TR&D Project 1: Virtual Patients

TR

• 批准号: 10551841
• 负责人: William P Segars
• 金额: $ 36.93万
• 依托单位: DUKE UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-04-01 至 2025-12-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10551841
• 关键词: 3-Dimensional; Adult; Age; Anatomy; Blood flow; Brain; Cardiac; Characteristics; Childhood; Clinical; Computed Tomography Scanners; Computers; Contrast Media; Data; Databases; Development; Disease; Dose; Evaluation; Exclusion; Exhibits; Future; Heart; Heart Diseases; Height; Heterogeneity; Hospitals; Image; Image Analysis; Imaging Device; Imaging technology; Individual; Intravenous; Kidney; Laboratories; Lesion; Libraries; Liquid substance; Liver; Lung; Magnetic Resonance Imaging; Malignant Neoplasms; Mathematics; Medical Imaging; Methods; Modality; Modeling; Motion; Normal tissue morphology; Organ; Paper; Patients; Performance; Perfusion; Phase; Physiological; Population; Pregnancy; Protocols documentation; PubMed; Race; Reader; Reading; Research; Research Personnel; Research Project Grants; Resources; Series; Source; Technology; Texture; Time; Tissue Model; Tissues; Universities; Update; Variant; Weight; Work; X-Ray Computed Tomography; clinical application; clinically relevant; computerized; cost; design; experimental study; human model; imaging approach; individual patient; innovation; patient population; phantom model; physiologic model; radiation absorbed dose; respiratory; sex; simulation; tool; virtual; virtual imaging; virtual patient; virtual platform

ABSTRACT – TRD1: Virtual Patients The broad objective of the Center for Virtual Imaging Trials is to develop and provide a comprehensive virtual platform for assessment of a medical imaging technology or clinical application from design to use, initially in Computed Tomography (CT). Virtual imaging trials, consisting of computerized patients, scanner models, and image analysis methods, provide an efficient means with which to determine the most effective and optimized use of emerging medical imaging technologies and approaches. Experiments can be conducted quickly and cost effectively on a computer with known ground truth (exact patient anatomy and physiological conditions are known) and with complete user control of the virtual imaging device and protocol. With the growing use of CT, the need for such a toolset is at an all-time high to optimize image quality versus dose, personalizing technologies to the characteristics of individual patients. Virtual imaging trials require a realistic virtual patient population. The team previously developed a series of 60 adult and 90 pediatric 4D whole-body computational human models known as XCAT phantoms (extended cardiac-torso). The phantoms have found wide use in numerous research projects both in the team's laboratory and by others, but they are limited in that they do not cover the full range of patient anatomies indicative of the population at large, including pregnancy, and lack sufficient realism in terms of anatomical textures, motion, blood flow, and contrast perfusion. All of these factors are significant determinants of image quality as well as absorbed radiation dose. This work of TRD1 will greatly expand the coverage of the XCAT library, creating a population of whole-body 4D phantoms to represent combinations of sex, age, race, height, and weight including anatomically variable models for stages of pregnancy. Further, TRD1 will update the XCAT motion models, incorporate blood flow to simulate intra-organ contrast perfusion within vital organs, and develop additional models for the tissue heterogeneity within the body including disease-related abnormalities. Such innovations greatly enhance the realism of the phantoms, giving researchers the ability to generate patient-quality CT data in order to assess image quality as well as dose in the presence of motion and contrast. The tools will be packaged into a framework from which researchers will be able to generate virtual populations, on demand, morphing between the templates to create 1000s of realistic phantoms with user-defined anatomical characteristics, lesions, physiologic motion, and contrast. The framework to produce realistic, vast populations of customizable virtual subjects will be combined with the Center's virtual scanners (TRD2) and virtual readers (TRD3) components. Through these advances, the proposed Center will have a significant impact on the development, evaluation, and optimization of clinically informed 3D and 4D CT imaging technologies with potential future expansion to other modalities.

摘要-TRD 1：虚拟患者 虚拟成像试验中心的广泛目标是开发和提供一个全面的虚拟 用于评估从设计到使用的医学成像技术或临床应用的平台，最初在 计算机断层扫描（CT）。虚拟成像试验，包括计算机化的患者，扫描仪模型， 图像分析方法，提供了一种有效的手段，以确定最有效和最优化的 使用新兴的医学成像技术和方法。实验可以快速进行， 在具有已知基础事实（精确的患者解剖结构和生理状况）的计算机上经济有效地 是已知的）并且具有对虚拟成像设备和协议的完全用户控制。随着越来越多地使用 CT，对这样一个工具集的需求是在历史上最高的，以优化图像质量与剂量，个性化 根据患者个体的特点。 虚拟成像试验需要真实的虚拟患者人群。该团队此前开发了一系列 60个成人和90个儿科4D全身计算人体模型，称为XCAT体模（扩展 心脏-躯干）。幻影已经在许多研究项目中得到了广泛的应用， 实验室和其他人，但他们是有限的，因为他们不涵盖病人解剖的全部范围 这表明人口普遍，包括怀孕，在解剖学方面缺乏足够的现实性。 纹理、运动、血流和造影剂灌注。所有这些因素都是形象的重要决定因素 质量以及吸收的辐射剂量。 TRD 1的这项工作将大大扩大XCAT库的覆盖范围，创造一个全身的群体。 4D体模代表性别、年龄、种族、身高和体重的组合，包括解剖学变量 怀孕阶段的模型。此外，TRD 1将更新XCAT运动模型，纳入血流 模拟重要器官内的器官内造影剂灌注，并为组织开发额外的模型 体内的异质性，包括疾病相关的异常。这些创新极大地提高了 幻象的真实性，使研究人员能够生成患者质量的CT数据，以评估 图像质量以及存在运动和对比度时的剂量。这些工具将打包到一个 框架，研究人员将能够生成虚拟人口，按需，变形之间 这些模板可创建1000个具有用户定义的解剖特征、病变 生理运动和对比度。 产生逼真的、大量可定制虚拟主题的框架将与 中心的虚拟扫描仪（TRD 2）和虚拟阅读器（TRD 3）组件。通过这些进步， 拟建中心将对临床的开发、评估和优化产生重大影响 3D和4D CT成像技术，未来可能扩展到其他模式。