A Biomedical Natural Language Processing Resource

生物医学自然语言处理资源

• 批准号: 6899974
• 负责人: CAROL FRIEDMAN
• 金额: $ 51.54万
• 依托单位: COLUMBIA UNIVERSITY HEALTH SCIENCES
• 依托单位国家: 美国
• 财政年份: 2005
• 资助国家: 美国
• 起止时间: 2005-07-01 至 2009-06-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6899974
• 关键词: artificial intelligence; automated medical record system; bioinformatics; biomedical automation; biomedical resource; clinical research; computer assisted patient care; computer program /software; computer system design /evaluation; data management; health science research; high throughput technology; human data

DESCRIPTION: The long-term aim of this project is to advance clinical care and biomedical research by establishing a natural language processing (NLP) resource for the biomedical community. A major bottleneck for development of automated tools for clinical applications and biomedical research is that most of the data and knowledge occur in the form of text, resulting in a lack of coded data. This NLP resource will make possible a host of automated applications by enabling high throughput access to coded biomedical knowledge and data. The foundation of this resource will be the MedLEE NLP system, which has been used operationally for almost a decade in healthcare settings for a broad range of applications that have proven to be valuable for clinical care. The NLP resource will also include BioMedLEE (a derivative of MedLEE), which encodes genotypic-phenotypic (GP) relations in the scientific literature. It currently focuses on GP relations associated with cancer and infectious diseases, and is being used to organize the extracted information to facilitate research, curation, and ontological development within model organism databases. This proposal will enable us to 1) disseminate our NLP resource to the community, 2) conduct technological research and development (R&D) to facilitate expansion and adaptation of the resource to new applications and specialties, 3) conduct R&D of tools that facilitate use of the extracted data and knowledge after coding, and 4) promote the resource, and provide service to users in the form of technical support, documentation, and tutorials. MedLEE and BioMedLEE are extendable systems that encompass the clinical and scientific communities. The dissemination of a proven NLP system that is applicable to the entire biomedical community provides an exceptional opportunity for multiple developers and researchers to work to unleash the true potential of NLP technology, increasing development of applications that aim to enhance scientific research and improve all levels of health.

描述： 该项目的长期目标是通过为生物医学界建立自然语言处理（NLP）资源来推进临床护理和生物医学研究。临床应用和生物医学研究自动化工具开发的一个主要瓶颈是大多数数据和知识以文本形式出现，导致编码数据的缺乏。该 NLP 资源将通过实现对编码生物医学知识和数据的高吞吐量访问，使大量自动化应用程序成为可能。该资源的基础将是 MedLEE NLP 系统，该系统已在医疗保健环境中运行了近十年，其广泛的应用已被证明对临床护理有价值。 NLP 资源还将包括 BioMedLEE（MedLEE 的衍生物），它对科学文献中的基因型-表型 (GP) 关系进行编码。它目前专注于与癌症和传染病相关的全科医生关系，并用于组织提取的信息，以促进模型生物数据库内的研究、管理和本体论开发。该提案将使我们能够 1）向社区传播我们的 NLP 资源，2）进行技术研究和开发（R＆D），以促进资源的扩展和适应新的应用程序和专业，3）进行促进使用的工具的研发编码后提取的数据和知识，4）推广资源，以技术支持、文档和教程的形式为用户提供服务。 MedLEE 和 BioMedLEE 是涵盖临床和科学界的可扩展系统。适用于整个生物医学界的经过验证的 NLP 系统的传播为多个开发人员和研究人员提供了一个绝佳的机会，以释放 NLP 技术的真正潜力，增加旨在加强科学研究和提高各个水平的应用程序的开发。健康。