Design and demonstration of a complex neonatal physiological model for testing of novel closed-loop inspired oxygen fraction controllers

Abstract

Recently published clinical trials document that manual control of oxygen fraction in the inspiratory gas in neonates is not prompt enough to react to the rapidly changing physiological status of a neonate. As a result, the arterial blood oxygen saturation exhibits significantly long periods when the actual oxygen saturation level goes outside the desired safe range. Simple closed-loop systems are able to optimize the inspiratory oxygen fraction in steady-state situations, but they do not perform well in the context of rapidly changing physiological parameters. As a consequence, new algorithms for the closed-loop control of the inspired oxygen fraction are being developed and are becoming available. The aim of our study was to create a physiologically-realistic model of a neonatal organism allowing more extensive bench testing of newly developed algorithms for oxygen control in neonates. The design of the model is based both on the theoretical and up-to-date knowledge of the physiological principles, as well as on the well-documented observations by the authors in the neonatal intensive care units. The simulated outputs of the model correlate well with the real situations observed in the clinical environment.