Comprehensive Summary
This paper introduces a semantic model derived from ontologies to facilitate disease surveillance and pandemic intelligence by combining fragmented health information from multiple sources such as public health reports, flight data, and mass gatherings. The model adopts a three-layer ontology architecture—reference, application-specific, and domain-specific—to enrich formal and informal information semantically, produce linked data, and build interoperable knowledge graphs. Demonstrated via a case study of a COVID-19 outbreak caused by an Italian football game, the system was shown to integrate contextual information (e.g., travel history, social activities, and health status) effectively to support inference, anomaly detection, and cross-domain comparison. While it is promising for enhanced pandemic preparedness and response, maintaining ontology, aligning it with different domains, and scaling up to broader scenarios are challenging.
Outcomes and Implications
The envisioned framework carries important medical consequences for public health observation and reaction. By making it easy to combine multiform health and contextual data seamlessly, it allows epidemiologists and healthcare policymakers to detect outbreaks earlier, track disease spreading more accurately, and forecast healthcare system loads. The knowledge graphs empowered by ontology allow for enhanced interpretation of patient-level data, such as symptoms, comorbidities, and exposure history, which can speed up diagnosis, case classification, and treatment planning. In addition, the system's capability to link epidemiological events (e.g., mass gatherings) with disease endpoints allows for targeted interventions, such as risk estimation, containment, and resource planning. Ultimately, this semantic approach has the potential to enhance evidence-based decision-making in clinical and public health care, presenting a scalable tool for managing current and future infectious disease threats.