Ontology based data integration and mapping for adverse drug reaction

There is an increasing interest to transform and integrate data obtained from variety of sources in the health care domain. This transformation and integration is particularly beneficial in adverse drug reaction (ADR) that may occur due to inhalation of a drug and requires quick treatment to reduce the risk of life. It has been proved in the literature that an expert system based on prior knowledge or supervised learning is beneficial for diagnosis of ADR. A ubiquitous semantic rich knowledge data set is essential for machine learning based clinical expert systems to deal with emergency in case of ADR. This paper highlights the method to integrate heterogeneous data from various clinical sources in ontological format for ADR. Thereafter, the data in ontological format is linked to various standard medical terminologies like SNOMED CT(Systematized Nomenclature of Medicine -Clinical Terms) and MedDRA (Medical Dictionary for Regulatory Activities) in order to make it more general, standardized and semantically rich with the help of TOP SPIN (TopQuadrant SPAQL Inferencing Notation) and SHACL (Shapes Constraint Language) rules. Furthermore the ontology is mapped to domain ontology OAE (Ontology for Adverse Events) to make it more powerful knowledgebase for query retrieval and searching. The objective behind this work is to obtain an integrated ontology or RDF (Resource Description Framework) Turtle format integrated data from heterogeneous resources (XML Data, Relational Data and ASCII Text Data). Further, it ensures linking to medical standards by utilizing different capabilities and plugins of Top Braid Composer maestro edition.