A semantic-based EMRs integration framework for diagnosis decision-making

Abstract

Discovering latent information from Electronic Medical Records (EMRs) for guiding diagnosis decision making is a hot issue in the era of big data. An EMR composes of various data (e.g., patient information, medical history, diagnosis, treatments, symptoms), but most of them are stored in the relational database. It is difficult to integrate the data and infer new knowledge based on existing data structures. Semantic technology (ST) is a flexible and scalable method for integrating heterogeneous, distributed information from big data. Taking advantage of these features, this paper proposes a framework that leverages ontology to improve EMRs decision-making. A case study shows that this framework is feasible to integrate information, and can provide specific and personalized information services for facilitating medical diagnosis.