Semantic Interoperability Framework for Multi-Center Healthcare Big Data Using Conceptual Modeling and SPARQL-Based Analysis

The exchange of health care data between multiple centers especially faces great challenges in attaining semantic integration. In the present research, this framework for semantic integration of multi-center real healthcare big data is contextualized with a conceptual modelling approach and subsequent facet analysis in SPARQL. Through conceptual modeling, the framework enshrines the provision of a consistent manner of presenting disparate health care data to facilitate data transfer across various systems. The semantic alignment process improves the integrated data by increasing the accuracy and consistency to meet the efficient querying and analysis through SPARQL. The study to assess the efficiency of the proposed framework reflects the real healthcare centers ten data with more than 100, 000 records. The semantic mapping achieved was 97.5%, the data coherence achieved was 98 percent, and the average of interoperability was 96.8 percent. Query performance was of order with average time of 4.2 seconds for complex queries which is 20% faster than normal ways. Some of the usual problems like lack of specific demographic information and variations between the clinical vocabularies were figured out and solved during the calibration process. Consequently, this study establishes the validity of the proposed framework of setting up a concept-based and ontology-configured HL7 FHIR system, showing how the integration of disparate health care data enhances data quality thereby enhancing the ability to make sound clinical decisions. The presented methodology provides a solution for managing the issue of data fragmentation in healthcare organizations on a large scale. Future work will focus on using machine learning techniques to automatically establish correspondences and fuses the two models.