Miss Tia Haddad

Research project: A privacy-preserving microservices architecture to support epidemiologic research using large health datasets

Abstract

An epidemiology research project's success depends on the efficiency of its data processing workflow and its ability to create value. It is often necessary to harmonise data from multiple sources, discover patterns and efficiently compute useful projections. There is a significant gap between the potential of these datasets and their actual use, despite the advancements in the software engineering domain. Considering the developments in the Microservices architecture (MSA), the Cloud computing paradigm and Artificial Intelligence tools, more scalable, cost-effective, and distributed processing approaches can be explored for healthcare data processing. This research is aimed at developing an innovative healthcare data processing architecture by utilising microservices-based, hybrid-cloud-hosted collaborative data processing pipelines to improve scalability, efficiency, and privacy-preserving characteristics. As case studies prominent epidemiological data where warehouse systems' workflow will be studied and modelled. We then wish to explore the viability of multiple microservices data processing pipeline designs to replicate the existing workflows, introduce process improvements and process reengineering, with further developments focusing on expanding the proposed architecture into a hybrid-cloud environment, adding in-memory processing capabilities and AI features for autonomous load balancing. This study will yield architecture designs and potential data processing frameworks that will be beneficial to multiple entities involved in healthcare networks.