A Blood Gas Hybrid Model for Ventilated Patients in ICU with New Formulations for Dead Space and Tidal Volume

A. Wang, M. Mahfouf, and G.H. Mills (UK)

Keywords

Ventilated patient model, blood gas, non-invasive measurements, neuro-fuzzy, dead space, tidal volume.

Abstract

SOPAVent (Simulation of Patients under Artificial Ventilation) is a blood gas model for ventilated patients in intensive care units (ICU). This model requires an explicit value of dead space. However, Dead space is not routinely measured in ICU, and hence its value had to be derived using invasive measurements as well as a slow secant algorithm. In this paper, a non-invasive method for dead space estimation is developed. A non-linear adaptive neuro-fuzzy inference system (ANFIS) model was used to describe the relationship between the relative dead space (Kd) and five routinely measured variables such as PaCO2, VT, RR, Pinsp, PEEP. Furthermore, a new estimation method has also been developed using ANFIS so that the tidal volume (VT) value can be updated as soon as the patient’s model inputs are changed. This is used during the validation of the new SOPAvent model. The SOPAVent model was validated against the blood gases measurements from real patients under different ventilator settings and the results show that, with the good predictions of dead space and tidal volume, the current SOPAVent model can represent the ventilated patient’s state and give matched predictions under different ventilator settings.

Important Links:

Go Back