Nonlinear electroencephalogram signal analysis method and device

Abstract

The invention discloses an electroencephalogram signal analysis method and an electroencephalogram signal analysis device to identify the brain state. The method comprises the following steps of: segmenting the acquired electroencephalogram signal; analyzing the state of each segment of electroencephalogram signal and preprocessing to remove interference; and calculating an ordering entropy of each segment of electroencephalogram signal, and determining the activity state of the brain according to ordering entropy values. The invention can be used for analyzing electroencephalogram signals under the conditions of different cognitive function activities of the brain and physiological statues, and can be particularly applied to studying the changes of brain activities in sleep and the like. Starting from a phase space, ordering entropies are adopted for statistical analysis of the structural change modes of the electroencephalogram signals to disclose the uncertainty, stable degree and information content of the electroencephalogram signals.

Description

technical field [0001] The invention relates to an electroencephalogram signal analysis method and device to identify brain states, in particular to a method and device for preprocessing and sorting entropy analysis of electroencephalogram sequences. Background technique [0002] EEG signals express the electrical activity of brain nerve cells, directly reflecting the electrophysiological activity of brain tissue and the functional state of the brain. Research has shown that different brain states can be identified through in-depth analysis of EEG signals. At present, the main methods of EEG signal analysis are: time-domain analysis, frequency-domain analysis, time-frequency analysis, information entropy and nonlinear dynamics. An overview is as follows: [0003] (1) The time-domain analysis method extracts features directly from the time domain, which is the earliest developed method. It is intuitive and has a relatively clear physical meaning. Therefore, there are still ...

AI Technical Summary

Problems solved by technology

Numerous studies have confirmed that MF and SEF are closely related to brain state; however, these parameters also vary among different individuals

[0005] (3) EEG signals are time-varying and non-stationary signals. Simple time-domain or frequency-domain representations cannot accurately express signals. Only by combining time and frequency for processing can we better extract useful information from EEG signals. information

Although the application of time domain and frequency domain methods can see significant results, the accuracy rate is not high. In addition, the premise of these methods is to assume that EEG is a linear stationary signal, while the actual EEG signal comes from the nonlinear dynamic system of the brain Non-linear and non-stationary signals, so information entropy and nonlinear dynamics methods have been developed rapidly and achieved good results. However, these analysis methods still need a lot of research to verify their effectiveness.

Images