Title :
Comparison of discrete wavelet and Fourier transforms for ECG beat classification
Author :
Dokur, Z. ; Ölmez, T. ; Yazgan, E.
Author_Institution :
Dept. of Electron. & Commun. Eng., Istanbul Tech. Univ., Turkey
fDate :
9/2/1999 12:00:00 AM
Abstract :
Two feature extraction methods, Fourier and wavelet analyses for ECG beat classification, are comparatively investigated. ECG features are searched by dynamic programming according to the divergence values. 10 types of ECG beat from an MTT-BIH database are classified with a success of 97% using a restricted Coulomb energy neural network trained by genetic algorithms
Keywords :
discrete Fourier transforms; discrete wavelet transforms; dynamic programming; electrocardiography; feature extraction; genetic algorithms; medical signal processing; neural nets; signal classification; Coulomb energy neural network; ECG beat classification; MTT-BIH database; discrete Fourier transform; discrete wavelet transform; dynamic programming; feature extraction; genetic algorithm;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:19991095
