MORPHOMETRY BIOMEDICAL INFORMATICS RESEARCH NETWORK

形态学生物医学信息学研究网络

• 批准号: 7724585
• 负责人: BRUCE R ROSEN
• 金额: $ 257.04万
• 依托单位: MASSACHUSETTS GENERAL HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-06-01 至 2009-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7724585
• 关键词: Anisotropy; Area; Arts; Atlases; Behavioral; Biomedical Informatics Research Network; Calibration; Clinical; Clinical Data; Collaborations; Collection; Computational Science; Computer Retrieval of Information on Scientific Projects Database; Computer software; Data; Databases; Development; Diagnostic; Diffusion; Diffusion Magnetic Resonance Imaging; Disease; Ensure; Environment; Evaluation; Fiber; Functional disorder; Funding; Genetic; Grant; Health; Human Genetics; Image; Image Analysis; Imagery; Incentives; Institution; Investigation; Knowledge; Machine Learning; Magnetic Resonance Imaging; Measurement; Methodology; Methods; Modality; Network-based; Noise; Numbers; Population; Population Control; Process; Protocols documentation; Range; Recruitment Activity; Research; Research Infrastructure; Research Personnel; Resolution; Resources; Services; Shapes; Site; Source; Structure; Study Subject; Techniques; Technology; Testing; Tissues; Translations; United States National Institutes of Health; Universities; Validation; Washington; Weight; Work; base; bioimaging; cluster computing; data acquisition; data integration; data management; data structure; design; improved; instrument; interoperability; morphometry; platform-independent; prognostic; shape analysis; tool; transmission process; white matter

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. DESCRIPTION (provided by applicant): Technological advances in imaging have revolutionized the biomedical investigation of illness. The tremendous potential that this methodology brings to advancing diagnostic and prognostic capabilities and in treatment of illnesses has as yet remained largely an unfulfilled promise. This potential has been limited by a number of technological impediments that could be in large part overcome by the availability of a federated imaging database and the attendant infrastructure. Specifically, the ability to conduct clinical imaging studies across multiple sites, to analyze imaging data with the most powerful software regardless of development site, and to test new hypotheses on large collections of subjects with well characterized image and clinical data would have a demonstrable and positive impact on progress in this field. The Morphometry BIRN (mBIRN), established in October 2001, has made substantial progress in the development of this national infrastructure to develop a data and computational network based on a federated data acquisition and database across seven sites in the service of facilitating multi-site neuroanatomic analysis. Standardized structural MRI image acquisition protocols have been developed and implemented that demonstrably reduce initial sources of inter-site variance. Data structure, transmission, storage and querying aspects of the federated database have been implemented. In this continuation of the mBIRN efforts, we propose three broad areas of work: 1) continuing structural MRI acquisition optimization, calibration and validation to include T2 and DTI; 2) translation of site specific state-of-the-art image analysis, visualization and machine learning technologies to work in the federated, multi-site BIRN environment; and 3) extension of data management and database query capabilities to include additional imaging modalities, clinical disorders and individualized human genetic covariates. These broad areas of work will come together in through key collaborations that will ensure utilization promotion by facilitating data entry into the federated database and creation of database incentive functionality. Our participating sites include MGH (PI), BWH, UCI, Duke, UCLA, UCSD, John Hopkins, and newly added Washington University and MIT. We have made a concerted effort to bridge the gap that can exist between biomedical and computational sciences by recruiting to our group leaders in both of these domains. Our efforts will be coordinated with those of the entire BIRN consortium in order to insure that acquisition and database functionality, and application-based disorder queries are interoperable across sites and designed to advance the capabilities to further knowledge and understanding of health and disease. Specifically, we propose the following Projects, with their associated specific aims: Project 1 Standardize and calibrate the acquisition of high-resolution structural MRI data to facilitate precise, quantitative, platform independent, multi-site evaluation of normal and pathological structural imaging data at multiple field strengths. 1.1 Develop methods to improve structural Tr and FSE-based PD, T2-weighted, and FLAIR weighted MRI acquisition protocols that maximize image quality, improve sensitivity, reduce noise and enable quantitative analysis of healthy and diseased tissue across sites and instruments. 1.2 Extend these methods to the acquisition of diffusion sensitive imaging, to improve diffusion MRI protocols and correction methods that minimize variability across sites while optimizing image quality and sensitivity for reconstructing fiber tracts and detecting abnormalities. Project 2 Continue to develop, integrate and deploy a suite of freely available software to enable scientific investigation of the morphological bases of function and dysfunction through increasingly sophisticated image analysis on increasingly large subject populations acquired at multiple research sites. 2.1 Adapt and apply automated and semi-automated tools to segment subcortical structures, delineate the cortex, and parcellate cortical functional and anatomical regions from a range of input image protocols by drawing on expertise and existing software of the participating institutions. 2.2 Adapt and apply shape-based morphometric tools to investigate clinical and control populations: continue to develop interoperability between segmentation and shape analysis tools through standardized data representation. 2.3 Integrate Diffusion Tensor Imaging (DTI), anisotropy measurements, white matter (WM) atlases, and automated tractography techniques into the BIRN morphometry analysis infrastructure. 2.4 Provide an integrated visualization tool to support detailed investigation of morphometry and other data types. 2.5 Develop a visualization-based query tool to facilitate knowledge discovery and development of scientific explanations. 2.6 Adapt and apply machine-learning techniques to identify statistically related subpopulations of study subjects based on biomedical images. Project 3 Create an infrastructure that will ensure efficient data management, reliable processing and dynamic access to imaging, behavioral, clinical and genetic data. 3.1 Continue to develop and deploy an extensible database, that can be adapted to fulfill a local site''s needs and that interoperates with the federated BIRN database infrastructure 3.2 Extend the BIRN infrastructure based on the capabilities and needs of the mBIRN collaborators to incorporate T2, FLAIR, and diffusion image data and genetic information. 3.3 Develop, test, and validate automated graphical protocols for data integration, pre- and postprocessing and display.

该副本是利用众多研究子项目之一 由NIH/NCRR资助的中心赠款提供的资源。子弹和 调查员（PI）可能已经从其他NIH来源获得了主要资金， 因此可以在其他清晰的条目中代表。列出的机构是 对于中心，这不一定是调查员的机构。 描述（由申请人提供）： 成像的技术进步彻底改变了对疾病的生物医学研究。这种方法带来促进诊断和预后能力的巨大潜力以及对疾病的治疗，但仍在很大程度上尚未实现。这种潜力受到许多技术障碍的限制，这些障碍可能会因联合成像数据库和随之而来的基础架构的可用性而很大程度上克服。具体而言，无论开发地点如何，都可以在多个站点进行临床成像研究，分析具有最强大软件的成像数据的能力，并在具有良好表征的图像和临床数据的大量受试者中测试新的假设，对该领域的进展具有明显的积极影响。成立于2001年10月的形态计量学BIRN（MBIRN）在该国家基础设施的开发方面取得了重大进展，以基于跨七个站点的联合数据采集和数据库来开发数据和计算网络，以促进多壁画神经瘤分析。已经开发和实施了标准化的结构MRI图像采集方案，这些方案明显地降低了地点间差异的初始来源。已经实施了联合数据库的数据结构，传输，存储和查询方面。在MBIRN努力的延续中，我们提出了三个广泛的工作领域： 1）持续结构性MRI获取优化，校准和验证，包括T2和DTI； 2）在联邦多站点的BIRN环境中使用特定网站的特定最先进的图像分析，可视化和机器学习技术； 3）扩展数据管理和数据库查询功能，包括其他成像方式，临床疾病和个性化的人类遗传协变量。这些广泛的工作领域将通过关键的协作聚集在一起，通过促进数据将数据输入到联合数据库中并创建数据库激励功能，从而确保利用促销。我们的参与网站包括MGH（PI），BWH，UCI，DUKE，UCLA，UCSD，JOHN HOPKINS和新添加的华盛顿大学和MIT。我们通过在这两个领域中招募我们的小组领导者来招募生物医学和计算科学之间可能存在的差距。我们的努力将与整个Birn联盟的努力进行协调，以确保获得和数据库功能，基于应用程序的疾病查询在跨站点可以互操作，并旨在提高能力，以进一步了解健康和疾病。 具体来说，我们提出以下项目，及其相关的特定目的： 项目1标准化并校准了高分辨率结构MRI数据的获取，以促进对多个田间强度下正常和病理结构成像数据的精确，定量，独立的，多站点评估。 1.1开发方法来改善结构性TR和FSE基于FSE的PD，T2加权和FLAIR加权MRI获取方案，以最大程度地提高图像质量，提高灵敏度，降低噪声并启用对跨站点和仪器的健康和患病组织的定量分析。 1.2将这些方法扩展到扩散敏感成像的获取，以改善扩散MRI方案和校正方法，以最大程度地减少位点的可变性，同时优化图像质量和敏感性以重建纤维区域并检测异常。 项目2继续开发，整合和部署一套免费的软件，以通过对在多个研究地点获得的越来越大的主题种群进行越来越复杂的图像分析来对功能和功能障碍的形态学基础进行科学研究。 2.1适应并应用自动化和半自动化的工具来分割皮层结构，描述皮层以及通过利用参与机构的专业知识和现有软件，从一系列输入图像协议中划分皮层功能和解剖区域。 2.2适应并应用基于形状的形态计量工具来研究临床和控制种群：通过标准化数据表示，继续开发分割和形状分析工具之间的互操作性。 2.3整合扩散量张量成像（DTI），各向异性测量，白质（WM）地图和自动拖拉技术技术中的BIRN形态分析基础架构。 2.4提供一个集成的可视化工具，以支持形态计量学和其他数据类型的详细研究。 2.5开发一种基于可视化的查询工具，以促进知识发现和开发科学解释。 2.6根据生物医学图像，适应并应用机器学习技术来识别研究对象的统计相关亚群。 项目3创建一个基础架构，以确保有效的数据管理，可靠的处理以及对成像，行为，临床和遗传数据的动态访问。 3.1继续开发和部署可扩展的数据库，可以适应以满足本地站点的需求，并与联合的BIRN数据库基础架构相互互操作3.2扩展了基于MBIRN合作者的功能和需求，扩展了BIRN基础架构，以合并T2，Flair，Flair，Flair和Genetic Image和GENETIC图像数据。 3.3开发，测试和验证自动图形协议，以进行数据集成，预处理和后处理和显示。