DuDoTrans: Dual-Domain Transformer for Sparse-View CT Reconstruction
DuDoTrans: Dual-Domain Transformer for Sparse-View CT Reconstruction
复制标题
DuDoTrans：用于稀疏视图 CT 重建的双域变压器
DOI：
10.1007/978-3-031-17247-2_9
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
S. K. Zhou
中科院分区：
文献类型：
--
作者：
Ce Wang;Kun Shang;Haimiao Zhang;Qian Li;S. K. Zhou
虽然CT对于临床诊断是必要的，但成像过程中的电离辐射会导致不可逆的损伤，从而促使研究人员研究稀疏CT重建。为了减少稀疏视角CT图像中出现的伪影，人们提出了迭代模型，但迭代模型的计算量较大。基于深度学习的方法以其良好的重建性能和计算效率得到了广泛的应用。然而，这些方法忽略了CNN的局部特征提取能力和正弦图的全局特征之间的不匹配。为了克服这一问题，我们提出了双域变换(DuDoTrans)，利用Transformer的远程依赖建模能力，同时恢复信息丰富的正弦图像，并用增强后的正弦图像和原始的正弦图像重建CT图像。在NIH-AAPM和新冠肺炎数据集上的重建结果表明，这种设计新颖的DuDoTrans方法，即使涉及的参数更少，也比竞争对手的方法更有效、更具泛化能力。最后，通过实验验证了该算法对噪声的鲁棒性。
While Computed Tomography (CT) is necessary for clinical diagnosis, ionizing radiation in the imaging process induces irreversible injury, thereby driving researchers to study sparse-view CT reconstruction. Iterative models are proposed to alleviate the appeared artifacts in sparse-view CT images, but their computational cost is expensive. Deep-learning-based methods have gained prevalence due to the excellent reconstruction performances and computation efficiency. However, these methods ignore the mismatch between the CNN’slocalfeature extraction capability and the sinogram’sglobalcharacteristics. To overcome the problem, we proposeDual-DomainTransformer (DuDoTrans) to simultaneously restore informative sinograms via the long-range dependency modeling capability of Transformer and reconstruct CT image with both the enhanced and raw sinograms. With such a novel design, DuDoTrans even with fewer involved parameters is more effective and generalizes better than competing methods, which is confirmed by reconstruction performances on the NIH-AAPM and COVID-19 datasets. Finally, experiments also demonstrate its robustness to noise.