A QRS-wave recognition system using time-delay neural networks

Pao Yue-kong Library Electronic Theses Database

A QRS-wave recognition system using time-delay neural networks

 

Author: Wong, Ming-yee
Title: A QRS-wave recognition system using time-delay neural networks
Degree: M.Sc.
Year: 1996
Subject: Electrocardiography -- Interpretation
Neural networks (Computer science)
Delay lines
Hong Kong Polytechnic University -- Dissertations
Department: Multi-disciplinary Studies
Pages: vii, 102 p. : ill. ; 31 cm
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b1231331
URI: http://theses.lib.polyu.edu.hk/handle/200/3824
Abstract: The Electrocardiogram (ECG) is a standard tool for analyzing cardiac diseases. Recognition and intepretation of QRS-wave from ECG are tedious routines for medical physicians. More than 100,000 of cardiac cycles per patient are recorded in a day. The physicians have to search for only a few abnormal cardiac cycles to interpret this large amount of ECG data. Therefore, an accurate, reliable automatic ECG interpretation system is urgently needed. Such a system requires a very accurate QRS-wave recognition rate because the QRS-wave is the reference of the cardiac cycle and it is important for interpretation of the ECG data. Technically speaking, recognition of QRS-wave is difficult, not only because of the physiological variability of the QRS-waves, but also because of the various types of noise that can be present in the ECG signal. Noise sources include random noise such as electrical interference or base line drift due to the cutaneous current or due to loose contacts in any place in the circuit. Traditional methods of recognizing QRS-wave can be classified into three main categories: (1) mathematical method, (2) syntactic method and (3) neural networks adaptive matched filtering method. All of these existing methods rely on the bandpass filter to reduce the noise in the ECG signal. However, the reduction of noise by filtering is not always effective and reliable. Thus we seek another approach of QRS-wave recognition to improve the performance in this aspects. Since Time-Delay Neural Networks (TDNN) has been proved very efficient for recognizing patterns with temporal structure. It is appropriate to apply TDNN for QRS-wave recognition. In this project, we implemented a new QRS recognition system by using a simplified TDNN. The new QRS-wave recognition system has been tested to recognize QRS-wave from standard ECG signals. The results are encouraging. Better detection perfomance than the existing algorithms can be achieved by the new method. The new method has several advantages over the existing algorithms by inheriting (1) higher detection accuracy, (2) higher noise immunity and (3) simple structure and operation. In our experiment, a simple QRS-wave classifier that consists of two QRS-wave detectors has also been implemented. It shows the system has a reliable waveform classification ability. This ability makes it very useful in the development of automatic ECG interpreting system for cardiac diseases diagnosis.

Files in this item

Files Size Format
b12313312.pdf 3.639Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

     

Quick Search

Browse

More Information