Development and application of Chinese medical ontology for diabetes mellitus

• DOI: 10.1186/s12911-023-02405-y
• Publication Date: 2024-01-19T13:02:17Z
• Authors: Jie Hu; Zixian Huang; Xuewen Ge; Yulin Shen; Yihan Xu; Zirui Zhang; Guangyin Zhou; Junjie Wang; Shan Lu; Yun Yu; Cheng Wan; Xin Zhang; Ruochen Huang; Yun Liu; Gong Cheng
• Keywords: Open Biomedical Ontologies

Abstract Objective To develop a Chinese Diabetes Mellitus Ontology (CDMO) and explore methods for constructing high-quality biomedical ontologies. Materials We used various data sources, including clinical practice guidelines, expert consensus, literature, hospital information system database schema, to build the CDMO. combined top-down bottom-up strategies integrated text mining cross-lingual ontology mapping. The was validated by experts development tools, its application through decision support natural language medical question answering. Results current CDMO consists of 3,752 classes, 182 fine-grained object properties with hierarchical relationships, 108 annotation properties, over 12,000 mappings other well-known ontologies in English. Based on we developed 200 rules diabetes diagnosis, treatment, diet, medication recommendations using Semantic Web Rule Language. By injecting knowledge, enhances performance T5 model real-world answering dataset related diabetes. Conclusion has semantic relationships extensive information, providing foundation artificial intelligence applications contexts, construction knowledge graphs, systems, automated Furthermore, process incorporated processing mapping improve quality improved efficiency. This workflow offers methodological reference efficient as well non-English