Wu Jia, Lin Jinzhao, Jiang Xiaoming, Zheng Wei, Zhong Lisha, Pang Yu, Meng Hongying, Li Zhangyong
School of Communications and Information Engineering, Chongqing University of Posts and Telecommunications, 400065, Chongqing, China.
School of Medical Information and Engineering, Southwest Medical University, 646000, Luzhou, China.
Sci Rep. 2025 May 15;15(1):16894. doi: 10.1038/s41598-025-02133-5.
Sparse-view CT reconstruction is a challenging ill-posed inverse problem, where insufficient projection data leads to degraded image quality with increased noise and artifacts. Recent deep learning approaches have shown promising results in CT reconstruction. However, existing methods often neglect projection data constraints and rely heavily on convolutional neural networks, resulting in limited feature extraction capabilities and inadequate adaptability. To address these limitations, we propose a Dual-domain deep Prior-guided Multi-scale fusion Attention (DPMA) model for sparse-view CT reconstruction, aiming to enhance reconstruction accuracy while ensuring data consistency and stability. First, we establish a residual regularization strategy that applies constraints on the difference between the prior image and target image, effectively integrating deep learning-based priors with model-based optimization. Second, we develop a multi-scale fusion attention mechanism that employs parallel pathways to simultaneously model global context, regional dependencies, and local details in a unified framework. Third, we incorporate a physics-informed consistency module based on range-null space decomposition to ensure adherence to projection data constraints. Experimental results demonstrate that DPMA achieves improved reconstruction quality compared to existing approaches, particularly in noise suppression, artifact reduction, and fine detail preservation.
稀疏视图CT重建是一个具有挑战性的不适定逆问题，其中投影数据不足会导致图像质量下降，噪声和伪影增加。最近的深度学习方法在CT重建中显示出了有前景的结果。然而，现有方法往往忽视投影数据约束，严重依赖卷积神经网络，导致特征提取能力有限且适应性不足。为了解决这些局限性，我们提出了一种用于稀疏视图CT重建的双域深度先验引导多尺度融合注意力（DPMA）模型，旨在提高重建精度，同时确保数据的一致性和稳定性。首先，我们建立了一种残差正则化策略，对先验图像和目标图像之间的差异施加约束，有效地将基于深度学习的先验与基于模型的优化相结合。其次，我们开发了一种多尺度融合注意力机制，该机制采用并行路径在统一框架中同时对全局上下文、区域依赖性和局部细节进行建模。第三，我们纳入了一个基于范围零空间分解的物理信息一致性模块，以确保遵守投影数据约束。实验结果表明，与现有方法相比，DPMA实现了更高的重建质量，特别是在噪声抑制、伪影减少和精细细节保留方面。
