国际疾病分类(肿瘤)智能编码体系及质量评价系统设计

Design of Automatic Coding and Quality Assessment System of International Classification of Diseases for Oncology

目的:针对肿瘤类疾病构建一套包含国际疾病分类(the 10th revision of the International Statisfical classification of Diseases and Related Health Problems,ICD-10)智能编码、编码质控及效果评价的编码体系,并将体系推广至其他类别疾病。方法:通过采集病案首页相关信息,根据肿瘤主要诊断选择原则、疾病编码原则,将相关指标建立内在逻辑关联,辅以语意识别,建立一套"嵌入精细判定诊断顺序逻辑规则+语意识别+机器学习+既往病史辅助判定+基于权重分析的人工校对"结合模式的肿瘤专科智能编码体系。结果:通过反复校对,获得一条预测人工校正分值的分割线,分割线以下个案(占比低于20%)进行人工编码,其余应用智能编码系统结果,预期编码准确性高于95%。结论:获得标准统一、高效、准确的肿瘤编码体系,并外推至全病种。

Objective: This study aims to set up a system including The 10 th Revision of the International Classification of Diseases(ICD-10) automatic coding, and assessment of coding quality basing on the oncological cases. For further development, this automatic coding and quality assessment system can be used in other sorts of diseases. Methods: The oncology specialist automatic coding system with multiple technics such as the logical fine decision rule of diagnostic orders, semantic recognition, machine learning, assisted judgment system with past medical record and manual check basing on results of weight analysis will be set up basing on the records from case pages, the main diagnostic selection principle and the disease coding principle. Results: The manual invention cut-off line of score from assessment system will be found by repeated checking. The cases below the cut-off line(<20%) will be coded manually, while the rest cases will accept the results from automatic coding system. The expected accuracy rate of auto coding is more than 95%. Conclusion: We will get the standard, effective and accurate oncology specialist automatic coding system, then we will modify it for the whole classification of diseases.