Open Health Natural Language Processing Collaboratory

项目摘要 利用电子健康记录（EHR）数据进行临床和翻译的主要障碍之一是 科学是普遍使用非结构化或半结构化的临床叙述， 信息.自然语言处理（NLP）从叙述中提取结构化信息， 受到了极大的关注，并发挥了关键作用，使二次使用的电子病历的临床和 翻译研究如ACT（临床应用的患者应计）等大规模工作所证明的那样， 试验），eMERGE和PCORnet，使用EHR数据进行研究依赖于强大的数据和 信息学基础设施，允许临床叙述的结构化，并支持临床 下游应用的信息。当前成功的NLP用例通常需要强大的信息学 团队（与NLP专家）与临床医生合作，提供他们的领域知识并构建定制的NLP 引擎迭代。这需要NLP专家和临床医生之间的密切合作，在 信息学支持有限的机构。此外， NLP系统仍然有限，部分原因是缺乏跨机构的EHR培训， 系统.电子健康记录数据的有限可用性限制了提高劳动力能力的培训 临床NLP我们的目标是通过扩大我们在以下方面的现有合作来应对上述挑战 开放健康自然语言处理（OHNLP）的多个CTSA中心，以共享 NLP工件（即，单词、n元语法、短语、句子、概念提及、概念和文本片段） 从跨多个机构的真实的EHR获得。我们将利用先进的隐私保护技术 iDASH的计算基础设施（集成数据分析，分析和共享），以保护隐私- 保留数据分析模型，并将与包括观察健康数据在内的各种社区合作 科学和信息学（OHDSI），精准医学计划（PMI），PCORnet和罕见疾病临床 研究网络（RDCRN）展示NLP在翻译研究中的实用性。CTSA的这一创新 RFA奖为我们提供了一个独特的机会，以应对临床NLP所面临的挑战， 通过与多个研究社区的强有力的伙伴关系和研究团队的领导作用， 临床NLP，我们设想，该项目的成功交付将扩大临床NLP的利用 在整个研究界。计划有四个目标：i）获得PHI抑制的NLP伪影， 保留多个机构的分销信息，并评估访问PHI的隐私风险- 抑制伪影，ii）生成用于临床叙述的探索性分析的合成文本语料库，以及 评估其在利用各种NLP挑战的NLP任务中的效用，iii）开发隐私保护计算 具有NLP的表型模型，以及iv）与不同的社区合作，以证明实用性 我们的转化研究项目。

Project Summary One of the major barriers in leveraging Electronic Health Record (EHR) data for clinical and translational science is the prevalent use of unstructured or semi-structured clinical narratives for documenting clinical information. Natural Language Processing (NLP), which extracts structured information from narratives, has received great attention and has played a critical role in enabling secondary use of EHRs for clinical and translational research. As demonstrated by large scale efforts such as ACT (Accrual of patients for Clinical Trials), eMERGE, and PCORnet, using EHR data for research rests on the capabilities of a robust data and informatics infrastructure that allows the structuring of clinical narratives and supports the extraction of clinical information for downstream applications. Current successful NLP use cases often require a strong informatics team (with NLP experts) to work with clinicians to supply their domain knowledge and build customized NLP engines iteratively. This requires close collaboration between NLP experts and clinicians, not feasible at institutions with limited informatics support. Additionally, the usability, portability, and generalizability of the NLP systems are still limited, partially due to the lack of access to EHRs across institutions to train the systems. The limited availability of EHR data limits the training available to improve the workforce competence in clinical NLP. We aim to address the above challenges by extending our existing collaboration among multiple CTSA hubs on open health natural language processing (OHNLP) to share distributional information of NLP artifacts (i.e., words, n-grams, phrases, sentences, concept mentions, concepts, and text segments) acquired from real EHRs across multiple institutions. We will leverage the advanced privacy-preserving computing infrastructure of iDASH (integrating Data for Analysis, Anonymization, and SHaring) for privacy- preserving data analysis models and will partner with diverse communities including Observational Health Data Sciences and Informatics (OHDSI), Precision Medicine Initiative (PMI), PCORnet, and Rare Diseases Clinical Research Network (RDCRN) to demonstrate the utility of NLP for translational research. This CTSA innovation award RFA provides us with a unique opportunity to address the challenges faced with clinical NLP and through strong partnership with multiple research communities and leadership roles of the research team in clinical NLP, we envision that the successful delivery of this project will broaden the utilization of clinical NLP across the research community. There are four aims planned: i) obtain PHI-suppressed NLP artifacts with retained distribution information across multiple institutions and assess the privacy risk of accessing PHI- suppressed artifacts, ii) generate a synthetic text corpus for exploratory analysis of clinical narratives and assess its utility in NLP tasks leveraging various NLP challenges, iii) develop privacy-preserving computational phenotyping models empowered with NLP, and iv) partner with diverse communities to demonstrate the utility of our project for translational research.