Quantitative high resolution cone beam CT for assessment of bone and joint health

项目总结/摘要 骨关节炎（OA）是最常见的退行性关节疾病，困扰着约3000万美国成年人， 目前缺乏有效治疗。放射学和MRI诊断OA主要依赖于 晚期疾病的生物标志物和早期检测OA的能力可以使治疗预防 不可逆转的软骨丧失。软骨下骨（SB）结构和骨密度（BMD）的变化 已被确定为早期疾病的可能生物标志物和治疗靶点。然而，这种潜力 OA诊断和治疗的进展受到缺乏结合高分辨率成像技术的阻碍。 表征小梁结构（<0.2 mm）所需的空间分辨率水平， 定量准确性，使可靠的BMD测量和成像的能力，在体内大关节。 专用四肢锥束CT（CBCT）是一项最新进展，可提供承重能力 与传统CT相比，成像、减少剂量和上级空间分辨率（~0.3 mm）， 使用我们机构开发的系统在临床研究中证明。我们假设一个先进的 用于肢体CBCT的系统，其利用新颖的CMOS检测器、基于模型的图像重建，以及 患者特异性散射校正将提供SB结构的同时体内评估， 骨密度和关节间隙形态用于OA的早期诊断和治疗反应评价。的 支持这一假设的系统的开发和临床转化涉及以下具体内容 目的：1）开发一种先进的肢体CBCT配置，用于采用 一个快速、低噪声、高分辨率的CMOS探测器，系统针对OA成像任务进行了优化， 与快速转换到肢体CBCT临床原型兼容; 2）增强空间分辨率， CBCT超越了传统的限制，基于模型的重建包含了系统模糊的模型， 无基准刚性运动校正，以及感兴趣体积、多分辨率对象表示， 约0.1 mm的最小可分辨细节尺寸，以便于准确的SB形态测定; 3）提高定量分析的准确性。 精确度和软组织对比度（例如，软骨和肌腱）使用患者特异性散射校正， 通过迭代去噪（预期校正时间约2 min/扫描）和以下方法加速蒙特卡罗计算： 开发一种新型的自动BMD校准系统，该系统与扫描仪外壳集成， 准确度优于3%;以及4）将目标1-3的系统集成到临床平台上，用于肢体 CBCT并在临床试点研究中评价系统，以评估潜在的诊断性能， SB结构和BMD测量的再现性。该项目将推进决议， CBCT的定量准确性转化为与OA早期检测和分期相关的试点研究， 在骨健康的其它领域例如类风湿性关节炎或骨质疏松症中具有潜在的应用。研究 是在约翰霍普金斯大学和Carestream Health之间的学术-工业联盟中进行的。

PROJECT SUMMARY / ABSTRACT Osteoarthritis (OA) is the most common degenerative joint disease, afflicting ~30 million US adults and currently lacking effective treatment. OA diagnosis with radiography and MRI relies predominantly on biomarkers of late disease, and capability for early detection of OA could enable therapies preventing irreversible cartilage loss. Changes in subchondral bone (SB) architecture and bone mineral density (BMD) have been identified as possible biomarkers of early disease and a target for therapy. However, this potential advance in OA diagnosis and treatment is hampered by the lack of imaging technology that combines the high level of spatial resolution required to characterize trabecular architecture (<0.2 mm) with a high degree of quantitative accuracy enabling reliable BMD measurement and the ability to image large joints in vivo. Dedicated extremity cone-beam CT (CBCT) is a recent advance providing the capability for weight-bearing imaging, reduced dose, and superior spatial resolution (~0.3 mm) compared to conventional CT, as demonstrated in clinical studies using a system developed at our institution. We hypothesize that an advanced system for extremity CBCT utilizing a novel CMOS detector, model-based image reconstruction, and patient-specific scatter correction will provide simultaneous in-vivo assessment of SB architecture, BMD, and joint space morphology for early diagnosis and evaluation of treatment response in OA. The development and clinical translation of the system supporting this hypothesis involve the following Specific Aims: 1) Develop an advanced extremity CBCT configuration for in-vivo assessment of bone health employing a fast, low-noise, high-resolution CMOS detector within a system optimized for OA imaging tasks and compatible with rapid translation to the extremity CBCT clinical prototype; 2) Enhance spatial resolution in CBCT beyond conventional limits with model-based reconstruction incorporating a model of system blur, fiducial-free rigid motion correction, and a volume-of-interest, multi-resolution object representation to yield minimum resolvable detail size of ~0.1 mm to facilitate accurate SB morphometry; 3) Improve quantitative accuracy and soft tissue contrast (e.g., cartilage and tendons) using a patient-specific scatter correction with Monte Carlo calculation accelerated with iterative de-noising (anticipated correction time ~2 min/scan) and by developing a novel, automatic BMD calibration system integrated with the scanner enclosure to yield BMD accuracy better than 3%; and 4) Integrate the systems from Aims 1-3 onto a clinical platform for extremity CBCT and evaluate the system in a clinical pilot study to assess potential diagnostic performance and reproducibility of the SB architecture and BMD measurement. The project will advance the resolution and quantitative accuracy of CBCT translated to pilot studies pertinent to the early detection and staging of OA, with potential application in other areas of bone health e.g. rheumatoid arthritis or osteoporosis. The research is conducted in an academic-industry consortium between Johns Hopkins University and Carestream Health.