Patient-specific, high-sensitivity spectral CT for assessment of pancreatic cancer

用于评估胰腺癌的患者特异性高灵敏度能谱 CT

• 批准号: 10491791
• 负责人: Jianan Grace Gang
• 金额: $ 61.38万
• 依托单位: UNIVERSITY OF PENNSYLVANIA
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-22 至 2025-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10491791
• 关键词: 3D Print; Achievement; Algorithms; Anatomy; Automobile Driving; Biological Markers; Biological Products; Calibration; Characteristics; Clinical; Clinical Trials; Complement; Complex; Computed Tomography Scanners; Computer software; Contrast Media; Data; Dependence; Detection; Diagnosis; Diagnostic; Disease; Disease Management; Enhancing Lesion; Evaluation; Functional Imaging; Goals; Hybrids; Image; Imaging Techniques; Iodine; Joints; Lead; Lesion; Magnetic Resonance Imaging; Malignant neoplasm of pancreas; Measurement; Measures; Methods; Modeling; Morphology; Noise; Organ; Outcome; Pancreas; Patients; Pattern; Performance; Perfusion; Physics; Property; Protocols documentation; Radiation Dose Unit; Radiation exposure; Reproducibility; Roentgen Rays; Selection Bias; Shapes; Source; Staging; System; Technology; Texture; Theoretical model; Tissues; Treatment Efficacy; Tweens; Vendor; Visualization; Work; X-Ray Computed Tomography; artificial intelligence algorithm; attenuation; base; clinical application; clinical decision-making; clinical translation; contrast enhanced; denoising; design; detector; fluorodeoxyglucose positron emission tomography; imaging approach; imaging biomarker; imaging modality; imaging system; improved; next generation; novel; novel therapeutics; perfusion imaging; predictive modeling; public health relevance; quantitative imaging; response; spectrograph; technology development; tumor; tumor microenvironment

Abstract. The proposed project concentrates on technology developments to enable high sensitivity, bias-tolerant spectral CT for accurate quantitation of iodine concentration. Spectral CT has the potential of providing true quantitative information of tissue composition and provides an avenue for combined functional and structural imaging. High- sensitivity spectral CT accommodates anatomical sites that are traditionally hard to image, and reliable meas- urements of iodine perfusion allow additional quantitative measures such as tissue texture to aid diagnosis and clinical decision making. In the case of pancreatic cancer, the complex tumor microenvironment and the conse- quential poor perfusion characteristics lead to difficulty in diagnosis, staging, and treatment assessment. The need for visualizing low-enhancing lesions and the benefit of extracting quantitative information directly from image data strongly motivate a high-sensitivity imaging modality for reproducible iodine measurements. The need for visualizing low-enhancing lesions and the benefit of extracting quantitative information directly from image data strongly motivate a high-sensitivity imaging modality for reproducible iodine measurements. How- ever, state-of-the-art spectral CT presents large quantitation bias, i.e., inaccuracies in measured iodine concen- tration compared to the truth. We identify three major sources that contribute to quantitation bias: imaging system (e.g., spectrum mismatch), post-processing (e.g., biased estimator), and patient (scatter, beam hardening). The bias effect in current spectral CT cannot be fully eliminated by increasing radiation exposure, and has complex dependencies on the imaging system, imaging techniques, patient habitus, and processing algorithms. This in- accuracy is a major impediment to pancreatic cancer management and quantitative applications in general. The overall goal of this proposal is to develop robust, high-sensitivity spectral CT solutions that will enhance sensi- tivity and reduce variability in iodine quantitation, which in turn enables accurate, high-performance spectral biomarkers for disease management. The following specific aims will be pursued: (1) to develop an end-to-end, modular theoretical model for robust spectral CT design and optimization, (2) to develop bias-tolerant processing pipeline, and (3) to implement and evaluate high performance, hybrid spectral CT solutions on an experimental CT bench. Completion of the proposed efforts enables robust, high sensitivity spectral CT for improved tumor detection and characterization through accurate, high performance spectral biomarkers. Vendor- and spectral technology-independent outcomes of the proposal include: optimized, patient-specific protocols; post-processing pipelines that are robust against quantitative bias and variability; and the next generation spectral CT system designs for enhanced iodine quantitation. Achievements from the proposed project will improve sensitivity and quantitation accuracy of iodinated contrast media in spectral CT which enables quantitative diagnostics and treatment assessment using robust iodine biomarkers across a broad range of clinical applications.

抽象的。 拟议的项目集中于技术开发，以实现高灵敏度、耐偏差光谱 CT 用于准确定量碘浓度。能谱 CT 有潜力提供真正的定量 组织成分的信息，并提供组合功能和结构成像的途径。高的- 灵敏度能谱 CT 可适应传统上难以成像的解剖部位，并提供可靠的测量 碘灌注的测定允许额外的定量测量，例如组织纹理，以帮助诊断和 临床决策。就胰腺癌而言，复杂的肿瘤微环境及其后果 潜在的不良灌注特征导致诊断、分期和治疗评估困难。这 需要可视化低增强病变以及直接从中提取定量信息的好处 图像数据强烈激发了高灵敏度成像模式的发展，以实现可重复的碘测量。这 需要可视化低增强病变以及直接从中提取定量信息的好处 图像数据强烈推动了高灵敏度成像模式的发展，以实现可重复的碘测量。如何- 迄今为止，最先进的光谱 CT 存在很大的定量偏差，即测量的碘浓度不准确。 与事实相比。我们确定了导致定量偏差的三个主要来源： 成像系统 （例如，频谱不匹配）、后处理（例如，有偏差的估计器）和患者（散射、光束硬化）。这 当前能谱CT中的偏差效应不能通过增加辐射暴露来完全消除，并且具有复杂的 依赖于成像系统、成像技术、患者习惯和处理算法。这在- 准确性是胰腺癌管理和定量应用的主要障碍。这 该提案的总体目标是开发强大的高灵敏度光谱 CT 解决方案，以增强灵敏度 活性并减少碘定量的变异性，从而实现准确、高性能的光谱 用于疾病管理的生物标志物。我们将追求以下具体目标：（1）开发端到端、 用于鲁棒能谱 CT 设计和优化的模块化理论模型，(2) 开发耐偏处理 管道，以及 (3) 在实验上实施和评估高性能混合能谱 CT 解决方案 CT 台。完成拟议的工作可以实现稳健、高灵敏度的能谱 CT，从而改善肿瘤 通过准确、高性能的光谱生物标志物进行检测和表征。供应商和频谱 该提案的独立于技术的成果包括：优化的、针对患者的方案；后处理 对定量偏差和变异性具有鲁棒性的管道；以及下一代能谱CT系统 增强碘定量的设计。拟议项目的成果将提高敏感性和 光谱 CT 中碘造影剂的定量准确性，可实现定量诊断和 在广泛的临床应用中使用强大的碘生物标志物进行治疗评估。