A CBCT sparse-angle reconstruction method and device based on a physical perception residual-guided diffusion model
By embedding a residual-guided diffusion model with a physical sensing module in CBCT sparse angle reconstruction, the problems of image artifacts and detail loss under sparse angles are solved, achieving high-quality, physically consistent reconstruction results, which are applicable to multiple medical and industrial inspection scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG LAB
- Filing Date
- 2026-06-11
- Publication Date
- 2026-07-17
AI Technical Summary
Existing sparse angle CBCT reconstruction algorithms are insufficient in terms of image artifacts and detail loss, making it difficult to meet the needs of clinical diagnosis and industrial testing. Traditional diffusion models lack physical imaging priors, leading to inconsistent reconstruction results.
A residual-guided diffusion model is constructed, embedding a back-projection-based physical perception module. By introducing physical processes such as projection geometry and X-ray attenuation into the diffusion model and training it with a multi-time-step loss function, image reconstruction under physical constraints is achieved.
It improves the anatomical rationality and numerical accuracy of reconstructed images, increases reconstruction efficiency, and meets the needs of clinical diagnosis and industrial testing.
Smart Images
Figure CN122415792A_ABST