Electroencephalogram signal identification method based on brain network and deep learning

Abstract

An electroencephalogram signal recognition method based on a brain network and deep learning comprises the steps that motor imagery electroencephalogram data and language imagery electroencephalogram data are obtained and preprocessed, and corresponding labels are obtained; based on the multi-lead electroencephalogram data, respectively calculating the phase synchronism of the multi-narrow-band inter-lead time sequence; setting a threshold value to obtain a plurality of narrowband connection matrixes based on phase synchronization; constructing a functional brain network through the connection matrix; a deep learning model is trained, the multi-narrow-band synchronous brain network serves as input of the convolutional neural network at the same time, and the structure and parameters of the network are optimized; according to the method, training and classification are carried out through a synchronous brain network classification model, the strong feature extraction capability and the time sequence signal processing capability of a deep learning algorithm are fully utilized, and the time sequence information hidden in the electroencephalogram signals is combined, so that the electroencephalogram signal recognition task of the multi-narrow-band brain network is completed; the sizes of a multi-input convolution layer and a convolution kernel are reasonably designed, and the classification effect is improved.

Description

technical field [0001] The present invention relates to the field of signal processing, in particular to an EEG signal recognition method based on brain network and deep learning. Background technique [0002] The human brain is a large-scale network. The biological operation system of the human brain requires the participation and interaction of multiple brain regions. Brain-computer interface technology can directly identify human intentions by detecting brain neural activity and convert them into Computer control instructions, so as to realize the human brain's operation control of external devices, is a new type of human-computer interaction control technology. However, the traditional method does not take into account the topological relationship between electrodes, and the interaction between signals on the scalp may not be fully exploited. , the research method of analyzing specific lead signals alone cannot reflect the information flow or causal relationship between ...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present application provides a method for identifying EEG signals based on brain networks and deep learning, aiming to solve the problem that the prior art cannot accurately identify EEG signals

Images