MULTIMODALITY NEUROIMAGING DATABASE SYSTEM

多模态神经影像数据库系统

• 批准号: 2872991
• 负责人: STEPHEN TC WONG
• 金额: $ 9.21万
• 依托单位: UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
• 依托单位国家: 美国
• 财政年份: 1998
• 资助国家: 美国
• 起止时间: 1998-02-01 至 2003-01-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/2872991
• 关键词: Internet; abstracting; bioimaging /biomedical imaging; brain imaging /visualization /scanning; computer graphics /printing; computer program /software; computer system design /evaluation; data management; epilepsy; health care professional practice; health science research; human data; image processing; information retrieval; information systems; interactive multimedia; magnetic resonance imaging; magnetoencephalography; medical education; multiple sclerosis; nuclear magnetic resonance spectroscopy; positron emission tomography

DESCRIPTION (Taken from application abstract): The past three decades have witnessed a remarkable growth of medical imaging modalities in patient care. Ironically, however, the field of medical imaging is threatened by its own success in that the sheer mass of information available is becoming unmanageable. The long-term objective of this application is to develop the next generation of medical image information systems. This five-year FIRST award project will focus on the design and implementation of a prototype neuroimaging database management system (NI-DBMS) for managing and indexing multimodality brain images, including MRI, PET, MR spectroscopy, and MEG. This new image DBMS intends to support flexible search and browsing of picture archiving and communication systems (PACS) image library, as well as clinical applications in epilepsy and multiple sclerosis. The hypothesis of this research is that different kinds of three-dimensional (3D) brain images and associated clinical data can be modeled, correlated, queried, and analyzed by computers on the World Wide Web (WWW) to aid neurologic education, practice, and research. The specific aims to investigate this hypothesis are: (I) design a new framework for multimodality neuroimage management; (ii) develop tools for extracting structural, functional, and temporal features of different brain images; (iii) investigate new content based search algorithms for epilepsy and multiple sclerosis imaging; (iv) develop a neuroimaging data model to allow access from the Web; and (v) evaluate system performance with well-defined experiments in epilepsy and multiple sclerosis imaging. The neuroimaging DBMS will be built on top of the large image archives in hospitals and will add values to the existing PACS with content-based information management and Web accessibility. The software product developed in this research will be generalizable and extensible so that it can be put into wide use by other medical centers and hospitals. Its open systems architecture will ensure image databases developed by the medical community will be sharable through the ubiquitous Web.

描述（摘自申请摘要）：过去三十年 见证了医学成像方式在患者护理中的显着增长。 然而，具有讽刺意味的是，医学成像领域正受到其自身的威胁。 成功之处在于，大量的可用信息正在变得越来越重要 难以管理。 该应用程序的长期目标是开发 下一代医学图像信息系统。 这个五年第一 奖项项目将重点关注原型的设计和实现 用于管理和索引的神经影像数据库管理系统 (NI-DBMS) 多模态脑部图像，包括 MRI、PET、MR 光谱和 MEG。 这个新的图像 DBMS 旨在支持灵活的搜索和浏览 图片存档和通信系统 (PACS) 图像库，以及 癫痫和多发性硬化症的临床应用。 本研究的假设是不同种类的三维空间 (3D) 大脑图像和相关临床数据可以建模、关联、 由计算机在万维网 (WWW) 上查询和分析以提供帮助 神经病学教育、实践和研究。 具体目标是 研究这个假设的方法是：（I）设计一个新的框架 多模态神经影像管理； (ii) 开发提取工具 不同大脑图像的结构、功能和时间特征； (iii) 研究基于新内容的癫痫搜索算法 多发性硬化症成像； (iv) 开发神经影像数据模型 从网络访问； (v) 通过明确的定义来评估系统性能 癫痫和多发性硬化症成像实验。 神经影像 DBMS 将建立在大型图像档案之上 医院并将通过基于内容的方式为现有 PACS 增加价值 信息管理和网络可访问性。 软件产品 这项研究中开发的内容将是可推广和可扩展的，以便 可以被其他医疗中心和医院广泛使用。 它的开放 系统架构将确保医疗机构开发的图像数据库 社区将通过无处不在的网络进行共享。