Identification and parameter estimation of the mechanical ventilatory system.
Oesophageal pressure and mouth volume were measured by oesophageal balloon and pneumotachography, respectively, in 10 men (both normal and diseases) under different experimental conditions, i.e. spontaneous ventilation, quasi-static manoeuvres and respiration at high frequencies. The recorded data were then analyzed by an identification and parameter estimation computer programme, to determine a mathematical model for the process. Contrary to other recently published papers dealing with frequency analysis, we studied a time-dependent model, represented by a linear time-invariant differential equation. The results were quite satisfactory since a relatively second-order model seems to be adequate for describing the process. Differences in the parameter values between normal and diseased subjects were characterized by increases in the patients' lung viscous component. In conclusion, our identification test results in the mechanical ventilatory system seem to indicate that a simple second-order mathematical model is consistent with the experimental data.