ArchEHR-QA

ArchEHR-QA数据集是一个新颖的专家注释数据集，基于重症监护室和急诊室环境中的真实世界患者案例。数据集包含患者向公共卫生论坛提出的问题、临床医生解释的对应问题、相关临床笔记摘录以及临床医生撰写的答案。该数据集旨在支持以患者为中心的电子健康记录（EHR）问答系统的开发和评估。数据集共包含134个患者案例，每个案例都包含患者提出的问题、相应的临床医生解释、临床笔记摘录以及临床医生撰写的答案。数据集的创建过程包括从公共卫生论坛和电子健康记录数据库中收集数据，并将患者提出的问题与相应的临床笔记进行对齐。数据集的应用领域是电子健康记录问答系统，旨在解决患者在住院期间对临床证据的信息需求。

The ArchEHR-QA dataset is a novel expert-annotated dataset based on real-world patient cases in intensive care unit (ICU) and emergency department (ED) settings. It contains questions raised by patients on public health forums, corresponding explanations from clinicians, excerpts of relevant clinical notes, and answers written by clinicians. This dataset aims to support the development and evaluation of patient-centric electronic health record (EHR) question answering (QA) systems. In total, the dataset includes 134 patient cases, each containing the patient’s question, corresponding clinician explanation, clinical note excerpts, and clinician-authored answers. The dataset was constructed by collecting data from public health forums and EHR databases, and aligning the patients’ questions with their corresponding clinical notes. Its application focuses on EHR QA systems, targeting the resolution of patients’ information needs regarding clinical evidence during their hospitalization.