June 11, 2020 – Using Distributed Big Data of Hospital System in Analytics Platform for Resilient Interactive Clinical Services – Presented by Dillon Chrimes, University of Victoria
Big data analytics (BDA) is important to improve operations and reduce health care costs. There are many challenges to a BDA platform with data aggregation, system integration, data translation and analysis for reporting and dashboards, and overall maintenance. The objective of this study was to establish an interactive BDA platform with simulated patient data of a hospital system that identifies utilization of its big data with the technological components for resilient clinical services. This study amalgamated metadata from admission, discharge, and transfer (ADT) database, which represented patient care and movement in hospital, with the discharge abstract database (DAD) that represented provider diagnoses, length-of-stay, and re-admissions for comprehensive data on daily operations. Using open-source software technologies, framework of the platform was constructed with Hadoop Distributed File System (HDFS) using HBase database. Distributed data structures were generated within the framework to form hospital-specific metadata of patient records with each row as patient encounters and columns for data fields in ADT and DAD databases.