CG-GLORE：一种用于稀疏视角CT重建的基于共轭梯度的全局-局部正则化网络
CG-GLORE: A Conjugate Gradient-Based Global-Local Regularization Network for Sparse-View CT Reconstruction
浏览论文内容
中文总结 AI 辅助
针对稀疏视角CT重建的条纹伪影问题，提出基于二阶优化的CG-GLORE框架，结合GLORE网络建模图像先验，在AAPM和DeepLesion数据集上取得更优重建效果。
中文摘要 AI 辅助
稀疏视角计算机断层扫描（CT）通过采集更少的投影视图来减少辐射剂量，但由此产生的逆问题高度不适定，常会产生严重的条纹伪影。现有的深度重建方法已取得良好性能，但许多方法依赖一阶更新或大型正则化网络，在病态环境中效果可能较差。我们提出CG-GLORE，这是一种受二阶优化启发的紧凑深度展开框架，用于稀疏视角CT重建。每个展开阶段使用基于结构化海森近似的共轭梯度（CG）求解线性系统：它保留了数据保真项中由物理诱导的曲率，同时对学习到的正则化项使用恒等近似。因此，该方法受二阶启发，而非针对完整学习目标的精确牛顿法。为建模图像先验，我们设计了全局-局部正则化网络（GLORE），它结合了卷积局部特征提取与基于稀疏分块和Nyström注意力的长程依赖表示模块。该设计在保持实际复杂度的同时，能捕捉解剖细节和非局部依赖关系。在AAPM和DeepLesion数据集上，针对多种稀疏视角和噪声设置的实验表明，与代表性重建方法相比，CG-GLORE实现了优异的定量性能、稳定的收敛性、更低的噪声功率以及更高的视觉保真度。
英文摘要
Sparse-view computed tomography (CT) reduces radiation dose by acquiring fewer projection views, but the resulting inverse problem is highly ill-posed and often produces severe streak artifacts. Existing deep reconstruction methods have achieved promising performance, yet many rely on first-order updates or large regularization networks, which can be less effective in ill-conditioned settings. We propose \textbf{CG-GLORE}, a compact deep unrolling framework inspired by second-order optimization for sparse-view CT reconstruction. Each unrolled stage uses a CG-solved linear system based on a structured Hessian surrogate: it retains the physics-induced curvature of the data-fidelity term while using an identity approximation for the learned regularization term. Thus, the method is second-order-inspired rather than an exact Newton method for the full learned objective. To model image priors, we design a Global-Local Regularization Network (GLORE), which combines convolutional local feature extraction with a Long-Range Dependency Representation module based on sparse patchification and Nyström attention. This design captures anatomical details and non-local dependencies while maintaining practical complexity. Experiments on AAPM and DeepLesion under multiple sparse-view and noise settings show that CG-GLORE achieves strong quantitative performance, stable convergence, lower noise power, and improved visual fidelity compared with representative reconstruction methods.
发表机构
- Nara Institute of Science and Technology（奈良科学技术研究所）
- University of Information Technology Ho Chi Minh City（胡志明市信息科技大学）
- Vietnam National University Ho Chi Minh City（胡志明市越南国家大学）
- CY Cergy Paris University（塞尔吉-蓬图瓦兹大学）
- ENSEA（法国高等电子与数字技术学院）
- CNRS（法国国家科学研究中心）
- Vietnam Institute for Advanced Study in Mathematics（越南高等数学研究院）
机构由 AI 辅助整理，请以论文原文为准。
