Healthcare data analytics using traditional and Hadoop frameworks for smart healthcare decision making

Abstract

This study aimed to describe the healthcare analytics data infrastructure used by organizations to carry out report and analysis regarding their healthcare data. Healthcare data involved data generated from electronic medical records which are structured and unstructured in nature. There is need for healthcare organization to make use of effective analytical techniques (i.e. big data analytics or traditional approach) in health sector which provides health practitioners and stakeholders with new insights that have the potential to advance personalized care to improve patient outcomes and avoid unnecessary costs. This study aimed to carry out a comparative analysis regarding healthcare data analytics between Traditional system that adapted relational database management system (RDBMs) using MySQL and big data approach using Hadoop/Mapreduce. Based on the comparison metrices, the analysis was conducted with a traditional relational databases tool installed on a stand-alone system, such as a desktop or laptop. But using big data Analytic tools and platforms, processing is broken down and executed across multiple nodes. However, after detailed comparative analyses between the two approaches have been done, the research found out that Hadoop is the best technique for handling Big Data compared to that of Mysql. As day by day, the data used increases and therefore a better way of handling such huge amount of data is becoming a hectic task.

Keywords: Big data, Big Data Analytics, Hadoop/MapReduce, HDFS, Healthcare data, Structured data, Unstructured data, RDBMs, Traditional Relational Database.