Screening method for motor imagery response ability based on resting-state electroencephalography features

Abstract

The invention discloses a screening method for motor imagery (MI) response ability based on resting-state electroencephalography (EEG) features. The method includes the following steps: preprocessingacquired EEG data, performing data segmentation, and extracting common spatial pattern features of MI tasks; establishing a multiple classification recognition model by using a support vector machine,calculating a classification correct rate through 10-fold cross-validation, and characterizing the MI response ability by using the classification correct rate; extracting normalized energy, power spectrum entropy and Lempel-Ziv complexity of each eye-open resting lead EEG in a 8-13 Hz frequency band (alpha frequency band) respectively; calculating the correlation between the above resting-stateEEG features and the MI response ability, and screening optimal features to establish a classification model and a regression prediction model; and based on the classification model and the regressionprediction model, screening the motor imagery response ability. The method can screen out ''BCI blindness'', avoid an invalid training process, and predict the MI response ability of a subject in advance and formulate a matching training program for the subject, thereby optimizing an experimental training process and ultimately reducing an error rate of a user in operating brain-computer interface (BCI).

Description

technical field [0001] The invention relates to the field of motor imagery brain-computer interfaces, in particular to a method for screening motor imagery responsiveness based on resting state EEG characteristics. Background technique [0002] Brain-Computer Interface (BCI) is a communication control system that does not depend on the normal output channels of peripheral nerves and muscles in the brain. Motor Imagery based BCI (MI-BCI) is a typical active BCI. Subjects induce different patterns of EEG in the brain by imagining the movement of a certain part of the body (such as imagining the movements of the left and right hands, legs or tongue). (Electroencephalography, EEG) signal, and then detect and identify the pattern characteristics of the generated EEG signal, and convert human movement thinking into corresponding pattern output instructions to control designated external equipment to perform predetermined tasks. Because MI-BCI has a wide range of application prosp...

AI Technical Summary

Problems solved by technology

Studies have shown that about 20% of the subjects could not successfully induce typical EEG features, so they could not use BCI effectively

Images