Biomedical Ontology Matching as a Service

Ontology matching is among the core techniques used for integration and interoperability resolution between biomedical systems. However, due to the excess usage and ever-evolving nature of biomedical data, ontologies are becoming large-scale, and complex; consequently, requiring scalable computational environments with performance and availability in mind. In this paper, we present a cloud-based ontology matching system for biomedical ontologies that provides ontology matching as a service. Our proposed system implements parallelism at various levels to improve the overall ontology matching performance especially for large-scale biomedical ontologies and incorporates third-party resources UMLS and Wordnet for comprehensive matched results. Matched results are delivered to the service consumer as bridge ontology and preserved in ubiquitous ontology repository for future request. We evaluate our system by consuming the matching service in an interoperability engine of a clinical decision support system (CDSS), which generates mapping requests for FMA and NCI biomedical ontologies.