Heart rate variability analysis using wavelet transform

Abstract

A method to analyze heart rate variability (HRV) signals using the wavelet transform is presented. The wavelet transform analysis is an alternative tool for analysis of nonstationary signals whose spectral characteristics change over time. The design and implementation includes a specially adapted decomposition-reconstruction filter bank for the frequency bands of interest. A time-scale representation is obtained for the very low frequency (VLF), low frequency (LF), medium frequency (MF), and high frequency (HF) bands. Results of the wavelet analysis are compared to classical methods, such as FFT and Blackman-Tukey, calculated on 1 hour of ECG recordings from the HMS-MIT-FFMS database. The authors' results suggest that the wavelet analysis could be an alternative tool to evaluate the time-frequency oscillatory components involved in the biological signals mentioned above.