A Constrained Optimization Reconstruction Model for X-ray Computed Tomography Metal Artifact Suppression
A Constrained Optimization Reconstruction Model for X-ray Computed Tomography Metal Artifact Suppression
复制标题
X射线计算机断层扫描金属伪影抑制的约束优化重建模型
DOI：
10.1166/jmihi.2015.1556
复制
发表时间：
2015-11-01
影响因子：
--
通讯作者：
Wu, Zhongyi
中科院分区：
文献类型：
--
作者：
Li, Ming;Zheng, Jian;Wu, Zhongyi
在计算机断层扫描中，金属物体引起的金属伪影会影响重建图像的质量。在本文中，我们开发了一种新的迭代金属伪影抑制（MAS）的方法，通过最小化建议的重建模型。它包括一个预定义的函数，调节图像梯度稀疏，同时投影数据保真度和图像非负性约束。建议MAS方法包括三个步骤的过程。首先，通过区域生长识别图像域中的金属植入物，并通过二值金属图像的前向投影确定原始投影中对应的金属痕迹。然后利用该重建模型从排除金属痕迹的投影数据中重建出无金属背景图像。最后，通过在背景图像上施加金属物体来产生校正图像。一系列的体模和临床研究的实施，以验证所提出的方法的性能。实验结果表明，该方法能够有效地抑制金属伪影，降低噪声，提高解剖结构的可见性。虽然所提出的方法提供了一个可行的策略来抑制金属伪影，它也可以扩展到处理其他的CT图像重建中的“丢失数据”的问题。
In computed tomography, metal artifacts caused by metallic objects will deteriorate the reconstructed image quality. In this paper, we develop a novel iterative metal artifact suppression (MAS) approach by minimizing the proposed reconstruction model. It includes a predefined function that regulates image gradient sparseness, with projection data fidelity and image nonnegativity constraints at the same time. The proposed MAS method incorporates three-step process. First, the metallic implants in the image domain are identified by region growing and the corresponding metal traces in the raw projections are determined through forward projection of the binary metal image. Then the metal-free background image is reconstructed from the metal-trace-excluded projection data using the proposed reconstruction Model. Finally, the corrected image is produced via imposing metallic objects on the background image. A series of phantom and clinical studies are implemented to verify the performance of the proposed method. The results demonstrate that the proposed method can successfully suppress metal artifacts, reduce noise and improve anatomical structure visibility. Although the proposed method offers a feasible strategy for suppressing metal artifacts, it can also be extended to deal with other "missing data" problems in CT image reconstruction.