CasiMedicos-Arg
收藏资源简介:
CasiMedicos-Arg是一个多语言的医学问答数据集,由法国蔚蓝海岸大学和西班牙巴斯克大学联合创建。该数据集包含558个临床案例,涵盖英语、西班牙语、法语和意大利语,每个案例都由医生提供正确和错误诊断的自然语言解释,并手动标注了论证结构(如前提、主张、支持关系和攻击关系)。数据集的创建过程包括从原始西班牙语数据自动翻译和手动校对为其他语言,并进行详细的论证结构标注。该数据集主要用于医学问答任务中的论证挖掘和生成技术研究,旨在帮助医生训练解释技能,特别是在需要解释诊断和治疗决策的场景中。
CasiMedicos-Arg is a multilingual medical question answering dataset jointly created by Université Côte d'Azur (France) and the University of the Basque Country (Spain). This dataset comprises 558 clinical cases spanning English, Spanish, French and Italian. For every case, physicians provide natural language explanations for both correct and incorrect diagnoses, and the argumentation structures—including premises, claims, support relations and attack relations—are manually annotated. The dataset development pipeline involves automatic translation from the original Spanish dataset and manual proofreading for the other three languages, followed by detailed annotation of argumentation structures. This dataset is primarily utilized for research on argument mining and generation techniques in medical question answering tasks, with the goal of assisting physicians in honing their explanation capabilities, particularly in scenarios requiring explanations for diagnostic and therapeutic decision-making.
Antidote CasiMedicos Datasets
数据集概述
该数据集基于CasiMedicos项目,该项目由医学专业人员组成的社区,自愿且免费地发布关于Resident Medical Intern考试(MIR)中可能答案的书面解释。这些考试通过多选题问卷评估医学生,有时通过简短的临床案例进行情境化。
数据集内容
- casimedicos-raw: 包含原始MIR考试和西班牙医学医生编写的解释的纯文本文档。
- casimedicos-exp: 对解释部分的标注,将解释中的部分与可能的答案相关联。
- casimedicos-arg: 对论证结构的标注,包括论证组件、关系、话语标记、医疗治疗和诊断。
casimedicos-raw
- 包含2011-2014年、2016年以及2018-2022年的1,561个带注释的问题。
- 选择包含临床案例的问题后,文档数量从1,561减少到622。
- 下载链接: casimedicos-raw
casimedicos-exp
- 基于casimedicos-raw中包含临床案例的文档。
- 手动标注了正确和错误答案解释的开始和结束位置。
- 提供了格式化的数据集: casimedicos-exp-squad-format
引用
-
casimedicos-raw: bibtex @inproceedings{Agerri2023HiTZAntidoteAE, title={HiTZ@Antidote: Argumentation-driven Explainable Artificial Intelligence for Digital Medicine}, author={Rodrigo Agerri and I{~n}igo Alonso and Aitziber Atutxa and Ander Berrondo and Ainara Estarrona and Iker Garc{i}a-Ferrero and Iakes Goenaga and Koldo Gojenola and Maite Oronoz and Igor Perez-Tejedor and German Rigau and Anar Yeginbergenova}, booktitle={SEPLN 2023: 39th International Conference of the Spanish Society for Natural Language Processing.}, year={2023} }
-
casimedicos-exp: bibtex @misc{goenaga2023explanatory, title={Explanatory Argument Extraction of Correct Answers in Resident Medical Exams}, author={Iakes Goenaga and Aitziber Atutxa and Koldo Gojenola and Maite Oronoz and Rodrigo Agerri}, year={2023}, eprint={2312.00567}, archivePrefix={arXiv} }
联系信息
- Rodrigo Agerri (rodrigo.agerri@ehu.eus)

- 1CasiMedicos-Arg: A Medical Question Answering Dataset Annotated with Explanatory Argumentative Structures法国蔚蓝海岸大学,CNRS,Inria,I3S · 2024年



