Park Hyoung Suk, Hwang Dosik, Seo Jin Keun
IEEE Trans Med Imaging. 2016 Feb;35(2):480-7. doi: 10.1109/TMI.2015.2478905. Epub 2015 Sep 15.
This paper proposes a new method to correct beam hardening artifacts caused by the presence of metal in polychromatic X-ray computed tomography (CT) without degrading the intact anatomical images. Metal artifacts due to beam-hardening, which are a consequence of X-ray beam polychromaticity, are becoming an increasingly important issue affecting CT scanning as medical implants become more common in a generally aging population. The associated higher-order beam-hardening factors can be corrected via analysis of the mismatch between measured sinogram data and the ideal forward projectors in CT reconstruction by considering the known geometry of high-attenuation objects. Without prior knowledge of the spectrum parameters or energy-dependent attenuation coefficients, the proposed correction allows the background CT image (i.e., the image before its corruption by metal artifacts) to be extracted from the uncorrected CT image. Computer simulations and phantom experiments demonstrate the effectiveness of the proposed method to alleviate beam hardening artifacts.
本文提出了一种新方法，用于校正多色X射线计算机断层扫描（CT）中因金属存在而导致的束硬化伪影，同时不会降低完整解剖图像的质量。由于X射线束的多色性导致的束硬化引起的金属伪影，随着医疗植入物在老龄化人口中越来越普遍，正成为影响CT扫描的一个日益重要的问题。通过考虑高衰减物体的已知几何形状，分析测量的正弦图数据与CT重建中理想前向投影器之间的不匹配，可以校正相关的高阶束硬化因子。在无需光谱参数或能量依赖衰减系数的先验知识的情况下，所提出的校正方法能够从未校正的CT图像中提取背景CT图像（即未被金属伪影破坏之前的图像）。计算机模拟和体模实验证明了所提方法减轻束硬化伪影的有效性。
