Extraction of the respiration influence from the heart rate variability signal by means of lattice adaptive filter

The respiratory sinus arrhythmia that influences the heart rate variability (HRV) signal is well quantifiable, by means of frequency analysis, in stationary conditions, when the respiration rate is uniform. In many physiological and pathological conditions, however, the respiration is characterized by transitions and variations which compromise the hypothesis of stationarity required for a correct spectral analysis. Here, an adaptive filter implemented in lattice form is employed for predicting the influence of respiration on the HRV and for separating this contribution from the other ones, namely from the low frequency (LF) oscillations also present in the HRV. The filter was employed for studying the sympatho-vagal balance in normal subjects during sleep, and in particular during transitions between different sleep stages.<>