Detection and frequency characterization of ventricular late potentials through wavelet transform

The wavelet transform (WT) is employed for a time-frequency analysis of ventricular late potentials (VLPs). Normal subjects (N) and post myocardial infarction (MI) patients underwent the study. The WT procedure allowed the authors to decompose the vector magnitude (M(t)) signal in frequency details in a multiresolution time-scale plane: a good time resolution is obtained at higher frequencies, while at lower frequencies a good frequency resolution is achieved. Graphical and numerical post-processing allowed the comparison between normal and pathological groups: the presence of VLPs was mainly identified in details D/sub 5/ and D/sub 4/, corresponding to a frequency range between 125 and 500 Hz. The time interval of VLPs presence was variable from patient to patient. The WT is a powerful tool able to well characterize the M(t) signal both in time and in frequency domains, thus allowing a better insight in the pathological phenomenon of VLP's.