Real-Time analysis of ECG data using mobile data stream management system

  • Authors:
  • Seokjin Hong;Rana Prasad Sahu;M. R. Srikanth;Supriya Mandal;Kyoung-Gu Woo;Il-Pyung Park

  • Affiliations:
  • Samsung Advanced Institute of Technology, South Korea;Samsung Advanced Institute of Technology, South Korea;Samsung Advanced Institute of Technology, South Korea;Samsung Advanced Institute of Technology, South Korea;Samsung Advanced Institute of Technology, South Korea;Samsung Advanced Institute of Technology, South Korea

  • Venue:
  • DASFAA'12 Proceedings of the 17th international conference on Database Systems for Advanced Applications - Volume Part II
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Monitoring and analyzing electrocardiogram(ECG) signals for the purpose of detecting cardiac arrhythmia is a challenging task, and often requires a Complex Event Processing (CEP) system to analyze real-time streamed data. Various server-based CEP engines exist today. However, they have practical limitations to be used in environments where network connectivity is poor and yet continuous real-time monitoring and analysis is critical. In this paper, we introduce a lightweight mobile-based CEP engine called Mobile Data Stream Management System (MDSMS) that runs on the smart phone. MDSMS is built on an extensible architecture with concepts such as lightweight scheduling and efficient tuple representation. MDSMS enables developers to easily incorporate domain specific functionalities with User Defined Operator (UDO) and User Defined Function (UDF). MDSMS also has other useful features, such as mechanisms for archiving streamed data in local or remote data stores. We also show effectiveness of our MDSMS by implementing a portable, continuous, and real-time cardiac arrhythmia detection system based on the MDSMS. The system consists of ECG sensor and a smart phone connected to each other via a wireless connection. MDSMS can detect and classify various arrhythmia conditions from ECG streams by executing arrhythmia detection algorithms written in Continuous Query Language.