of CBCT Input Representationpreprint2026paper
Towards reconstructing experimental sparse-view X-ray CT data with diffusion modelsICIP2026paper
3D CBCT Artefact Removal Using Perpendicular Score-Based Diffusion ModelsDGM4MICCAI2025paper
Adaptive Diffusion Models for Sparse-View Motion-Corrected Head Cone-beam CTTRPMS2025paper · project
Advancing Limited-Angle CT Reconstruction Through Diffusion-Based Sinogram CompletionICIP2025paper

Browse all 68 Diffusion & Generative Priors papers in the searchable directory

Datasets, benchmarks and toolkits

Entries from the directory whose tasks or tags match this topic.

Guides

Guides that cover this topic.

Other topics