An improved blood gas intelligent hybrid model for mechanically ventilated patients in the intensive care unit

Authors:
A. Wang;G. Panoutsos;M. Mahfouf;G. H. Mills
Affiliations:
Department of Automatic Control and Systems Engineering, University of Sheffield, Sheffield, UK;Department of Automatic Control and Systems Engineering, University of Sheffield, Sheffield, UK;Department of Automatic Control and Systems Engineering, University of Sheffield, Sheffield, UK;Intensive Care Unit, Royal Hallamshire Hospital, Sheffield, UK
Venue:
BIEN '07 Proceedings of the fifth IASTED International Conference: biomedical engineering
Year:
2007

Citing 3
Cited 1

Adaptive ventilator Fi02 advisor: use of non-invasive estimations of shunt

Artificial Intelligence in Medicine
A blood gas hybrid model for ventilated patients in ICU with new formulations for dead space and tidal volume

BioMed'06 Proceedings of the 24th IASTED international conference on Biomedical engineering
SIVA: a hybrid knowledge-and-model-based advisory system for intensive care ventilators

IEEE Transactions on Information Technology in Biomedicine

Intelligent model-based advisory system for the management of ventilated intensive care patients. Part II: Advisory system design and evaluation

Computer Methods and Programs in Biomedicine

Quantified Score

Hi-index	0.00

Visualization

Abstract

SOPAVent (Simulation of Patients under Artificial Ventilation) is a blood gas model for mechanically ventilated patients in intensive care units (ICU). A set of parameters is required before the model is able to predict the patient's blood gases. Although most of the parameters required can be obtained from routine ICU measurements, some have to be estimated, such as shunt, relative dead space (Kd) and carbon dioxide production (VCO2), for such parameters measurements are not always available. In this paper, some recent work on improving the model parameter prediction and data collection is presented, with the aim of enabling the model to be used in routine ICU environments (i.e. be independent of any invasive measurements). Modifications include the development of a new data-driven model for VCO2 prediction, the update of the current Kd model and shunt estimation method and finally the development of automatic data collection algorithms. Based on these improvements, the new SOPAVent model is validated against blood gas measurements from 12 patients under different ventilator settings and the results show that the new SOPAVent model performance is better compared to the previous model, hence providing more accurate and reliable predictions of blood gases.