MIMIC-IV-Ext-SEQ

MIMIC-IV-Ext-SEQ是一个专为医疗健康领域设计的序列建模数据集，由荷兰埃因霍温理工大学的研究团队开发。该数据集通过将MIMIC-IV临床数据库中的异构临床记录转换为统一的交互事件流格式，以便于机器学习模型理解和预测未来的医疗事件。数据集包含87899个词汇的交互词汇表，这些词汇代表了医生和患者之间的交互，适用于强化学习等机器学习任务，旨在提高医疗决策的自动化和智能化水平。

MIMIC-IV-Ext-SEQ is a sequence modeling dataset tailored for the healthcare domain, developed by a research team at Eindhoven University of Technology in the Netherlands. This dataset converts heterogeneous clinical records from the MIMIC-IV clinical database into a unified interactive event stream format, facilitating machine learning models to comprehend and predict future medical events. It encompasses an interactive vocabulary of 87,899 terms, which represent interactions between healthcare providers and patients, and is applicable to machine learning tasks such as reinforcement learning, aiming to enhance the automation and intelligence of medical decision-making.