A New Recongnition System Based on Gabor Wavelet Transform for Shockable Electrocardiograms
Takayuki Okai *
Tokyo City University, 1-28-1 Tamazutsumi, Setagaya-ku, Tokyo, Japan
Shonosuke Akimoto
Tokyo City University, 1-28-1 Tamazutsumi, Setagaya-ku, Tokyo, Japan
Hidetoshi Oya
Tokyo City University, 1-28-1 Tamazutsumi, Setagaya-ku, Tokyo, Japan
Kazushi Nakano
The University of Electro-Communications, 1-5-1 Chofugaoka, Chofu-shi, Tokyo, Japan.
Hiroshi Miyauchi
Foundation for Ambulance Service Development, 4-6 Minamiosawa, Hachioji-shi, Tokyo, Japan
Yoshikatsu Hoshi
Tokyo City University, 1-28-1 Tamazutsumi, Setagaya-ku, Tokyo, Japan.
*Author to whom correspondence should be addressed.
Abstract
This paper presents a new recognition system for shockable arrhythmias for patients suffering from sudden cardiac arrest. In order to develop the recognition system, lots of electrocardiogram (ECGs) have been analyzed by using gabor wavelet transform (GWT). Although, there is a huge number of spectrum feature parameters, recognition performance for all combinations for spectrum feature parameters are evaluated, and on the basis of the evaluation results, useful and effective spectrum features for ECGs are extracted. As a result, the proposed recognition system based on the selected effective spectrum feature parameters can achieved good performance comparing with the existing results.
Keywords: Recognition system, shockable ECGs, effective spectrum feature parameters, improvement of recognition performance, wavelet transform