Electroencephalogram signal recognition method and system combining recurrence plot and CNN

Abstract

The invention belongs to the technical field of medical treatment. The invention discloses an electroencephalogram signal recognition method and system combining a recurrence plot and a CNN, and the method comprises the steps: decomposing a motor imagery electroencephalogram signal into intrinsic mode functions of different scales through the empirical mode decomposition of preprocessed electroencephalogram signal data, calculating a multi-scale recurrence plot of an intrinsic mode component of each scale, and obtaining a first-stage feature; regarding the reconstructed multi-scale recurrenceplot as image features of left and right hand EEG signals, regarding the multi-scale recurrence plot features as input of a convolutional neural network, classifying and recognizing the recurrence plot through the convolutional neural network, and extracting second-level features capable of better expressing motor imagery electroencephalogram signals from the first-level features. The electroencephalogram signal recognition rate is high, and electroencephalogram signals can be better recognized; the delay time result determined by the mutual information method is more accurate. The ReLU activation function is adopted, and when the input is a positive number, the problem of gradient saturation does not exist.

Description

technical field [0001] The invention belongs to the field of medical technology, and in particular relates to an EEG signal recognition method and system combining a recursive graph and a CNN. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] The Brain-Computer Interface (BCI) system realizes the establishment of a direct interaction channel between the human brain and the external environment without the support of the peripheral nervous system by studying the operation mode of the electric current signal sent by the human brain in the cerebral cortex. The key to realizing this technology lies in how to effectively identify and classify various EEG signals generated by the brain using reasonable signal processing algorithms. [0004] In recent years, the development trend of artificial intelligence has also been increasing. For example, a series of major events such as IBM Watson cognitive computing a...

AI Technical Summary

Problems solved by technology

[0011] In the current research on brain-computer interface systems, in the signal acquisition stage, the collected EEG signals contain a lot of noise, which cannot effectively preserve the real signal. It is necessary to find better preprocessing techniques and better denoising algorithms to Improve the signal-to-noise ratio of EEG to ensure the reliability of the BCI system

[0012] (2) The system recognition rate needs to be improved

[0014] (3) Hardware technology needs to be improved

[0016] (4) System adaptive problem

[0021] (1) There are two ways to collect EEG signals, implanted and non-implanted. The implanted type requires electrodes to be implanted in the cerebral cortex. This method will cause harm to the subject, so now the non-implanted Implantable EEG signal acquisition, non-implantable acquisition method places electrodes on the subject's scalp for signal recording, this method will not cause harm to the subject but the collected signal is susceptible to noise , thus affecting the accuracy of the entire BCI system

Images