Power Spectral Study of Heart Rate Variability Time Series by the Adaptive Modified Continuous Morlet Wavelet Transform

Abstract

In this paper, a newly developed method referred as adaptive modified continuous Morlet wavelet transform has been proposed to improve the energy concentration in the time-frequency domain. The improvement is achieved by introducing the shape parameter to the Morlet function that is known as modified Morlet function. To make it adaptive, the modified Morlet function was optimized by concentration measure based on automatic determination algorithm. The proposed method was validated on set of synthetic time series signals like stationary signal and non-stationary signals with fast changing frequency and slow varying frequency components of signals. Also, this method is tested on all synthetic signals contaminated with additive white Gaussian noise with signal to noise ratio (SNR) of 30dB. The results show that the proposed method improves energy concentration in time-frequency domain compared to standard Morlet wavelet transform, standard Stockwell Transform (S-Transform), adaptive S-Transform and modified S-Transform. Further, this method has been used for analysis of heart rate variability (HRV) time series signals for estimating the value of mean power in VLF (0.004 Hz–0.04 Hz), LF (0.04 Hz–0.15 Hz), HF (0.15 Hz–0.4 Hz) and LF/HF ratio in frequency band of HRV spectrum. For this analysis, two group of healthy subjects, 21 young [10 M self-recorded, 11 (5 M+6 F) Fantansia database, age range 23–32] and 16 elderly [(9 M+ 7 F) Fantansia database, age range 70–82)] are used.

Keywords: Adaptive continuous Morlet wavelet transform, energy concentration, global method, a shape parameter

Cite this Article

R.S. Singh, B.S. Saini, R.K. Sunkaria. Power Spectral Study of Heart Rate Variability Time Series by the Adaptive Modified Continuous Morlet Wavelet Transform. Research & Reviews: Journal of Medical Science and Technology. 2017; 6(2): 5–20p.