Introduction
Mathematical models based on the physiology, namely, physiological models, have a crucial role in understanding the underlying mechanisms. When programmed as a software with a graphical user interface, these so-called computational models, or virtual patients, could be used to teach cardiorespiratory physiology and ventilation, determine optimal ventilation management as well as forecast the effect of various ventilatory support strategies (1–3). Furthermore, due to its virtual nature, prediction over a large amount of time can be summarized in a few minutes, making the assessment of the model more convenient than a real time assessment. Currently, there is no validated virtual patient specifically designed for modelling children cardiorespiratory system. Thus, our research team developed a cardiorespiratory simulator for children called “SimulResp” (figure 1) (2,4). SimulResp provides cardiorespiratory parameters, such as blood gases values, while simulating spontaneous and artificial ventilation situations for patients of various ages and weights with several pathological conditions. This simulator is based on physiological principles. Before a widespread use in respiratory status forecasting, this simulator must be validated (5). According to Summers et al. (6,7), the quality of a physiologic model is evaluated by three specific criteria: 1) qualitatively, which relates to the model’s ability to provide directionally appropriate predictions; 2) quantitatively in steady states and 3) in dynamics, which is the ability of the model to provide accurate predictions in steady state situations as well as dynamic transitions. The purpose of this study was to evaluate the quality of SimulResp according to these criteria in a pediatric critical care population.