Development Of Feature Extraction Framework For Real Time Wireless ECG Signal For Automated Patient Monitoring System

Abstract

Electrocardiogram (ECG) is considered as one of the most important parameter during diagnosis. ECG has
been considered as the most basic and low cost bio signal tool in automated patient monitoring system.
Accurate diagnosis in most critical scenarios depends on analysis of patience vitals which are captured through
various devices. These vitals when transmitted through wired technology has very little chances of error in
transmission. However, for wireless transmission, these vital can be significantly altered due to noise in the
environment, properties of wireless medium, delay in wireless transmission, resulting in erroneous diagnosis.
Effective feature extraction of the transmitted signal on the receiver side reduces the risk of signal alteration in
wireless transmission. In this paper, an efficient feature extraction framework of real time wireless ECG signal
for automated patient monitoring system has been investigated and developed. The Finite Ridgelet Transform
(FRIT), for feature extraction provides 100% accuracy as compared to existing other feature extraction
methods.