Brain-controlled rehabilitation system motion imagery recognition system integrating complex network and graph convolution

Abstract

The invention discloses a brain-controlled rehabilitation system motion imagery recognition system integrating a complex network and graph convolution. According to the system, a testee carries out motion imagery through watching a hand fist clenching and stretching motion video, and the motor imagery EEG signals of the testee is collected through an EEG signal collection device; a motion intention recognition module is used for constructing a multi-entropy complex network for the obtained motion imagery EEG signals; the characteristics in the aspects of symbol fluctuation, frequency energy distribution and amplitude fluctuation in the motion imagery EEG signals are extracted; and the characteristics are input into a graph convolutional neural network to classify and identify fist clenching motion imagery EEG signals and hand stretching motion imagery EEG signals, and a classification result is transmitted to a brain-controlled rehabilitation system to prompt the testee to execute handfist clenching and stretching actions. According to the brain-controlled rehabilitation system, a closed-loop path is formed between motion intention and limb feeling, the muscle strength and the nerve conduction speed of a testee are gradually enhanced, recovery of a damaged brain movement area is promoted, and the movement capacity is gradually recovered.

Description

technical field [0001] The invention relates to a motor imagery recognition system. In particular, it involves a brain-controlled rehabilitation system motor imagery recognition system that integrates complex networks and graph convolution. Background technique [0002] The Brain-Computer Interface (BCI) system provides a connection between the human brain and external devices. The system first collects brain activity signals, then detects the user's intention through the signal processing part, and finally converts the intention into instructions to control the external device. Motor imagery is a classic BCI paradigm. When a person imagines a limb movement, it will cause the activation of a certain area of ​​the brain's motor perception cortex. The activation of the motor perception cortex will trigger a change in cortical potential. This electrical signal is called a motor imagination signal. Different motor imagery tasks observed oscillatory activity in different regions...

AI Technical Summary

Problems solved by technology

However, due to the characteristics of weak and noisy EEG signals, it is still impossible to accurately classify motor imagery signals. The accurate extraction of motor imagery signal features needs further exploration.

Images