Real-time estimation of depth of anaesthesia using the mutual information of electroencephalograms

Proposes a novel approach to monitor the depth of anesthesia by predicting the response to incision during isoflurane anaesthesia using mutual information (MI) time series or electroencephalograms (EEGs) and their complexity analysis. The MI between four lead electrodes was first computed using the EEG time series. The Lempel-Ziv complexity measures, C(n)s, were extracted from the MI time series. Prediction was made by means of artificial neural network (ANN). Training and testing the ANN used the 'drop-one-patient' method. 98 consenting patient experiments show the system was able to correctly classify purposeful response in average accuracy of 91.84% of the cases and the method has a better performance than other methods, such as spectral edge frequency, median frequency, and bispectral analysis. This method is computationally fast and acceptable real-time clinical performance was obtained.