Estimation and power spectral analysis of heart instantaneous frequency (HIF) - a wavelet approach

Abstract

Investigation of heart rate variability (HRV) has been the subject of considerable interest in the management of cardiac arrhythmia. Digital signal processing using instantaneous frequency and multiresolution analysis (Daubechies (db4) wavelet transforms) has been applied to estimate HRV without R point detection. The proposed method requires less memory for long term analysis as the ECG was measured with only 4 Hz sampling. Such a saving in memory makes it especially convenient to apply for long term analysis. The analysis of HRV using wavelet transforms was found to be very effective over conventional IIR filters. The sensitivity of the LF/HF ratio parameter increases with the Daubeches (db) based wavelet filter.