Modeling and Development of Ontology-Based Information System to Support Medical Activity

Abstract

Medical information systems face significant challenges in semantic interoperability, knowledge reuse, and intelligent decision support. This paper presents the design and experimental evaluation of an ontology-based information system (OBIS) for supporting clinical activity. We construct a custom dataset of 100 patient records encompassing 24 clinical attributes, spanning 13 ontology disease classes and 19 medical departments. The proposed system integrates ICD-10, SNOMED CT, and LOINC terminologies within a unified ontological framework modeled using OWL 2 and SPARQL-based inference. Experimental results demonstrate a mean ontology inference accuracy of 89.57% and a semantic similarity score of 85.34%, with a mean query response time of 125.77 ms. A Random Forest classifier achieves the highest classification performance with 45.0% accuracy on the 13-class task. Comparative evaluation shows the proposed system surpasses traditional SQL databases, rule-based expert systems, and keyword-search IS by 17–24 percentage points in query accuracy. These findings validate the feasibility of ontology-driven architecture for enhancing semantic search, automated clinical coding, and decision support in healthcare information systems.