gms | German Medical Science

33rd International Congress on Electrocardiology

International Society of Electrocardiology

Time-frequency profiles of QRS complex with wavelet transform for estimating arrhythmogenic signals and its meanings assessed by computer simulation

Meeting Abstract

  • corresponding author presenting/speaker D. Wakatsuki - Showa University, Yokohama, Japan
  • T. Tsutsumi - Showa University, Yokohama, Japan
  • H. Shimojima - Showa University, Yokohama, Japan
  • Y. Takeyama - Showa University, Yokohama, Japan
  • Y. Okamoto - Chiba Institute of Technology, Narashino-shi, Japan

33rd International Congress on Electrocardiology. Cologne, 28.06.-01.07.2006. Düsseldorf, Köln: German Medical Science; 2007. Doc06ice085

The electronic version of this article is the complete one and can be found online at:

Published: February 8, 2007

© 2007 Wakatsuki et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Purpose: There is a possibility that any arrhythmogenic information is concealed in the time-frequency profile of QRS. Under this supposition, it was measured by wavelet transform( WT ) from the patients with anterior myocardial infarction( MI ). To assess the results of clinical study, a computer simulation was performed.

Methods: Standard 12 leads electrocardiogram( ECG ) were recorded under condition of filter off, sampling rate at 20KHz from normal volunteers( n=25 ), the patients with MI( n=25 ) and complicated with ventricular arrhythmias( MI-A , n=7 ). The time-frequency profile of ECG using WT from 10 to 500Hz was measured by our developed PC software. The time-frequency power and the integral values between QRS( 100msec interval ) were calculated. Number of spectral hole were counted in the WT signals. Computer simulation were performed using a square shaped two dimensional myocardium whose electrical properties was reproduced by 65000 Luo-Rudy phase 1 model cells and bidomain model. Disk-shaped micro-necroses modeled as passive resistance with various radii were scattered in the model and also the model assigned to slow conduction zone around the micro-necroses.

Results: Number of spectral holes were recognized 12% in NV, 63%in MI and 67% in MI-A in the lower frequency range( 100Hz or less). The appearances of the cut-off frequency adjacent to 100Hz were seen in MI-A. From the simulation study, the cut-off frequencies of the power spectrum were equal to the inverse of passing time of the excitation front through the simulated necroses. If the slow conduction zone was incorporated, the cut-off frequency was shifted to the lower frequency range where the spectral power was increased. Conclusion: Appearances of spectral hole and cut-off frequency may show the discontinuous nature of impulse propagation., that will offer the new information of arrhythmogenic signals in QRS.