Author(s)

D.J. Botes, P.J. Cilliers & G.J. Ras

Abstract

Computer aided parameter estimation for lung function diagnosis D.J. Botes", P.J. Cilliers" & G.J. Ras& ^Department of Electrical and Electronic Engineering, University of Pretoria, 0002 Pretoria, South Africa ^Department of Internal Medicine, University of Pretoria, 0002 Pretoria, South Africa ABSTRACT The flow resistance and compliance of the pulmonary system could be useful in assessing lung conditions. These parameters may be time-variant and non-linear. A comprehensive model to simulate all parts of the pulmonary system, including the endotracheal tube is presented here. Its feasibility for real-time estimation of resistance and compliance during assisted breathing is demonstrated. The model uses recursive estimation in a noisy environment to determine these lung parameters in real time. Results are presented which show a realistic prediction of lung parameters, as well as the non-linearity and varying nature thereof. INTRODUCTION In the assessment of

Keywords