Model-Fitting for Automated and Robust Determination of Pulmonary Oxygen Exchange

A methodology was created for quantifying a subject's pulmonary oxygen exchange from P-IO2 and S-aO2 measurements, where P-IO2 is the partial pressure of inspired oxygen and S-aO2 the oxygen saturation level in arterial blood. Adapted from the procedures described by Sapsford and Jones in 1995, the algorithm involves fitting a two-compartment model of the respiratory system over available datapoints. However, compared with existing implementation strategies, the new program adopts a least squares criterion for model-fitting, and was demonstrated to be both robust and automated. In the future, it is envisaged that the program will be extended to consider models with higher degrees of freedom, and that it will be made freely available for general use.