门诊结构化病历高质量数据集

该数据为医院在门诊医生使用语音录入系统生成病历时产生，包含了语音识别后的文本以及经过后结构化处理形成的标准化病历数据（如主诉、现病史、诊断的结构化字段）。数据形态为文本和结构化数据，用于优化语音识别模型和自然语言处理（NLP）结构化引擎。业务场景应用于全院各门诊科室。该数据集可支撑持续提升语音识别的准确率和医学专业术语的识别能力，并实现病历内容的深度结构化，助力医生解放双手、提高效率，同时为临床科研提供高质量数据基础。

This dataset is generated when outpatient physicians create medical records using speech dictation systems in hospitals. It contains both the text output after speech recognition and standardized medical record data processed through post-structuring, including structured fields for chief complaints, present medical histories, and diagnoses. The data is available in text and structured formats, and is designed to optimize speech recognition models and natural language processing (NLP) structured engines. Its application scenarios cover all outpatient departments across the hospital. This dataset can support the continuous improvement of speech recognition accuracy and medical professional term recognition capabilities, realize deep structuring of medical record content, help doctors reduce their workload and improve work efficiency, and provide a high-quality data foundation for clinical research.

搜集汇总

背景与挑战

背景概述

该数据集源自医院门诊语音录入系统，包含语音识别文本及结构化病历数据，用于优化语音识别和自然语言处理模型。其可提升医学术语识别准确率，支持临床诊疗效率提升和科研数据应用。