糖尿病领域本体构建与语义推理实现

目的：构建糖尿病领域本体库与糖尿病诊疗规则库，在此基础上实现语义推理，糖尿病领域知识的重用、潜在知识的揭示及共享，为后续糖尿病管理系统研究奠定基础，进而为基层全科医生诊疗糖尿病提供决策支持。方法：以国内糖尿病领域相关临床指南和领域专家知识为依据，抽取其中的概念以及概念之间的关系，借鉴七步法和骨架法，在斯坦福大学Protégé平台构建糖尿病领域本体，编写SWRL诊疗规则库，继而使用JESS推理机实现语义推理。结果：构建了较为完整的糖尿病领域本体库与糖尿病诊疗规则库，包含概念233条、实例205条、实例间关系16条、数值属性18条、SWRL规则28条，并在此基础上实现了语义推理。结论：构建的糖尿病领域本体能够实现语义推理任务，是将本体技术应用于慢病诊疗领域的有益探索。

Development of diabetes mellitus ontology and realization of semantic inference

Objective To lay the foundation for the successive research on diabetes mellitus (DM) management system and provide support for the general doctors at grass root level to make decisions by developing DM ontology database and DM diagnosis and treatment database for semantic inference, reuse of DM knowledge, revealing and sharing potential DM knowledge. Methods The DM ontology was established on the Stanford University Protégé Platform according to the 7-step method and skeletal method by extracting the concepts of DM ontology database and DM diagnosis and treatment database, and their relationship from domestic DM-related clinical guidelines and knowledge of DM experts. The SWRL diagnosis and treatment rules were then composed and the semantic inference was realized using the JESS inference engine. Results The developed DM ontology database and DM diagnosis and treatment database included 233 concepts, 205 examples, 16 relationships between examples, 18 data value properties, 28 SWRL rules, which could thus realize the semantic inference. Conclusion The developed DM ontology can realize semantic inference and is thus beneficial for the application of ontology technology in diagnosis and treatment of chronic disease.