Bispectral Analysis of Electroencephalogram Using Neural Networks to Assess the Depth of Anesthesia


Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription Access

Abstract

The article reviews algorithms of bispectral analysis of the electroencephalogram (EEG) signal of a patient to determine the level of brain activity during sedative-assisted treatment. The proposed algorithms are based on construction of multiple convolutions of complex amplitudes of the EEG signal, combined into so-called bispectra. Artificial neural networks (ANNs) are used to perform bispectral analysis and form a conclusion on the degree of patient brain activity. The article also shows individual results of functioning of the algorithms on real EEG signals and compares these results with expert judgments of doctors (anesthesiologists and neurophysiologists).

About the authors

N. G. Lavrov

Triton Electronics LLC; Krasovsky Institute of Mathematics and Mechanics; Ural Federal University

Author for correspondence.
Email: lavrov_ng@mail.ru
Russian Federation, Ekaterinburg; Ekaterinburg; Ekaterinburg

V. V. Bulaev

Triton Electronics LLC

Email: lavrov_ng@mail.ru
Russian Federation, Ekaterinburg

E. N. Solouhin

Triton Electronics LLC

Email: lavrov_ng@mail.ru
Russian Federation, Ekaterinburg

S. A. Taratuhin

Triton Electronics LLC

Email: lavrov_ng@mail.ru
Russian Federation, Ekaterinburg

A. V. Chistyakov

Triton Electronics LLC

Email: lavrov_ng@mail.ru
Russian Federation, Ekaterinburg

Supplementary files

Supplementary Files
Action
1. JATS XML

Copyright (c) 2016 Springer Science+Business Media New York