Encoding and Processing Patient Allergy Information in EHRs

摘要 过敏症影响五分之一的美国人，是美国第五大慢性病。 每年，过敏症占超过1700万门诊就诊。虽然 在电子健康记录（EHR）中记录和交换过敏信息， 我们的地位日益重要，但我们仍然面临多重挑战。其中包括：缺乏 广泛采用的代表过敏的标准术语，频繁输入过敏 信息作为自由文本，并且没有用于协调过敏信息的现有过程。在这 在这项研究中，我们将通过实现以下具体目标，为这些挑战提供解决方案： 1)对标准术语和大型过敏症资料库进行分析， 用于表示过敏信息的综合知识库; 2）设计、开发和 评估用于提取和编码自由文本的自然语言处理（NLP）模块 过敏信息，并将其与现有的NLP系统集成; 3）衡量可行性 以及拟采用的自然语言处理系统在过敏症调解新程序中的效率;以及 4)分发我们的方法和工具，以便其他研究人员广泛使用， 医疗机构用于非商业用途。

Abstract Allergies affect one in five Americans and are the 5th leading chronic disease in the U.S. Each year, allergies account for more than 17 million outpatient office visits. Although documenting and exchanging allergy information in electronic health records (EHRs) is becoming increasingly important, we still face multiple challenges. These include: lack of well-adopted standard terminologies for representing allergies, frequent entry of allergy information as free-text, and no existing process for reconciling allergy information. In this study, we will provide solutions to these challenges by addressing the following specific aims: 1) conduct analyses on standard terminologies and a large allergy repository to build a comprehensive knowledge base for representing allergy information; 2) design, develop and evaluate a natural language processing (NLP) module for extracting and encoding free-text allergy information and integrate it with an existing NLP system; 3) measure the feasibility and efficiency of the proposed NLP system for the new process of allergy reconciliation; and 4) distribute our methods and tool, so they are widely available to other researchers and healthcare institutions for non-commercial use.