Technology Research Site for Advanced, Faster Quantitative Imaging for BACPAC

项目摘要/摘要 脊柱疾患对社会有巨大的影响；既有身体上的疾病，也有 由于生产力下降和医疗费用增加，对个人和经济都造成了影响。尽管意义重大 对于这个问题，在许多患者中，症状的病因是多样和不清楚的，几乎没有可靠的证据。 前瞻性地确定适当的病人护理过程和客观评估的方法 各种干预措施的有效性。造成这一重大医疗困境的挑战包括 背部疼痛的来源众多，难以使用任何单一成像技术显示相关组织 疼痛定位的技术和困难及其促成的分子过程。磁共振 成像(MR)已被用来表征椎间盘、肌肉、神经和正电子发射断层扫描(PET)。 已被用于研究下腰痛受试者的骨转换和小关节疾病。 UH2/UH3中提出的研究和工具开发在临床上迈出了关键的下一步 翻译下腰痛患者更快的定量磁共振成像(MR)。新的 需要优化的技术和患者研究来研究其量化的临床潜力 描述与下腰痛有关的组织特征，并客观评估疼痛。我们的建议 美国国立卫生研究院背痛联盟的多学科技术研究网站(Tech Site)将 开发第四阶段TTM(技术优化研究和开发)，以利用两个关键 技术进步-开发基于机器学习的更快的MR采集方法，以及 机器学习用于图像分割和从图像中提取与目标疾病相关的特征。我们 将开发、验证和部署基于端到端深度学习的技术(TTM)以实现加速映像 重建、组织分割、脊柱退变检测，便于自动化、健壮 评估脊柱特征、神经认知疼痛反应、 以及患者报告的结果。为了完成这一重要项目，我们汇集了一支经验丰富的 多学科研究团队结合了广泛的MR生物工程专业知识、先进的MRI数据 分析、放射学、神经科学、神经外科、整形外科、多维分析和 现有的与产业界的研究协议。研究设施和环境包括临床和 成功完成拟议翻译项目所需的研究基础设施。这支队伍有 以前向学术界传播工具，与工业界密切合作，并积极与 随着该财团计划的发展，BACPAC。

PROJECT SUMMARY/ABSTRACT Disorders of the spine have a tremendous impact on society; both physically through the morbidity of afflicted individuals, and financially, through lost productivity and increased health care costs. Despite the significance of this problem, the etiology of symptoms is diverse and unclear in many patients, and there are few reliable methods by which to prospectively determine the appropriate course of patient care and to objectively evaluate the effectiveness of various interventions. Challenges contributing to this major healthcare dilemma include numerous sources of back pain, difficulty in visualization of responsible tissues using any single imaging technique and difficulty in the localization of pain and contributing molecular processes. Magnetic Resonance imaging (MR) has been used to characterize disc, muscle, nerves and Positron Emission Tomography (PET) has been used to study bone turnover, and facet disease in subjects with lower back pain. The research and tool development proposed in this UH2/UH3 takes the critical next step in the clinical translation of faster, quantitative magnetic resonance imaging (MR) of patients with lower back pain. New optimized techniques and patient studies are required to investigate its clinical potential for quantitatively characterizing the tissues implicated in lower back pain, and objective evaluation of pain. Our proposed multidisciplinary Technology Research Site (Tech Site) of the NIH Back Pain Consortium (BACPAC) will develop Phase IV TTMs (Research and Development for Technology Optimization) to leverage two key technical advancements – development of machine learning based faster MR acquisition methods, and machine learning for image segmentation and extraction of objective disease related features from images. We will develop, validate, and deploy end-to-end deep learning-based technologies (TTMs) for accelerated image reconstruction, tissue segmentation, detection of spinal degeneration, to facilitate automated, robust assessment of structure-function relationships between spine characteristics, neurocognitive pain response, and patient reported outcomes. To accomplish this important project, we have assembled a highly-experienced multidisciplinary research team combining extensive expertise MR bioengineering, advanced MRI data analysis, radiology, neuroscience, neurosurgery, orthopedic surgery, multi-dimensional analytics and have existing research agreements with industry. The research facilities and environment include the clinical and research infrastructure required for successful completion of the proposed translational project. The team has disseminated tools before to academia, worked closely with industry and are motivated to totally work with BACPAC as the plans of the consortium evolve.