Natural Language Question Understanding for Electronic Health Records

电子健康记录的自然语言问题理解

• 批准号: 9479293
• 负责人: Kirk Edward Roberts
• 金额: $ 24.9万
• 依托单位: UNIVERSITY OF TEXAS HLTH SCI CTR HOUSTON
• 依托单位国家: 美国
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-05-01 至 2020-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9479293
• 关键词: Artificial Intelligence; Award; Blood Glucose; Caring; Clinical; Collection; Computer software; Data; Databases; Development Plans; Electronic Health Record; Ensure; Glucose; Goals; Health; Intuition; Knowledge; Literature; Manuals; Medical; Medical History; Medical Informatics; Mentors; Methods; Natural Language Processing; Pathway interactions; Patient Care; Patients; Pharmaceutical Preparations; Physicians; Process; Research; Research Personnel; Semantics; Specific qualifier value; Structure; System; Techniques; Testing; Text; TimeLine; Training; United States National Institutes of Health; Work; abstracting; base; biomedical informatics; career; career development; crowdsourcing; data access; design; electronic structure; health data; illiterate; improved; information organization; lexical; natural language; open source; operation; patient portal; phrases; prototype; syntax

 DESCRIPTION (provided by applicant): Patient information in the electronic health record (EHR) such as lab results, medications, and past medical history is the basis for physician decisions about patient care. It also helps patients better understand and manage their care. Efficient access to this patient information is thus essential. One of the most intuitive ways of accessing data is by asking natural language questions. A significant amount of work in medical question answering has been conducted, yet little work has been performed in question answering for EHRs. Natural language questions can be represented in logical forms, a standard structured knowledge representation technique. This project proposes to take natural language EHR questions, both for doctors and patients, and automatically convert them to a logical form. The logical forms can then be converted to a structured query such as those used by EHRs. A major obstacle to this approach is the lack of data containing questions annotated with logical forms. This project hypothesizes that a small set of questions can be manually annotated, and then paraphrases can be produced for each annotated question. Since paraphrasing is a simpler task than logical form annotation, crowd-sourcing techniques can be used to collect thousands of question paraphrases. This question paraphrase corpus will then be used to build a semantic grammar capable of recognizing the logical structure of EHR questions. To ensure a robust, generalizable grammar, existing NLP techniques will be used to pre-process questions, simplifying their syntactic structure and abstracting their medical concepts. In order to develop such a method, the candidate, Dr. Kirk Roberts, requires additional training and mentoring in natural language processing and biomedical informatics. This application for the NIH Pathway to Independence Award (K99/R00) describes a career development plan that will allow Dr. Roberts to achieve the goals of this project as well as transition to a career as an independent researcher. He will be mentored by Dr. Dina Demner-Fushman, a leading medical NLP researcher, and co-mentored by Dr. Clement McDonald, a leading EHR and medical informatics researcher. The specific aims of the project are: (1) Build a paraphrase collection of EHR questions, where each prototype question will have many unique paraphrases. The paraphrases encompass different lexical and syntactic means of conveying the same logical form. (2) Construct a semantic grammar for EHR questions. The grammar can then be used to convert a natural language question to a logical form. (3) Implement an end- to-end question analyzer that generalizes EHR questions for improved parsing, parses the question into a logical form using the grammar, and converts the logical form into a leading structured EHR query format.

 描述（由申请人提供）：电子健康记录（EHR）中的患者信息，如实验室结果、药物和既往病史，是医生做出患者护理决策的基础。它还有助于患者更好地了解和管理他们的护理。因此，有效地获取这些患者信息至关重要。访问数据的最直观的方式之一是询问自然语言问题。大量的工作在医疗问题回答已经进行，但很少的工作已经在电子病历的问题回答。自然语言问题可以用逻辑形式表示，这是一种标准的结构化知识表示技术。该项目建议采用自然语言EHR问题，无论是医生还是患者，并自动将其转换为逻辑形式。然后可以将逻辑形式转换为EHR所使用的结构化查询。这一方法的一个主要障碍是缺乏包含以逻辑形式注释的问题的数据。这个项目假设可以手动注释一小部分问题，然后可以为每个注释的问题生成释义。由于释义比逻辑形式注释更简单，因此可以使用众包技术来收集数千个问题释义。该问题释义语料库将被用来建立一个语义语法能够识别的逻辑结构的电子病历问题。为了确保一个强大的，可推广的语法，现有的NLP技术将用于预处理问题，简化其句法结构和抽象的医学概念。 为了开发这种方法，候选人Kirk Roberts博士需要在自然语言处理和生物医学信息学方面进行额外的培训和指导。NIH独立之路奖（K99/R 00）的申请描述了一个职业发展计划，该计划将使罗伯茨博士实现该项目的目标，并过渡到职业生涯， 独立研究员。他将由Dina Demner-Fushman博士指导，Dina Demner-Fushman博士是一位领先的医学NLP研究员，Clement McDonald博士是一位领先的EHR和医学信息学研究员。 该项目的具体目标是：（1）建立一个释义集的EHR问题，其中每个原型问题将有许多独特的释义。这些释义包含表达同一逻辑形式的不同词汇和句法手段。(2)为EHR问题构建语义语法。然后可以使用语法将自然语言问题转换为逻辑形式。(3)实现端到端问题分析器，其概括EHR问题以改进解析，使用语法将问题解析为逻辑形式，并将逻辑形式转换为领先的结构化EHR查询格式。

项目成果

Toward a Neural Semantic Parsing System for EHR Question Answering

• DOI: 10.48550/arxiv.2211.04569
• 发表时间: 2022-11
• 期刊: AMIA ... Annual Symposium proceedings. AMIA Symposium
• 作者: Sarvesh Soni;Kirk Roberts

A content-based dataset recommendation system for researchers-a case study on Gene Expression Omnibus (GEO) repository.

• DOI: 10.1093/database/baaa064
• 发表时间: 2020-01-01
• 期刊: Database : the journal of biological databases and curation
• 作者: Patra BG;Roberts K;Wu H
• 通讯作者: Wu H

Deep Patient Representation of Clinical Notes via Multi-Task Learning for Mortality Prediction.

• 发表时间: 2019
• 期刊: AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science
• 作者: Yuqi Si;Kirk Roberts

Understanding spatial language in radiology: Representation framework, annotation, and spatial relation extraction from chest X-ray reports using deep learning.

• DOI: 10.1016/j.jbi.2020.103473
• 发表时间: 2020-08
• 期刊: Journal of biomedical informatics
• 影响因子: 4.5
• 作者: Datta S;Si Y;Rodriguez L;Shooshan SE;Demner-Fushman D;Roberts K
• 通讯作者: Roberts K

A dataset of 200 structured product labels annotated for adverse drug reactions.

• DOI: 10.1038/sdata.2018.1
• 发表时间: 2018-01-30
• 期刊: Scientific data
• 影响因子: 9.8
• 作者: Demner-Fushman D;Shooshan SE;Rodriguez L;Aronson AR;Lang F;Rogers W;Roberts K;Tonning J
• 通讯作者: Tonning J