Modelling and multivariable control in anaesthesia using neural-fuzzy paradigms. Part I. Classification of depth of anaesthesia and development of a patient model.

Abstract

The first part of this research relates to two strands: classification of depth of anaesthesia (DOA) and the modelling of patient's vital signs. First, a fuzzy relational classifier was developed to classify a set of wavelet-extracted features from the auditory evoked potential (AEP) into different levels of DOA. Second, a hybrid patient model using Takagi-Sugeno Kang fuzzy models was developed. This model relates the heart rate, the systolic arterial pressure and the AEP features with the effect concentrations of the anaesthetic drug propofol and the analgesic drug remifentanil. The surgical stimulus effect was incorporated into the patient model using Mamdani fuzzy models. The result of this study is a comprehensive patient model which predicts the effects of the above two drugs on DOA while monitoring several vital patient's signs. This model will form the basis for the development of a multivariable closed-loop control algorithm which administers "optimally" the above two drugs simultaneously in the operating theatre during surgery.