Ontology-Aligned Representation of Adverse Events in a Multiple Myeloma Clinical Trial.

To explore the potential of ontologies in improving adverse event identification, we developed ontology-aligned representations of adverse events extracted from electronic health records (EHRs) of fifty patients participating in a multiple myeloma trial. We reused relevant biomedical ontologies that are built upon Basic Formal Ontology (BFO) and constructed representations in Resource Description Framework/Web Ontology Language (RDF/OWL). Adverse events for the selected cohort were extracted from both structured and unstructured EHR data stored in the Arkansas Clinical Data Repository (AR-CDR). This ontology-aligned approach supports automated reasoning and facilitates semantic interoperability.