An encoding methodology for medical knowledge using SNOMED CT ontology

Journal of King Saud University: Computer and Information Sciences

Mohammed Elmogy¹  Shaker El-Sappagh² 

[1] Faculty of Computers and Information, Mansoura University, Egypt;Faculty of Computers and Information, Minia University, Egypt;

关键词: Clinical decision support system (CDSS);    SNOMED CT (SCT) coding;    Semantic data retrieval;    Ontology;    Case-based reasoning (CBR);    Diabetes diagnosis;   

DOI  :  10.1016/j.jksuci.2015.10.002

来源: DOAJ

【 摘 要 】

Knowledge-Intensive Case Based Reasoning (KI-CBR) systems mainly depend on ontology. Using ontology as domain knowledge supports the implementation of semantically-intelligent case retrieval algorithms. The case-based knowledge must be encoded with the same concepts of the domain ontology. Standard medical ontologies, such as SNOMED CT (SCT), can play the role of domain ontology to enhance case representation and retrieval. This study has three stages. First, we propose an encoding methodology using SCT. Second, this methodology is used to encode the case-based knowledge. Third, all the used SCT concepts are collected in a reference set, and an OWL2 ontology of 550 pre-coordinated concepts is proposed. A diabetes diagnosis is chosen as a case study of our proposed framework. SCT is used to provide a pre-coordination concept coverage of ∼75% for diabetes diagnosis terms. Whereas, the uncovered concepts in SCT are proposed. The resulting OWL2 ontology will be used as domain knowledge representation in diabetes diagnosis CBR systems. The proposed framework is tested by using 60 real cases.

【 授权许可】

Unknown