Electronic Health Record Integration

Abstract

The advancement in Information Technology has increased the adoption of Electronic Medical Record (EMR) systems in the healthcare industry. The improved access of readily available information in EMR provides a well-coordinated and collaborative effort that contributes greatly to the timeliness of care and better quality of healthcare delivery. However, the high volume of healthcare data that are being produced and shared have led to a need to condense data so that they are easily manageable and perceivable with higher performance and better storage efficiency than currently possible in traditional relational databases. Linked data is one of the many use cases which uses universal RDF/JSON format to describe things and to connect data from heterogeneous systems or open data for theoretical and advanced analytics. Recent literatures showed that the emerging NoSQL (Not Only SQL) databases offer significant advantages such as simple and automatic scaling, greater performance with high availability which address the limitations of relational databases. Yet, there are limited studies that have evaluated the applicability of NoSQL database systems in healthcare domain. The present study aimed to propose a novel way of visualizing and linking EMR data by developing a NoSQL graph database using Neo4j. Overall, the study demonstrated that NoSQL systems such as Neo4j graph database have many technical and financial advantages over relational database systems and are more suitable for data visualization as well as better data storage and management alternative for large scale data-intensive applications such as EMR database systems.