(4) compared with electroencephalography under EC conditions, the frequency band with power spectrum energy reduced by more than 20% under EO conditions is considered as personalized Alpha band partitioning (IAF);

(5) according to the IAF division, taking the range from 3Hz to the lower boundary of the IAF as a Theta waveband, and taking the range from the upper boundary of the IAF to 18Hz as the Beta waveband;

(6) calculating the low-frequency/high-frequency energy ratio of the ith segment, namely Theta/Beta ratio (TBR), according to the segment data of the resting electroencephalogram under the EO condition:

(7) collecting forehead electromyogram signals, and recording the forehead brow muscle-forehead muscle position. Cutting the signal into sections with the length of 1s without superposition; calculating the mean square energy of the electromyographic signal of the ith segment:

(8) marking the state of the electromyographic signals, distinguishing the steady state from the muscle action state:

E_thr＝μ+Tδ

wherein mu and delta are the average value and standard deviation of mean square energy of all segmented myoelectricity, and T is used for adjusting tolerance degree. Here, we take T-3.

(9) The mean square energy of the resting electroencephalogram TBR and the electromyogram signal in a stable state is used as a training baseline.

(10) In the training process, the electroencephalogram of the occipital area and the myoelectricity of the forehead are collected in real time, the TBR based on the IAF is calculated in real time, and the parameters are as follows: the window length is 1s, the superposition is 50%, 1024-point fast Fourier transform is performed, and a Hamming window is adopted to reduce frequency spectrum leakage; the mean square energy of the forehead electromyogram signal is real-time, the window length is 0.5s, no coincidence exists, 512-point fast Fourier transform is performed, and a Hamming window is adopted to reduce frequency spectrum leakage.

(11) The feedback control strategy is: awarding a reward when both conditions one and two are met, otherwise awarding a penalty: compared with a resting baseline, the TBR is reduced by 20 percent; the mean square energy of the real-time myoelectricity does not exceed the sum of the average value of the forehead myoelectricity in a stable state and 3 times of standard deviation.

In the neural feedback training process, the acquisition equipment respectively acquires the electroencephalogram signals and the forehead electromyogram signals, and respectively analyzes the signals to extract corresponding characteristics.

The technical core is to fuse the forehead myoelectric index obtained by calculation with the individual rhythm ratio index to obtain an accurate state index of the cranial nerve system, and then control a nerve feedback system to ensure that audio and video feedback is consistent with the current requirement of cranial nerve feedback training. Finally, a real-time forehead myoelectric and cranial nerve signal acquisition-real-time signal processing-feature fusion-feedback control nerve feedback training technology is formed.

The technology adopts the individualized rhythm ratio as one of the training indexes, overcomes the problem of low effective rate caused by inaccurate rhythm division of the cranial nerve system caused by adopting fixed frequency band division, and simultaneously avoids the safety problem caused by subsequent inaccurate training; the technology integrates the individualized rhythm ratio and the forehead myoelectric signal, integrates the nerve psychology emotion characterization advantage of the forehead myoelectric, and improves the effectiveness of nerve feedback training.

The invention has the advantages that:

compared with the traditional division mode adopting fixed brain rhythms, the method has the advantages that the individualized rhythms can reflect the real state of the brain nervous system, so that the effectiveness and the safety of the neural feedback training are ensured; another great advantage is that the neural rhythm information and the forehead myoelectric information are fused, and the brain nervous system development disorder is described from more dimensions, so that the effectiveness of the neural feedback training is improved.

Detailed Description

FIG. 1 is a flow chart of the neural feedback method fusing individualized brain rhythm ratio and forehead myoelectric energy according to the present invention.

Detailed Description

According to the invention, the forehead myoelectric signals are fused for nerve feedback, so that on one hand, the physiological value of the forehead myoelectric signals is fully utilized; on the other hand, the effect of myoelectricity on brain signals can also be reduced. Therefore, the fused neurofeedback training system has higher safety and effectiveness.

The nerve feedback method fusing the individualized brain rhythm ratio and the forehead myoelectric energy comprises the following specific working procedures:

(1) collecting resting brain electricity before training, recording brain electricity of electrodes P1, P2, Pz, O1 and O2 by taking binaural connection as reference; open eye conditions (EO) and closed Eye Conditions (EC) were each three minutes. The recorded data is divided into length signals of 4s, and the division sections are overlapped by 50 percent;

(8) a threshold for marking the state of the electromyographic signals and distinguishing the stable state from the muscle action state:

E_thr＝μ+Tδ

wherein mu and delta are the average value and standard deviation of mean square energy of all segmented myoelectricity, and T is used for adjusting tolerance degree. Here, we take T ═ 3;

(9) using the mean square energy of the resting electroencephalogram TBR and the electromyogram signal in a stable state as a training baseline;

(10) in the training process, the electroencephalogram of the occipital area and the myoelectricity of the forehead are collected in real time, the TBR based on the IAF is calculated in real time, and the parameters are as follows: the window length is 1s, the superposition is 50%, 1024-point fast Fourier transform is performed, and a Hamming window is adopted to reduce frequency spectrum leakage; real-time forehead electromyogram signal mean square energy, wherein the parameter is the window length of 0.5s, no coincidence exists, 512-point fast Fourier transform is performed, and a Hamming window is adopted to reduce frequency spectrum leakage;

In the neural feedback training process, the acquisition equipment respectively acquires the electroencephalogram signals and the forehead electromyogram signals, and respectively analyzes the signals to extract corresponding characteristics. The technical core is to fuse the forehead myoelectric index obtained by calculation with an individual rhythm ratio index to obtain an accurate state index of a cranial nerve system, then control a nerve feedback system to ensure that audio and video feedback is consistent with the current requirement of cranial nerve feedback training, and finally form a real-time forehead myoelectric and cranial nerve signal acquisition-real-time signal processing-feature fusion-feedback control nerve feedback training technology.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.

Claims

1. A neurofeedback system for children fusing individualized brain rhythm ratio and forehead myoelectric energy, characterized in that it comprises the following modules:

the system comprises a forehead myoelectricity and cranial nerve signal acquisition module, a real-time signal processing module, a characteristic fusion module and a feedback control module;

the acquisition module is used for respectively acquiring the electroencephalogram signal and the forehead electromyogram signal and respectively analyzing the signals to extract corresponding characteristics;

fusing the forehead myoelectric index obtained by calculation with the individual rhythm ratio index to obtain an accurate cerebral nervous system state index, and then controlling a nerve feedback system to ensure that audio/video feedback is consistent with the current cerebral nerve feedback training requirement;

the working method of the system is as follows:

(1) collecting resting brain electricity before training, recording brain electricity of electrodes P1, P2, Pz, O1 and O2 by taking binaural connection as reference, wherein the sampling rate is 1 KHz; the recorded data are divided into signal sections with the duration of 4s, and the sections are overlapped by 50 percent;

(3) respectively calculating resting state electroencephalogram sequences X of each channel under two conditions EO and EC to respectively obtain power spectral density curves of each segment:

where N is the number of points in each segment of data, x_nData points in the brain electrical sequence X;

(4) averaging PSDs of all channels on each frequency point, calculating electroencephalogram under EC condition, regarding frequency bands with power spectrum energy reduced by more than 20% under EO condition as personalized Alpha frequency band division (IAF), marking lower boundaries of IAF as lower, and marking upper boundaries of IAF as upper;

(7) collecting forehead electromyogram signals, wherein recording positions are forehead frown muscle-forehead muscle positions, and the signals are divided into sections with the length of 1s without superposition; calculating the mean square energy of the electromyographic signal Y of the ith segment:

wherein N is the length of the myoelectric data segment Y, Y_iData points for Y;

(8) marking the state of the electromyographic signals, and distinguishing a threshold value of a steady state and a muscle action state:

E_thr＝μ+Tδ

wherein mu and delta are the average value and standard deviation of mean square energy of all segmented myoelectricity, T is used for adjusting tolerance degree, and T is taken as 3;

(9) averaging the plurality of data segments TBR under the EO condition in the resting state as the baseline of condition 1 in step (11); for forehead electromyogram signal a plurality of data segments E_iCalculating to obtain the average value and standard deviation of the mean-square energy of a plurality of data segments, and using the average value and standard deviation to calculate the base line of the condition 2 in the step (11);

(10) in the training process, collecting electroencephalogram of a top occipital area and forehead myoelectricity in real time, and calculating TBR based on IAF in real time;

(11) real-time feedback of electroencephalogram and forehead myoelectricity controls the effect of audio and video;

wherein, the TBR based on IAF is calculated in real time, and the parameters are as follows: signal duration is 1s, 1024-point fast Fourier transform is carried out, and a Hamming window is adopted to reduce frequency spectrum leakage; the mean square energy of the forehead electromyogram signal is real-time, the parameter is signal duration 1s, and no coincidence exists;

the individualized electroencephalogram rhythm ratio and forehead myoelectricity energy fusion feedback control strategy is as follows: awarding a reward when both conditions one and two are met, otherwise awarding a penalty: compared with the EO baseline in the resting state, the TBR is reduced by 20 percent; the mean square energy of the real-time myoelectricity does not exceed the sum of the mean square energy of the forehead myoelectricity in a stable state and 3 times of standard deviation.