Real-Time Analytics Pipelines for Healthcare using AWS Kinesis and Lambda

Authors: Anusha Joodala

DOI: https://doi.org/10.37082/IJIRMPS.v11.i1.232967

Short DOI: https://doi.org/

Country: United States

Full-text Research PDF File:   View   |   Download

Abstract: The need for immediate data processing and real time decision support arises from the rapid development of electronic health records, wearable devices, and IoT healthcare devices. This builds upon AWS Kinesis and AWS Lambda for serverless, scalable and event driven architecture for real time health care analytics. This analytics architecture allows the real time processing of health data and potential emergency health event. The architecture incorporates Kinesis Data Streams for stream ingestion, Kinesis Data Analytics for transformation and stream analytics, and Lambda functions for serverless computation to achieve low latency processing for near real time clinical insight. Vital sign data, telemetry, diagnostic images and data analytics streams identifying potential anomalies like cardiac and respiratory distress constitute real time data streams. This architecture relieves routine maintenance of the hardware by offering automatic scalable and highly available infrastructure, while load and service exposure are controlled through public partitioning with Amazon storage services, DynamoDB, and Quick Sight for analytics visualization. During experimental runs, the latency of the entire system was reduced by 45% and remained constant during varying data loads when compared to classical batch processing systems.

Keywords: Real-time analytics, AWS Kinesis, AWS Lambda, healthcare IoT, serverless computing, predictive diagnostics.

Paper Id: 232967

Published On: 2023-02-03

Published In: Volume 11, Issue 1, January-February 2023

Share this