Effects of propofol anesthesia on nonlinear properties of EEG: Time-lag and embedding dimension

Abstract

Depth of anesthesia monitors frequently use linear features. Though some non-linear methods such as spectral entropy and bispectrum analysis are also used; the application of methods derived after Chaos Theory is still pending. This paper focuses on the analysis of the embedding parameters required for the reconstruction of the chaotic atractor associated to the EEG signal. After clinical research and ethical committee approval and signed informed consent, EEG data was collected from 6 patients scheduled for surgery under general anesthesia and 7 ones scheduled for ambulatory endoscopic procedures. One differential channel of EEG (+ in mastoids M1/M2, - in the middle-line of the forehead FPz/AFz and referenced to F7/F8) was amplified with a gain of 10000 (Biopac MP100-EEG100B), low-pass filtered at 300 Hz (2^nd order Butterwoth) and acquired at 2.5 KHz. Data were digitally filtered (50 Hz comb filter and 120-taps linear phase FIR lowpass filter at 85 Hz) and decimated to 250 Hz prior to analysis under OpenTstool. Time-lag was evaluated through autocorrelation and auto-mutual information functions. False nearest neighbors, and Higuchi Fractal Dimension were used for the study of the embedding dimension. After analyzing the data with different window lengths (250, 500, 1000 and 2000 samples), time lag was found to decrease from the basal level with anesthesia levels. Embedding dimension after Cao’s method was not found to vary, due to the short length of the time series. Higuchi fractal dimension was able to show a slight dimensionality reduction with anesthesia. These results seem to indicate a decrease on the data-rate of the brain channel due to the reduction of the number of active neurons and the decrease of the neural conduction velocity induced by the anaesthetic agent used (propofol) at the GABAA receptors.