Comparative Analysis of Methods of Nonlinear Dynamics for the Problem of Monitoring the State of Anesthesia During an Operation

Monitoring the state of anesthesia is in important task during operations. During operations required maintaining state of deep anesthesia and preventing early release of him. Follow for the state of anesthesia can with analysis of EEG signal. Due to the fact that the development coming out of anesthesia leads to serious consequences for the human body the development of new algorithms for monitoring anesthesia is an argent task. In our previous studies we used entropy characteristics to analyze the stages of anesthesia [1], [2], but it limited the assessment of the possibility of using the severity of chaotic components in EEG signals. In this work we conducted a comparative analysis of the effectiveness of various methods of nonlinear dynamics for recognizing the depth of anesthesia EEG signals. The goal of this work is to develop and study nonlinear dynamics methods for medical computer systems capable of implementing monitoring the depth of anesthesia using EEG. To achieve this goal, the choice of nonlinear methods was justified, and a set of classification criteria was formed, on the basis of which the task was solved. As a result of the work were received diagnostic indicators were obtained that can be used to assess the severity of chaotic components. The effectiveness of using nonlinear methods is compared and it is shown that these methods provide high accuracy in recognizing the stages of anesthesia.