LongHealth

LongHealth数据集是由慕尼黑工业大学诊断与介入放射学系等机构合作创建的，包含20个详细的虚构患者案例，每个案例涉及5090至6754个单词。该数据集旨在评估大型语言模型处理长篇临床数据的能力，特别是在信息提取、否定和排序任务上的表现。数据集的创建过程由经验丰富的医生参与，确保案例的真实性。LongHealth数据集的应用领域主要集中在医疗健康领域，特别是用于提高医疗专业人员处理大量患者记录的效率和准确性。

The LongHealth dataset was collaboratively developed by institutions including the Department of Diagnostic and Interventional Radiology at the Technical University of Munich and other partner organizations. It consists of 20 detailed fictional patient cases, with each case containing between 5090 and 6754 words. This dataset is designed to evaluate the performance of large language models (LLMs) in processing long-form clinical data, particularly on tasks including information extraction, negation handling, and ranking. Experienced clinicians participated in the dataset creation process to ensure the authenticity of the patient cases. The primary application scenarios of the LongHealth dataset are concentrated in the healthcare field, specifically aimed at improving the efficiency and accuracy of medical professionals when dealing with large volumes of patient records.