Brain-computer interface method with additional Chinese character silent reading and exercise imagery tasks

Abstract

The invention relates to a brain-computer interface method with additional Chinese character silent reading and exercise imagery tasks, which comprises the following steps of: acquiring electroencephalogram signals of a subject in a plurality of groups of additional Chinese character silent reading and exercise imagery tasks, taking a plurality of Chinese characters as stimulation in a single task, comprising an idle stage, a preparation stage, a prompt stage and a silent reading imagery stage; preprocessing the electroencephalogram signals, obtaining the energy of the electroencephalogram signals of the subject, and determining the optimal filtering frequency range according to the energy of the electroencephalogram signals; preprocessing the electroencephalogram signals again, and performing filtering processing on the electroencephalogram signals according to the optimal filtering frequency range in the re-preprocessing process; using the preprocessed electroencephalogram signals as electroencephalogram data samples, performing feature extraction and classification on the electroencephalogram data samples through a convolutional neural network, and using classification results for driving external equipment interacting with the brain. According to the method, an exercise imagery task normal form with additional Chinese character slient reading is provided, and the classification accuracy of the electroencephalogram signals is improved.

Description

technical field [0001] The invention belongs to the technical field of brain-computer interface of biomedical engineering, and in particular relates to a brain-computer interface method for silent reading of Chinese characters and motor imagination tasks. Background technique [0002] Brain Computer Interface (BCI) is an information transmission technology that realizes the direct interaction between the brain and external devices by decoding the information of human brain nerve activity. The realization of BCI technology is to convert the electrophysiological activity characteristics of the cerebral cortex into corresponding Instructions, and then realize the control of the brain to external devices. BCI technology has been used in many fields such as military, medical, entertainment, etc. For example, in the military field, electroencephalogram (EEG) can be used to assist in the control of weapons and equipment. In the medical field, BCI technology can be used to help phys...

AI Technical Summary

Problems solved by technology

At present, the BCI technology based on MI-EEG has the defect of less operational level, and the extracted EEG features are limited. Therefore, with the increase of imagination tasks, the classification accuracy will drop sharply; Accurate manipulation of the external equipment of the BCI system, resulting in the inability of the user to effectively and spontaneously drive the external equipment

[0004] The BCI system generally includes five parts: EEG acquisition, signal preprocessing, feature extraction, classification and discrimination, and control interface. The core of the BCI system is to improve the classification accuracy of EEG. Unstable and other characteristics, the frequency range of the EEG signal characteristics of each user is different, so the same frequency range is used for filtering the EEG signals of all users, and it cannot be guaranteed that the EEG of each user can reach the optimal level. good classification effect

Images