A CONTINUOUSLY UPDATED HYBRID BLOOD GAS MODEL FOR VENTILATED PATIENTS

摘要

SOPAVent (Simulation of Patients under Artificial Ventilation) is a blood gas model for ventilated patients in intensive care units. In this paper, a continuously updated blood gas model based on SOPAVent is developed and extended to simulate patients' states from a stable to an unstable status. The changes in the patient's physiological state are simulated by updating the shunt and relative dead space in the patient model. The model was validated against blood gas measurements from real patients during a 7-hour period and with changes to the ventilator settings. Results show that, with a selector based on a time criterion, the autonomous model can represent the real patient with a changing physiological state and give good blood gas predictions.