A Semantic Ontology-Driven Architecture for Personalized Health Insurance Assignment in Smart Healthcare Ecosystems

Abstract

Traditional health insurance models are often static, reactive, and poorly aligned with the dynamic and personalized nature of patient health data. These limitations are increasingly evident in modern healthcare ecosystems that demand real-time, context-aware services. Addressing this gap, this study proposes a novel semantic, ontology-driven architecture for personalized health insurance assignment, designed to adapt dynamically to electronic health records (EHR), clinical diagnoses, and real-time physiological data. Unlike prior systems that lack semantic reasoning or integration with heterogeneous data sources, the proposed framework integrates OWL-based ontologies, Semantic Web Rule Language (SWRL), and SPARQL querying into a multi-layered architecture involving IoT devices, fog computing, and cloud services. A key innovation lies in its ability to infer and assign composite insurance policies based on patient-specific conditions, supporting personalized, adaptive, and explainable decision-making. A prototype system was implemented using Protégé, Apache Jena, and the Pellet reasoner, and tested on five representative patient scenarios. Evaluation results demonstrate sub-second reasoning latency, high semantic accuracy, and robust policy alignment with clinical profiles. This framework advances value-based health insurance by enabling real-time policy automation, enhanced transparency, and scalable integration across smart healthcare domains. These results suggest that the system can support scalable integration into national health insurance platforms and improve patient-centered policy delivery.