Limited-Angle CBCT Reconstruction via Geometry-Integrated Cycle-domain Denoising Diffusion Probabilistic Models

2025-06-16

AI Summary

In cone-beam CT (CBCT) for radiotherapy, limited-angle acquisition (≤90°) and low-dose protocols induce motion artifacts, structural blurring, and degraded image quality. To address this, we propose the Geometry-Embedded Dual-Domain Cyclic Denoising Diffusion model (LA-GICD). LA-GICD innovatively integrates analytical cone-beam forward/back-projection operators to jointly reconstruct missing projection data and refine anatomical structures in the image domain; cyclic cross-domain mapping, regularized by geometric priors, enables synergistic optimization of data fidelity and anatomical plausibility. Evaluated on 78 clinical CT cases, LA-GICD achieves MAE = 35.5 HU, SSIM = 0.84, and PSNR = 29.8 dB. A single 90° scan suffices to produce images comparable in quality to full-angle acquisitions, reducing scanning time and radiation dose by 75% while significantly enhancing soft-tissue contrast.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Cone-beam CT (CBCT) is widely used in clinical radiotherapy for image-guided treatment, improving setup accuracy, adaptive planning, and motion management. However, slow gantry rotation limits performance by introducing motion artifacts, blurring, and increased dose. This work aims to develop a clinically feasible method for reconstructing high-quality CBCT volumes from consecutive limited-angle acquisitions, addressing imaging challenges in time- or dose-constrained settings. We propose a limited-angle (LA) geometry-integrated cycle-domain (LA-GICD) framework for CBCT reconstruction, comprising two denoising diffusion probabilistic models (DDPMs) connected via analytic cone-beam forward and back projectors. A Projection-DDPM completes missing projections, followed by back-projection, and an Image-DDPM refines the volume. This dual-domain design leverages complementary priors from projection and image spaces to achieve high-quality reconstructions from limited-angle (<= 90 degrees) scans. Performance was evaluated against full-angle reconstruction. Four board-certified medical physicists conducted assessments. A total of 78 planning CTs in common CBCT geometries were used for training and evaluation. The method achieved a mean absolute error of 35.5 HU, SSIM of 0.84, and PSNR of 29.8 dB, with visibly reduced artifacts and improved soft-tissue clarity. LA-GICD's geometry-aware dual-domain learning, embedded in analytic forward/backward operators, enabled artifact-free, high-contrast reconstructions from a single 90-degree scan, reducing acquisition time and dose four-fold. LA-GICD improves limited-angle CBCT reconstruction with strong data fidelity and anatomical realism. It offers a practical solution for short-arc acquisitions, enhancing CBCT use in radiotherapy by providing clinically applicable images with reduced scan time and dose for more accurate, personalized treatments.

Problem

• Reconstruct high-quality CBCT from limited-angle scans
• Reduce motion artifacts and dose in CBCT imaging
• Improve soft-tissue clarity with dual-domain diffusion models

Innovation

• Geometry-integrated dual-domain DDPM framework
• Analytic cone-beam forward and back projectors
• Limited-angle scans with reduced artifacts