Quantification of the impact of anesthesia by multivariate data analysis and nonlinear analysis of human brain electrical activity

Brain electrical activity (BEA) is modified during anesthesia in different areas and with different intensities depending of the drugs being used. Therefore, quantification of BEA is related to drug-effect and could be related to drug-concentration. However, the quantification of drug-effect is not a straightforward procedure. Furthermore, there is no capability to measure the drug concentration of an intravenous (iv) anesthetic in real time.The entropy content was estimated by using a fast and sensitive method to calculate changes in the EEG entropy as a function of the predicted drug plasma concentration. The detection and quantification of burst suppression was estimated by changes in EEG gaussianity. The use of the EEG as a surrogate measure of effect-site drug concentration in real time was calculated using the empirical mode decomposition (EMD) method, which separates the EEG intrinsic modes of oscillation (IMF), followed by the analysis of gaussianity changes in the IMF high frequency component (IMF1). These approaches were evaluated using data from 12 human volunteers who were anesthetized with propofol in a staircase fashion using a Target-Controlled Infusion (TCI) system.As a novel technique, the EMD and gaussianity analysis of IMF1 was found to be useful as a surrogate measure of effect-site drug concentration in real time. This novel noninvasive technique could be useful for monitoring and controlling anesthetic administration in anesthesia and intensive care.The purpose in this work was to analyze the behavior of the frontal BEA during different levels of anesthesia in human volunteers to explore the opportunity to use signals from this area as real time indicators of anesthetic state. This analysis is divided in three different approaches. As a primary approach, the quantification of the level of consciousness was made using entropy changes in the electroencephalogram (EEG) as a function of drug concentration. When the level of consciousness is significantly depressed, the EEG signal patterns fundamentally change and a secondary approach for the automatic detection and quantification of burst suppression was developed. Finally, as a third approach, the use of the EEG as a surrogate indicator of effect-site drug concentration in real time was determined.