Assessment of frequency shifts in R-R interval variability and respiration with complex demodulation

A complex demodulation (CDM) method for continuous assessment of frequency shifts and time-dependent changes in amplitude in the rhythmic components existing in predefined frequency bands was proposed and applied to the analysis of high-frequency (HF) and low-frequency (LF) components of the R-R interval and to the analysis of respiration via impedance spirogram. Simulation studies revealed that this CDM technique furnishes mathematical features well suited to the investigation of non-stationary R-R interval signals and can delineate time-dependent fluctuations in both amplitude and frequency, accurately differentiating between HF and LF components. Analysis of data during paced breathing at different respiratory frequencies revealed that the estimated frequency of the HF component and respiration faithfully reflected the frequency of paced breathing. Analysis of data during dynamic exercise with increasing workload (20 W/min) showed that the frequency of the HF component was elevated with exercise and that both HF and LF amplitudes were reduced progressively with advancing load. CDM-derived frequency and amplitude of respiration were highly correlated to direct breath-by-breath respiratory frequency and tidal volume measurements. We conclude that this method could provide a powerful means for continuously assessing time-dependent changes in both cardiovascular and respiratory variations.