Title :
A method for the ventricular late potentials detection and characterisation using wavelets
Author :
Batista, Arnaldo ; English, Michael
Author_Institution :
Faculdade de Ciencias e Tecnologia, Monte de Caparica, Portugal
fDate :
31 Oct-3 Nov 1996
Abstract :
In this paper we present a method for the ventricular late potentials (VLP) characterisation using the discrete wavelet transform with the Daubechies wavelets and the Newland´s (1993) harmonic and musical wavelets. The method is based on a wavelet multiresolution scheme where the energy content of the beat-by-beat ST and the TP segments of the high resolution electrocardiogram (HR-ECG) are compared at different resolutions. An objective figure, the factor of normality is generated for ready clinical use. These results were compared to the ones obtained with the application of the spectrogram, and the Choi-Williams transform, using the spectro-temporal Mapping, for the VLP´s study. A high resolution system for the acquisition of the high resolution electrocardiogram (HR-ECG) has been set up, with amplitude and group delay error less than 1% for which a multirate technique was used. In this way we could guarantee that our data were adequate to be submitted to the time-frequency and time-scale (wavelet) analysis. The factor of normality generated by the multiresolution scheme proved to be more sensitive, in our study cases, but the method requires further clinical testing
Keywords :
electrocardiography; medical signal processing; signal representation; signal resolution; time-frequency analysis; wavelet transforms; Daubechies wavelets; Newland´s wavelets; TP segments; beat-by-beat ST; discrete wavelet transform; energy content; factor of normality; harmonic wavelets; high resolution electrocardiogram; multirate technique; musical wavelets; ventricular late potentials characterisation; ventricular late potentials detection; wavelet multiresolution scheme; Biomedical engineering; Delay; Discrete wavelet transforms; Energy resolution; Hafnium; Noise level; Signal resolution; Spectrogram; Time frequency analysis; Wavelet analysis;
Conference_Titel :
Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE
Conference_Location :
Amsterdam
Print_ISBN :
0-7803-3811-1
DOI :
10.1109/IEMBS.1996.647462