Luo Shouhua, Wu Huazhen, Sun Yi, Li Jing, Li Guang, Gu Ning
Southeast University, Nanjing, People's Republic of China.
Phys Med Biol. 2017 Mar 7;62(5):1810-1830. doi: 10.1088/1361-6560/aa56b5. Epub 2017 Jan 4.
The beam hardening effect can induce strong artifacts in CT images, which result in severely deteriorated image quality with incorrect intensities (CT numbers). This paper develops an effective and efficient beam hardening correction algorithm incorporated in a filtered back-projection based maximum a posteriori (BHC-FMAP). In the proposed algorithm, the beam hardening effect is modeled and incorporated into the forward-projection of the MAP to suppress beam hardening induced artifacts, and the image update process is performed by Feldkamp-Davis-Kress method based back-projection to speed up the convergence. The proposed BHC-FMAP approach does not require information about the beam spectrum or the material properties, or any additional segmentation operation. The proposed method was qualitatively and quantitatively evaluated using both phantom and animal projection data. The experimental results demonstrate that the BHC-FMAP method can efficiently provide a good correction of beam hardening induced artefacts.
束硬化效应会在CT图像中产生强烈伪影，导致图像质量严重下降，出现强度（CT值）错误。本文开发了一种有效且高效的束硬化校正算法，该算法融入了基于滤波反投影的最大后验算法（BHC - FMAP）。在所提出的算法中，束硬化效应被建模并纳入到最大后验算法的前向投影中，以抑制束硬化引起的伪影，图像更新过程通过基于费尔德坎普 - 戴维斯 - 克雷斯方法的反投影来执行，以加速收敛。所提出的BHC - FMAP方法不需要关于束光谱或材料特性的信息，也不需要任何额外的分割操作。使用体模和动物投影数据对所提出的方法进行了定性和定量评估。实验结果表明，BHC - FMAP方法能够有效地对束硬化引起的伪影进行良好校正。
