By Topic

ECG beat classification using features extracted from teager energy functions in time and frequency domains

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $33
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
C. Kamath ; Electron. & Commun. Dept., Manipal Inst. of Technol., Manipal, India

It is hypothesised that a key characteristic of ECG signal is its non-linear dynamic behaviour and that the non-linear component changes more significantly between normal and arrhythmia conditions than the linear component. This study makes an attempt to analyse ECG beats from an energy point of view by accounting for the features derived from non-linear component in time and frequency domains using Teager energy operator (TEO). The key feature of TEO is that it models the energy of the source that generated the signal rather than the energy of the signal itself. Hence any deviations in the regular rhythmic activity of the heart get reflected in the Teager energy function. To show the validity of appropriate choice of features, t-tests and scatter plot are used. The Mests show significant statistical differences and scatter plot of mean of Teager energy in time domain against mean of Teager energy in frequency domain for the ECG beats evaluated on selected Manipal Institute of Technology-Beth Israel Hospital (MIT-BIH) database, which reveals an excellent separation of the features into five different classes: normal, left bundle branch block, right bundle branch block, premature ventricular contraction and paced beats. The neural network results achieved through only two non-linear features exhibit an average accuracy that exceeds 95%, average sensitivity of about 80% and average specificity of almost 100%.

Published in:

IET Signal Processing  (Volume:5 ,  Issue: 6 )