EEG UNDER ANESTHESIA A general method for calculation of depth of anesthesia

摘要

This paper investigated the problem of automatic depth of anesthesia (DOA) estimation from electroencephalogram (EEG) recordings. Compared with the Bispectral Index (BIS), time-frequency domain signal processing technique and nonlinear dynamical analysis were combined for DOA assessment, multiple features were extracted from EEG, Lasso and Logistic regression were used to classify and calculate the index of DOA and evaluate its relationship with EEG features. In emulation and clinical practice, the index of DOA is very close to BIS. This method can enhance existing monitoring devices and work as a general method to find effective features for calculation of DOA.