Electroencephalogram signal abnormity monitoring system and method

Abstract

The invention provides an electroencephalogram signal abnormity monitoring system and method. The electroencephalogram signal abnormity monitoring method comprises the following steps: filtering and denoising acquired electroencephalogram signals; performing empirical mode decomposition on the preprocessed electroencephalogram signals to obtain a series of intrinsic mode functions, and determining the number of decomposition layers of empirical mode decomposition of the preprocessed electroencephalogram signals; performing multi-resolution analysis on the preprocessed electroencephalogram signal according to the decomposition layer number to obtain a series of smooth signals and detail signals; performing signal fusion on the intrinsic mode function, the smooth signal and the detail signal to obtain a series of fusion components; respectively carrying out weighting processing according to the sample entropy and the correlation coefficient of the fusion component to obtain a weighted feature vector; training an extreme learning machine by taking the weighted feature vector as a training sample; and calculating a weighted feature vector of the to-be-judged electroencephalogram signal, inputting the weighted feature vector into the trained extreme learning machine, and outputting whether the to-be-judged electroencephalogram signal is abnormal or not. According to the method, misjudgment and missed judgment of the abnormal electroencephalogram signals are reduced.

Description

technical field [0001] The invention relates to the technical field of EEG signal monitoring, in particular to a system and method for monitoring abnormal EEG signals. Background technique [0002] Brain-computer interface is an EEG signal information exchange control channel established between the human brain and a computer or other electronic equipment by means of engineering technology. EEG signals are the external manifestations of human brain neural activity, which can be used to monitor brain activity and diagnose abnormal brain states. field of study. However, the existing abnormal EEG monitoring technology has the following shortcomings and deficiencies: [0003] 1. The abnormal monitoring of EEG signals in the related art still relies on the empirical determination of the degree of wavelet decomposition of the signal, and the signal cannot be decomposed adaptively. [0004] 2. The indicators for judging EEG signals are too single, and there is no objective and u...

AI Technical Summary

Problems solved by technology

[0003] 1. The abnormal monitoring of EEG signals in related technologies still depends on experience to determine the degree of wavelet decomposition of the signal, and cannot decompose the signal adaptively

[0004] 2. The indicators for judging EEG signals are too single, and there is no objective and unified monitoring standard, which is prone to misjudgment and missed judgment of abnormal EEG signals

[0005] 3. When monitoring EEG signals, the dynamic characteristics of EEG signals are not considered

Images