har1/MTS_Dialogue-Clinical_Note
收藏资源简介:
--- license: mit task_categories: - text2text-generation - feature-extraction - summarization language: - en tags: - medical pretty_name: clinical-note-generator --- ## MTS Dialogue (Clinical Note Summarisation) Main Dataset The MTS-Dialog dataset is a new collection of 1.7k short doctor-patient conversations and corresponding summaries (section headers and contents). The training set consists of 1,201 pairs of conversations and associated summaries. The validation set consists of 100 pairs of conversations and their summaries. The "dialogue" column contain Doctor-Patient conversation. The "section_text" column contains the Clinical Note of the Doctor-Patient conversation. This clinical note is of the format : * Symptoms: * Diagnosis: * History of Patient: * Plan of Action: *** N/A is given if no information is found for each of the sections. This dataset is a modified version of the [MTS-Dialog dataset](https://github.com/abachaa/MTS-Dialog). The dataset was modified to satisfy the needs of the [har1/HealthScribe-Clinical_Note_Generator](https://huggingface.co/har1/HealthScribe-Clinical_Note_Generator) model. This is a fine-tuned version of [facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn). The main contributors of this dataset i.e. the modified version are : * [Aleena Patani](https://www.linkedin.com/in/aleena-patani-47aa69267/), * [Amigashabnam F](https://www.linkedin.com/in/amigashabnamf/), * [Harikrishnan K C](https://www.linkedin.com/in/harikrishnan-kc-a0a0441b9/), * [Sreeja S](https://www.linkedin.com/in/sreeja-s-9b3b43287/), * [Sujith Jayaprakash](https://www.linkedin.com/in/sujith-jayaprakash-291453288/).
license: MIT许可证 task_categories: - 文本到文本生成 - 特征提取 - 摘要生成 language: - 英语 tags: - 医疗 pretty_name: 临床笔记生成器 --- ## MTS对话(临床笔记摘要)主数据集 MTS-Dialog数据集是包含1700组医患对话及对应摘要(含章节标题与内容)的全新合集。 训练集包含1201组对话与关联摘要对。 验证集包含100组对话与摘要对。 其中"dialogue"列存储医患对话内容,"section_text"列存储对应医患对话的临床笔记,该临床笔记格式如下: * 症状: * 诊断: * 患者病史: * 行动方案: *** 若某一章节无相关信息,则标注为"N/A"。 本数据集是[MTS-Dialog数据集](https://github.com/abachaa/MTS-Dialog)的修改版本。 本次修改是为适配[har1/HealthScribe-Clinical_Note_Generator](https://huggingface.co/har1/HealthScribe-Clinical_Note_Generator)模型的需求,该模型是[facebook/bart-large-cnn](https://huggingface.co/facebook/bart-large-cnn)的微调版本。 本修改版数据集的主要贡献者为: * [Aleena Patani](https://www.linkedin.com/in/aleena-patani-47aa69267/) * [Amigashabnam F](https://www.linkedin.com/in/amigashabnamf/) * [Harikrishnan K C](https://www.linkedin.com/in/harikrishnan-kc-a0a0441b9/) * [Sreeja S](https://www.linkedin.com/in/sreeja-s-9b3b43287/) * [Sujith Jayaprakash](https://www.linkedin.com/in/sujith-jayaprakash-291453288/)
数据集概述
基本信息
- 许可证: MIT
- 任务类别:
- 文本生成
- 特征提取
- 摘要生成
- 语言: 英语
- 标签: 医疗
- 名称: clinical-note-generator
数据集详情
- 名称: MTS Dialogue (Clinical Note Summarisation)
- 描述: 包含1.7k对医生与患者对话及其对应摘要(章节标题和内容)的数据集。
- 训练集: 包含1,201对对话及其摘要。
- 验证集: 包含100对对话及其摘要。
- 数据结构:
- 对话列: 包含医生与患者的对话。
- 章节文本列: 包含医生与患者对话的临床笔记,格式为:
- 症状
- 诊断
- 患者病史
- 行动计划
- *** 若无信息,则标记为N/A。
数据集来源与修改
- 原始数据集: MTS-Dialog dataset
- 修改目的: 满足har1/HealthScribe-Clinical_Note_Generator模型的需求。
- 模型基础: facebook/bart-large-cnn的微调版本。
主要贡献者




