Computer algorithm for tracking ECG spectral dynamics in ventricular tachyarrhythmias

ECG spectral dynamics might reveal information about the organization of the underlying reentrant circuits in ventricular arrhythmias. We present a computer algorithm for tracking changes in the spectral dynamics of ventricular tachyarrhythmias applied to 5 recordings each of monomorphic ventricular tachycardia (MVT) and ventricular fibrillation (VF). Peaks in the spectra were continually tracked throughout the recording if the change in frequency at each time step was equal to or less than the spectral resolution (0.12 Hz) and the peak amplitude exceeded the spectral noise. The algorithm was used to quantify the number, frequency, amplitude and durations of the spectral peaks. In all cases MVT was characterised by a single peak lasting for the full duration of the recordings. VF had many more peaks with on average 4.6 peaks. The peaks evolved and diminished with continuous evolution of the longest duration peaks of on average 21.9 s.