Learned Alternating Minimization Algorithm for Dual-domain Sparse-View CT Reconstruction
Learned Alternating Minimization Algorithm for Dual-domain Sparse-View CT Reconstruction
复制标题
DOI：
10.48550/arxiv.2306.02644
复制
发表时间：
2023-06
期刊：
影响因子：
--
通讯作者：
Chi-Jiao Ding;Qingchao Zhang;Ge Wang;X. Ye;Yunmei Chen
中科院分区：
文献类型：
--
作者：
Chi-Jiao Ding;Qingchao Zhang;Ge Wang;X. Ye;Yunmei Chen
提出了一种新的学习交替最小化算法（LAMA）用于双域稀疏CT图像重建。LAMA是自然引起的CT重建与可学习的非光滑非凸正则化，这是参数化的图像和正弦图域的深度网络的复合功能的变分模型。为了最小化模型的目标，我们将平滑技术和残差学习架构的LAMA的设计。我们表明，LAMA大大降低了网络的复杂性，提高了内存效率和重建精度，并证明收敛可靠的重建。大量的数值实验表明，LAMA优于现有的方法在多个基准CT数据集上的一个很大的保证金。
We propose a novel Learned Alternating Minimization Algorithm (LAMA) for dual-domain sparse-view CT image reconstruction. LAMA is naturally induced by a variational model for CT reconstruction with learnable nonsmooth nonconvex regularizers, which are parameterized as composite functions of deep networks in both image and sinogram domains. To minimize the objective of the model, we incorporate the smoothing technique and residual learning architecture into the design of LAMA. We show that LAMA substantially reduces network complexity, improves memory efficiency and reconstruction accuracy, and is provably convergent for reliable reconstructions. Extensive numerical experiments demonstrate that LAMA outperforms existing methods by a wide margin on multiple benchmark CT datasets.