A Simulation Framework for X-Ray Phase-Contrast Imaging

X射线相衬成像(XPCI)可以显著提高X射线软组织对比度 医学成像。尽管全世界都在努力开发新的xpci系统，但我们还没有 一种用于严格预测临床xPCI系统性能的数值框架 人体尺度。在本研究中，我们建议开发这样一个仿真框架来完成 这利用了两个主要组成部分的突破：一个逼真的人类尺度的数值 XPCIPhantom和基于波动光学的精确传播X射线波模拟器 通过一个逼真的人类数字幻影。 对于数字体模来说，最大的挑战是将各种器官与多个器官结合在一起。 人体尺寸体模中的比例结构和定义人体的材料属性 各部位处于正常和各种病理生理条件下。为了解决这些问题，我们 将扩展xCAT体模，这是一种在医学成像中广泛使用的人体体模 模拟，结合亚器官结构和组织纹理，并指定适当的 用于XPCI模拟的各种组织的材料属性。 对于模拟器，我们最近演示了在人类身上进行xPCI模拟的可能性 通过将基于波动光学的成像模型应用于xCAT体模来进行缩放。在建议的 项目，我们将开发一个通用的XPCI模拟器，可以用于各种 几何学、X射线光学和组织模型。使用我们已经在 基于同步加速器的高性能XPCI设置，我们将验证模拟器，以便其 预测与实验获得的数据相匹配。最后，我们将使用一个 图形用户界面，以便用户可以方便地模拟自己的XPCI系统。 源代码和详细的用户手册也将分发。

X-ray phase-contrast imaging (XPCI) can dramatically improve soft tissue contrast in X-ray medical imaging. Despite worldwide efforts to develop novel XPCI systems, we do not yet have a numerical framework to rigorously predict the performance of a clinical XPCI system at a human scale. In this study, we propose to develop such a simulation framework to accomplish this using the breakthroughs in two main components: a realistic human-scaled numerical phantom for XPCI, and a wave optics-based simulator for accurately propagating X-ray wave through a realistic human numerical phantom. For the numerical phantom, the biggest challenges are to incorporate various organs with multi- scale structures in a human-size phantom and to define material properties for human body parts under normal and a variety of pathophysiological conditions. To address these issues, we will extend the XCAT phantom, a human phantom that is widely used in medical imaging simulation, to incorporate sub-organ structures and tissue textures, and to assign appropriate material properties to various tissues for XPCI simulation. For the simulator, we have recently demonstrated the possibility of XPCI simulation at a human scale by applying a wave optics-based imaging model to XCAT phantom. In the proposed project, we will develop a general-purpose XPCI simulator that can be used with variety of geometries, X-ray optics, and tissue models. Using the data we have already acquired on a synchrotron-based, high-performance XPCI setup, we will validate the simulator so that its predictions match experimentally acquired data. Finally, we will distribute the program with a graphical user interface in order that the users can easily simulate their own XPCI systems. Source codes and a detailed user manual will be distributed as well.