The use of sinusoidally changing workloads as an aid to identify dynamic properties of physiological systems

The methodology of using frequency analysis in order to identify types of system(s) is discussed. A number of models with their characteristic Nyquist plots are presented graphically, to facilitate the identification of various types of systems by pattern recognition. A small series of measurements of oxygen consumption and heart rate has been carried out to illustrate the possibilities and limitations of the use of frequency analysis in exercise physiology. With a step function method one can only establish whether a system has multiple components in parallel or not, but with the frequency analysis method one can identify whether a system has multiple components in parallel or in series. It was found that under moderate exercise, the response of oxygen consumption to exercise is a system with multiple components in series, but the response of heart rate to exercise is a system with multiple components in parallel. The data that were obtained are in agreement with those produced by other authors, but the curves to fit the observed data were obtained using second order linear transfer functions.