Dimensional Characterization of Anesthesia Dynamic in Reconstructed Embedding Space

The depth of anesthesia quantification has been one of the most research interests in the field of EEG signal processing and nonlinear dynamical analysis has emerged as a novel method for the study of complex systems in the past few decades. In this investigation we use the concept of nonlinear time series analysis techniques to reconstruct the attractor of anesthesia from EEG signal which have been obtained from different hypnotic states during surgery to give a characterization of the dimensional complexity of EEG by Correlation Dimension estimation. The dimension of the anesthesia strange attractor can be thought of as a measure of the degrees of freedom or the 'complexity' of the dynamics at different hypnotic levels. The results imply that for awaked state the correlation dimension is high, On the other hand, for light, moderate and deep hypnotic states these values decrease respectively; which means for anesthetized situation we expect lower correlation dimension.