Real-time estimation of the spectral parameters of Heart Rate Variability

Spectral Heart Rate Variability (HRV) parameters, LF (low frequency) and HF (high frequency), have an important role in interpreting slower and faster heart rate modulations. An online analysis method of HRV spectral parameters based on the modified Hilbert–Huang Transform (HHT) is proposed in the paper. A number of novel methods have been put forward to meet the demand of causal pre-processing of interbeat time intervals (IBI) series prior to application of HHT. Also in the real-time implementation of the HHT which is the combination of the Empirical Mode Decomposition and Hilbert spectral analysis an original extrapolation method of intrinsic mode function related to LF and HF spectral parameters was applied. The proposed algorithm allows temporal estimation of HRV spectral parameters in real-time with delays being reduced up to 60% with respect to the Short Time Fourier Transform (STFT) analysis. Such reduction in analysis delay can have an important significance in a number of cardiologic invasive procedures, e.g. in cardio-resynchronisation therapy (CRT).