MedDec
收藏资源简介:
MedDec数据集由马萨诸塞大学洛厄尔分校和麻省理工学院的专家团队创建,包含451份出院总结,涵盖11种疾病类型和10种医疗决策类别。数据集通过专家标注,旨在提取和分类临床笔记中的医疗决策,以改善临床实践和生物NLP技术。该数据集的创建过程严格遵循DICTUM分类法,通过三位专家的独立标注和仲裁确保数据质量。MedDec数据集的应用领域广泛,包括但不限于医疗决策支持系统、健康政策制定和临床决策分析。
The MedDec dataset was developed by a team of experts from the University of Massachusetts Lowell and the Massachusetts Institute of Technology. It contains 451 discharge summaries covering 11 disease types and 10 medical decision categories. Annotated by domain experts, this dataset aims to extract and classify medical decisions from clinical notes, so as to advance clinical practice and biomedical natural language processing (BioNLP) technologies. Its construction strictly follows the DICTUM taxonomy, and data quality is ensured via independent annotation and arbitration conducted by three experts. The MedDec dataset has a wide range of application scenarios, including but not limited to medical decision support systems, health policy formulation and clinical decision analysis.




