ABSTRACT
The unstable activation wavefront from the tissue responsible for the production of ventricular tachycardia (VT) gives rise to beat-variable signals components that are eluded during the averaging step of high resolution ECG (HRECG). We used a mathematical model of small variations in a largely repetitive waveform to evaluate the beat-to-beat variance of the HRECG signal. The ability of the Morlet Wavelet Transform to discriminate the different factors responsible for the beat-variability (the alignment error and the beat-variable signal component) has been assessed on simulated signals. The performance evaluation on real ECG signals from normal subjects and patients with a documented history of ventricular tachycardia showed that the dynamics of the arrhythmogenic substrate as revealed by wavelet transform offers a significant improvement in ventricular tachycardia risk assessment.
