Segmentation-free statistical image reconstruction for polyenergetic x-ray computed tomography with experimental validation
Segmentation-free statistical image reconstruction for polyenergetic x-ray computed tomography with experimental validation
复制标题
DOI：
10.1088/0031-9155/48/15/314
复制
发表时间：
2003-08-07
影响因子：
3.5
通讯作者：
Fessler, JA
中科院分区：
文献类型：
--
作者：
Elbakri, IA;Fessler, JA
本文描述了一种用于X射线CT的统计图像重建方法，该方法基于一个物理模型，该模型考虑了多能X射线源能谱和由能量相关衰减引起的测量非线性。与我们以前的工作不同，提出的算法不需要将对象预先分割成各种组织类别(例如，骨骼和软组织)，并且允许混合像素。每个体素的衰减系数被建模为其未知密度和依赖于能量的质量衰减系数的加权和的乘积。我们建立了该多能量模型的惩罚似然函数，并开发了一种迭代算法来估计每个体素的未知密度。将该方法应用于包含骨和软组织的物体的模拟X射线CT测量，与传统的射束硬化校正方法相比，产生的图像具有显著减少的射束硬化伪影。我们还将该方法应用于从包含不同浓度的磷酸二氢钾溶液的体模获得的真实数据。该算法以不同浓度的准确密度值重建图像，展示了其在定量CT应用中的潜力。
This paper describes a statistical image reconstruction method for x-ray CT that is based on a physical model that accounts for the polyenergetic x-ray source spectrum and the measurement nonlinearities caused by energy-dependent attenuation. Unlike our earlier work, the proposed algorithm does not require pre-segmentation of the object into the various tissue classes (e.g., bone and soft tissue) and allows mixed pixels. The attenuation coefficient of each voxel is modelled as the product of its unknown density and a weighted sum of energy-dependent mass attenuation coefficients. We formulate a penalized-likelihood function for this polyenergetic model and develop an iterative algorithm for estimating the unknown density of each voxel. Applying this method to simulated x-ray CT measurements of objects containing both bone and soft tissue yields images with significantly reduced beam hardening artefacts relative to conventional beam hardening correction methods. We also apply the method to real data acquired from a phantom containing various concentrations of potassium phosphate solution. The algorithm reconstructs an image with accurate density values for the different concentrations, demonstrating its potential for quantitative CT applications.