Parameter extraction from heart rate and arterial blood pressure variability signals in dogs for the validation of a physiological model.

The paper describes an automatic procedure for improving the extraction of parameters in heart rate (HR) and arterial blood pressure (ABP) beat-to-beat variability signals. Auto- and cross-spectral analysis of such signals is carried out through parametric models and the distribution of the power of the spectra and the phase relationships are compared in various physiological situations induced in the dogs via drug infusion or surgical interventions which do influence the control mechanisms of HR and ABP. This "black-box" approach allows the obtention, directly from the processing of the above-mentioned signals, of the estimation of parameters relative to cardiovascular models, as the ones described by simple equations (Windkessel and Starling laws are introduced as examples). These parameters seem to validate significantly the capability of such models to describe the physiological interactions existing between the two considered signals. Applications may be foreseen both for research and clinical purposes.