A Diffusion-Based Generative Prior Approach to Sparse-view Computed Tomography

2026-02-11

AI Summary

This work proposes a deep generative prior (DGP) framework that integrates diffusion generative models with iterative optimization to address artifacts and distortions in X-ray CT reconstruction under sparse-view or limited-angle conditions. By incorporating a diffusion model into the DGP formulation and co-designing the image generation process, network architecture, and optimization strategy, the method substantially enhances reconstruction quality under extremely sparse sampling. The approach effectively balances the expressive power of generative models with the interpretability of iterative optimization, enabling high-fidelity CT image reconstruction while significantly suppressing artifacts and structural distortions in highly undersampled scenarios.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The reconstruction of X-rays CT images from sparse or limited-angle geometries is a highly challenging task. The lack of data typically results in artifacts in the reconstructed image and may even lead to object distortions. For this reason, the use of deep generative models in this context has great interest and potential success. In the Deep Generative Prior (DGP) framework, the use of diffusion-based generative models is combined with an iterative optimization algorithm for the reconstruction of CT images from sinograms acquired under sparse geometries, to maintain the explainability of a model-based approach while introducing the generative power of a neural network. There are therefore several aspects that can be further investigated within these frameworks to improve reconstruction quality, such as image generation, the model, and the iterative algorithm used to solve the minimization problem, for which we propose modifications with respect to existing approaches. The results obtained even under highly sparse geometries are very promising, although further research is clearly needed in this direction.

Problem

• sparse-view CT
• image reconstruction
• limited-angle tomography
• artifacts
• data scarcity

Innovation

• diffusion-based generative model
• sparse-view CT reconstruction
• deep generative prior
• iterative optimization
• sinogram