Quantitative, Image-Based Osteoarthritis Biomarkers Software Resubmission

项目总结 肌肉骨骼疾病在美国很常见，尤其是老年人和老年人 较低的社会经济地位，对国家的整体健康状况造成很大影响。骨病是 通过研究病人的病史和体格检查，以及实验室检查，骨骼 活组织检查和成像测试。骨成像测试提供了一种检查骨骼结构的非侵入性方法。然而， 成像数据通常是定性的或依赖于操作员的，而不是自动或定量的 测量。这些定量测量不够灵敏，无法检测到骨骼的细微变化。 与早期疾病进展相关的质量。我们建议发展高性能、多模式、 以及自动化的3D骨骼角色化工具，可通过Web浏览器访问。范围很广 的研究人员和临床医生可以利用这些工具来获得高通量、可重复性的生物标志物 统计敏感的研究研究。该系统将自动分割骨骼和软骨并量化 来自感兴趣区域的生物标志物。拟议的系统将具有卓越的高吞吐量能力 超过现有的骨骼图像分析套件，它将为研究人员提供访问最先进的算法 没有编程能力。除了为研究界提供强大的资源外，我们还将 将这一完整、简化的分析解决方案商业化，将其作为按图像收费的在线处理提供 服务。我们的系统将通过演示我们可以在临床前检测到骨骼退化来验证 研究，这可能导致新的临床试验的新的治疗和诊断方法 人类。我们将测试该系统可以自动识别膝关节图像中的骨关节炎的假设 并在微型计算机断层扫描图像中区分血友病。 拟议项目的最终目标是导致更好的预防战略和更好的进展。 监测骨关节炎及相关疾病。

PROJECT SUMMARY Musculoskeletal diseases are common in the United States, especially among the elderly and individuals of low socioeconomic status, and they take a large toll on the Nation's overall health status. Bone disorders are diagnosed by exploring a patient's medical history and by physical exam, alongside laboratory tests, bone biopsies, and imaging tests. Bone imaging tests provide a non-invasive way to examine at bone structure. However, imaging data is often evaluated qualitatively or with operator dependence as opposed to automated or quantitative measurements. These quantitative measurements are not sensitive enough to detect subtle variations in bone quality associated with early disease progression. We propose the development of high performance, multimodal, and automated 3D bone characterization tools, which are accessible through a web browser. A broad range of researchers and clinicians can leverage these tools to obtain high-throughput, reproducible biomarkers for statistically sensitive research studies. The system will automatically segment bone and cartilage and quantify biomarkers from the regions of interest. The proposed system will have superior high-throughput capabilities over existing bone image analysis suites, and it will provide access to state-of-the-art algorithms for researchers without programming abilities. In addition to providing a powerful resource to the research community, we will commercialize this complete, streamlined analytical solution by offering it as an online fee-per-image processing service. Our system will be validated by demonstrating that we can detect skeletal deterioration in preclinical studies, which can potentially lead to new clinical trials for novel therapeutic and diagnostic approaches in humans. We will test the hypothesis that the system can automatically identify osteoarthritis in knee images from the Osteoarthritis Initiative database and differentiate hemophilia in micro-computed tomography images. The ultimate goal of the proposed project is to lead to better preventive strategies and improved progression monitoring of osteoarthritis and related diseases.