ICD编码的德国肿瘤诊断数据集
收藏资源简介:
本研究的数据集包含2024个独特的德国肿瘤诊断文本描述,每个描述都对应一个ICD-10代码和ICD-O拓扑代码。数据集是从当地肿瘤文档系统中提取的,用于评估经过指令微调的LLMs在医疗文档任务中的编码能力。数据集的创建使用了德国版本的ICD-10-GM、ICD-O-3和OPS目录,并包含了超过500,000个基于这些目录的问题-答案对。数据集的应用领域是提高LLMs在德国语言肿瘤诊断文本编码中的准确性,旨在解决德国医疗文档中肿瘤诊断编码自动化的问题。
The dataset for this study comprises 2,024 unique German textual descriptions of tumor diagnoses, with each description paired with one ICD-10 code and one ICD-O topographic code. This dataset was extracted from a local oncology documentation system, and is utilized to evaluate the coding performance of instruction-tuned Large Language Models (LLMs) on medical documentation tasks. The dataset was developed using the German editions of the ICD-10-GM, ICD-O-3, and OPS coding catalogs, and contains more than 500,000 question-answer pairs grounded in these coding standards. The intended use of this dataset is to enhance the coding accuracy of LLMs when processing German-language tumor diagnosis texts, with the goal of resolving the challenges associated with automating tumor diagnosis coding in German medical documentation.




