A brain-computer interface system based on selective attention to vibratory stimuli

Abstract

The application discloses a brain-computer interface system based on vibration stimulation selective attention, and the system comprises a vibration tactile stimulation device, an electroencephalogram acquisition system and a computer for stimulation interface presentation and data processing; a subject sits on a chair about 70 cm in front of the stimulation interface, and vibration stimulation is applied to the index fingers of the left and right hands according to an experimental paradigm process, and the stimulation frequencies are respectively set as optimal specificity stimulation frequencies screened in advance; the subject selectively concentrates attention on a target side according to a prompt of the stimulation interface; the electroencephalogram acquisition system acquires electroencephalogram data of the subject, the acquired data are subjected to subsequent pretreatment, feature extraction and classification to identify the target stimulation concerned by the subject, and the identification result is fed back to the subject in the form of vibration stimulation. The application can decode the user intention by using the electroencephalogram signal feature difference induced by different vibration stimulation selective attention tasks, and then convert the user intention into a target instruction, so that the user can interact with the external environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of brain-computer interface (BCI), and particularly relates to a brain-computer interface system based on selective attention of vibration stimulation. BACKGROUND

[0002] The BCI system does not rely on the human body's own neuromuscular pathway, and can directly convert central nervous system activity into artificial output. It acquires user intention by collecting and decoding the electroencephalogram (EEG) generated by the human brain when performing certain thinking activities or being stimulated by external stimuli, so as to realize the interaction between the central nervous system and the external environment. At present, the BCI system has been applied in military, transportation, medical and entertainment fields, and has a broad future.

[0003] The BCI has multiple control strategies, and several commonly used paradigms have advantages and disadvantages. For example: the BCI based on motor imagery requires users to have good motor imagery strategy and ability, is affected by the user's subjective consciousness, has large individual differences, and has poor stability; the BCI based on auditory evoked potential requires the user's auditory pathway to be normal, is not suitable for users with impaired or lost hearing, and has high requirements for the use environment, so that the applicable scenarios and actual communication speed are limited; the BCI based on visual evoked potential depends on external visual stimulation, is easy to cause visual fatigue after long-term use, is not suitable for patients with visual impairment, especially for patients with Locked-In Syndrome (LIS) or Amyotrophic Lateral Sclerosis (ALS) in the late stage who have damage or loss of eye movement control, and they are the potential users who need the BCI system assistance. As a potential supplement or alternative, the BCI based on somatosensory evoked potential depends on the somatosensory system, uses the tactile channel with wide distribution and strong sensitivity in the human body as the information transmission pathway, does not occupy the user's visual and auditory channels, and is not easy to cause visual fatigue. It is suitable for patients with visual and auditory diseases, especially for LIS and ALS patients whose somatosensory system is still intact, and it has important practical significance to carry out related research.

[0004] According to the difference of somatosensory stimulation frequency, somatosensory evoked potential can be divided into transient somatosensory evoked potential (TSEP) (1-10 Hz) and steady-state somatosensory evoked potential (SSSEP) (more than 10 Hz). Compared with TSEP, SSSEP has advantages in stability and speed. Applying somatosensory stimulation of a specific frequency on different limbs can induce SSSEP signals of the same frequency at the corresponding position of the contralateral somatosensory cortex. The signal characteristics are stable, the spatial separability is good, and the signal amplitude can be modulated by attention. In related BCI systems, the method of electric stimulation is generally used to induce SSSEP. However, the electric stimulation has high requirements for the performance parameters of the device, and the induced signal is easily disturbed by electromagnetic signals. Strong electric stimulation may also cause pain and spasm in users, and has certain traumatic and dangerous characteristics, low comfort, and poor user experience. Studies have shown that the amplitude and signal-to-noise ratio of SSSEP are highly dependent on the stimulation frequency, and have large individual differences. In related applications, compared with using a unified stimulation frequency, using the individual optimal specificity stimulation frequency that can induce the highest amplitude SSSEP can obtain better BCI performance. However, the related research generally uniformly specifies the electric stimulation frequency for inducing SSSEP, and does not consider the influence of individual differences. SUMMARY

[0005] The application provides a brain-computer interface system based on selective attention of vibration stimulation, which can induce SSSEP characteristics and event-related desynchronization (ERD) characteristics in the cerebral cortex. By using the differences in electroencephalogram signal characteristics induced by different vibration stimulation selective attention tasks, the user's intention can be decoded, and then converted into target instructions to realize the interaction between the user and the external environment. Details are described below:

[0006] A brain-computer interface system based on selective attention of vibration stimulation, the system comprises: a vibration haptic stimulation device, an electroencephalogram acquisition system, and a computer for stimulation interface presentation and data processing;

[0007] The subject sits on a chair about 70 cm in front of the stimulation interface, and the left and right index fingers are subjected to vibration stimulation according to the experimental paradigm process, and the stimulation frequency is set to the optimal specificity stimulation frequency selected in advance; the subject selectively concentrates attention on the target side according to the prompt of the stimulation interface;

[0008] The electroencephalogram acquisition system acquires electroencephalogram data of a subject, and the acquired data is preprocessed, feature extracted and classified to identify a target stimulus of interest of the subject, and the identification result is fed back to the subject through vibration stimulation.

[0009] In the experiment, the target stimulus position is the left or right index finger, the frequency screening process of the bilateral index fingers is performed respectively, the vibration stimulation duration in a single trial is 4s, the inter-stimulation interval is 2s, and the stimulation frequency is a random frequency in the range of 17 to 35Hz with a step of 2Hz.

[0010] First, a 4s black screen rest is maintained, then a 3.5s circle is presented in the center of the screen to prompt the subject to concentrate attention, and after the circle disappears, plus-minus numbers appear in the center of the screen, the subject is required to concentrate attention on the plus-minus calculation of the numbers, and answer the calculation result during the black screen rest after the numbers disappear, and continue the experiment after the circle prompt appears again.

[0011] In the experiment, every 10 trials is a group, covering 10 random stimulation frequencies from 17 to 35Hz, and every 5 groups is a block; the left and right index fingers are respectively subjected to 6 blocks of experiments to improve the test accuracy, and each index finger includes 30 trials of data under the same stimulation frequency; the subjects are rested and adjusted according to the fatigue condition between blocks.

[0012] In the experiment, every 10 trials is a group, covering 10 random stimulation frequencies from 17 to 35Hz, and every 5 groups is a block; the left and right index fingers are respectively subjected to 6 blocks of experiments to improve the test accuracy, and each index finger includes 30 trials of data under the same stimulation frequency; the subjects are rested and adjusted according to the fatigue condition between blocks.

[0013] Further, the difference between the two frequencies needs to be at least 4Hz, and the best specific stimulation frequency of the left and right index fingers of each subject is determined, and the best specific stimulation frequency is applied to the vibration stimulation selective attention experiment.

[0014] In the experiment, the target stimulus position is the left and right index finger, and the stimulation frequency of the two channels of the vibration tactile stimulation device is set to the best specific stimulation frequency screened.

[0015] Further, at the beginning of each trial, a cross in the center of the screen is presented for 1s to prompt the subject to be ready; after the cross disappears, the vibration tactile stimulation device applies vibration stimulation to the index fingers of the left and right hands at the same time, a circle appears on the screen and lasts for 1s to prompt the subject to be ready to perform the selective attention task; after the circle disappears, a selective attention direction prompt appears on the screen, which is an arrow pointing to the left or an arrow pointing to the right, and the direction prompt disappears after 1s to present a black screen; the subject immediately performs the corresponding task according to the prompt after seeing the direction prompt; when the direction prompt is an arrow pointing to the left, the subject needs to focus attention on the vibration stimulation of the left index finger while ignoring the vibration of the right index finger; when the direction prompt is an arrow pointing to the right, the subject needs to focus attention on the vibration stimulation of the right index finger while ignoring the left index finger.

[0016] In the selective attention task, the subject needs to roughly count the number of vibrations of the motor; the vibration stimulation lasts for 6s in the whole trial, and the subject focuses on the vibration stimulation of the target side according to the prompt direction after seeing the selective attention task direction prompt until the vibration stimulation ends; in the training stage, the interval between trials is 2-3s, and the subject remains at rest during this period and adjusts the state to wait for the next trial. The brain electrical data of 10 blocks is collected in the training stage.

[0017] The beneficial effects of the technical scheme provided by the application are:

[0018] 1. The vibration stimulation method with strong operability, easy control, non-invasiveness and high safety factor is used in the system to induce SSSEP, so as to improve the comfort of the user; considering the individual differences among users, the individual optimal specific stimulation frequency of each user is determined through vibration stimulation frequency screening, and the screening result is applied to the vibration stimulation selective attention experiment in the subsequent design, so as to improve the performance of the system.

[0019] 2. Compared with the traditional BCI system, the brain-computer interface system based on vibration stimulation selective attention adopted by the application does not need to occupy the visual and auditory channels, and is also suitable for patients with visual and auditory disorders but with intact somatosensory system, and can be used as a potential supplement and alternative scheme of the brain-computer interface system, which has certain research significance and application value.

[0020] 3. Compared with the method of inducing SSSEP by using electric stimulation, the method of inducing SSSEP characteristics by using vibration stimulation avoids the problems that electric stimulation is easily interfered by electromagnetic signals and that strong electric stimulation easily causes pain and spasm of the subject, and has high safety, strong operability, easy control and improved comfort of the subject.

[0021] 4、Compared with the related research using a unified stimulation frequency to induce SSSEP, the present application considers the influence of individual differences, selects the individual optimal specific stimulation frequency for each subject, and obtains better system performance;

[0022] 5、Through experimental verification, the brain-computer interface system designed by the present application can stably induce obvious SSSEP characteristics, effectively identify the user's thinking intention, and to some extent, improve the system classification performance; the vibration feedback designed by the present application can timely report the identification result to the subject, which helps to increase the subject's participation. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 It is a structural schematic diagram of a brain-computer interface system based on vibration stimulation selective attention;

[0024] Figure 2 It is a physical diagram of a vibration tactile stimulation device;

[0025] Among them, (a) is a control circuit board; (b) is a stimulation end fixed mode; (c) is an external vibration motor and a thin film pressure sensor; (d) is an OLED display of pressure detection data.

[0026] Figure 3 It is a design schematic diagram of a vibration stimulation frequency screening paradigm;

[0027] Figure 4 It is a vibration stimulation frequency screening spectrum diagram of a typical subject's left hand index finger;

[0028] Figure 5 It is a design schematic diagram of a vibration stimulation selective attention experiment paradigm. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical scheme and advantages of the present application clearer, the embodiments of the present application are further described in detail below.

[0030] Figure 1 It is a structural schematic diagram of a brain-computer interface system based on vibration stimulation selective attention designed by the present application. The system mainly includes: a vibration tactile stimulation device, an electroencephalogram acquisition system, and a computer for stimulation interface presentation and data processing. In application, the subject sits on a chair about 70 cm in front of the stimulation interface, and the left and right hand index fingers are subjected to vibration stimulation according to the experimental paradigm process, and the stimulation frequency is set to the optimal specific stimulation frequency selected in advance. The subject needs to selectively focus attention on the target side according to the stimulation interface prompt. The electroencephalogram acquisition system acquires the electroencephalogram data of the subject, and the acquired data is subjected to subsequent preprocessing, feature extraction and classification to identify the target stimulation concerned by the subject, and the identification result is fed back to the subject in the form of vibration stimulation.

[0031] I. Vibratory tactile stimulation device

[0032] The vibratory tactile stimulation device applied in the embodiment of the present application comprises two stimulation channels and mainly has two working modes of pressure display and vibration stimulation. In use, the stimulation device is connected to a power supply lithium battery, one end of which is connected to a computer presenting a stimulation interface through a USB line, and the other end is connected to a thin film pressure sensor placed below the vibration motor. The two are fixed on the index finger of a subject through an elastic band with soft texture, good air permeability and high elasticity, in a manner as shown in Figure 2 (b). The host computer uses Matlab to write a control program to output control instructions. First, the stimulation device is adjusted to the pressure display mode, and the detection data is displayed on the OLED display screen, as shown in Figure 2 (d). The OLED display screen displays the data measured by the pressure sensor, with the first and third lines being channel markers, and the second and fourth lines being measured voltage values (mv) and pressure values (g). The tightness of the elastic band is adjusted to keep the fingertip pressure within a consistent range, preventing the fixation pressure from being too loose or too tight to affect the induction effect. After the fixation is completed, the stimulation device is adjusted to the vibration mode, and the two-channel vibration motor will generate vibration stimulation of the specified frequency to the target limb according to the experimental procedure.

[0033] II. Vibration stimulation frequency screening

[0034] The vibration stimulation frequency screening results are relatively stable, and considering the overall experimental time, the vibration stimulation frequency screening experiment can be performed in advance. In the experiment, the target stimulation position is the index finger of the left or right hand, and the frequency screening process of the bilateral index fingers is performed respectively, and the experimental paradigm is as shown in Figure 3 . The vibration stimulation duration in a single trial is 4s, the inter-stimulus interval is 2s, and the stimulation frequency is a random frequency in the range of 17 to 35Hz with a step of 2Hz. To avoid attention modulation caused by the subject's attention to the vibration stimulation, the subject needs to perform a mental calculation task while receiving the vibration stimulation. In the mental calculation task, first, a 4s black screen rest is maintained, then a 3.5s circle appears in the center of the screen, prompting the subject to concentrate attention, and after the circle disappears, an addition and subtraction number appears in the center of the screen, and the subject is required to concentrate attention on the calculation of the number, and answer the result of the calculation during the black screen rest after the number disappears. After the circle prompt appears again, the experiment continues. In the experiment, every 10 trials is a group, covering 10 random stimulation frequencies from 17 to 35Hz, and every 5 groups is a block. The left and right index fingers respectively undergo 6 blocks of experiments to improve the testing accuracy, and each index finger contains 30 trials of data under the same stimulation frequency. A block of experiment takes about 6 minutes, and the subjects can rest and adjust according to their fatigue during the block.

[0035] The EEG data collected in the experiment was pre-processed, and the average FFT results of 30 trials under the same stimulation frequency were calculated. The EEG response spectrograms of the contralateral prefrontal and midparietal F, FC, C, CP leads where the activation phenomenon was more obvious were analyzed. The stimulation frequency that could induce the highest SSSEP amplitude under the selected 10 frequencies was selected as the best specific stimulation frequency of the subject's target limb. Figure 4 The spectrograms of the EEG responses induced at the F2, F4, F6, FC2, FC4, FC6, C2, C4, C6, CP2, CP4, CP6 leads in the contralateral brain region of the typical subject during the vibration stimulation frequency screening of the left index finger are shown. Figure 4 The horizontal axis is the frequency (Hz), and the vertical axis is the amplitude (μV). Each curve represents the SSSEP response induced by the stimulation frequency with a step of 2 Hz in the range of 17-35 Hz. As can be seen from the figure, vibration stimulation at all frequencies successfully induced SSSEP at the corresponding frequency, showing a significant increase in amplitude at the stimulation frequency on the amplitude-frequency curve. The frequency error of the induced SSSEP response is within 1 Hz. By comparing the induced amplitudes of SSSEP at the above leads, it can be seen that among all the frequency combinations of the leads, the SSSEP amplitude at the F2 lead at 25 Hz is the highest. Therefore, 25 Hz is finally selected as the best specific stimulation frequency for the left index finger of the subject. Similarly, the right index finger of the subject can be subjected to vibration stimulation frequency screening. To ensure that the stimulation frequency has a certain difference, the minimum difference between the two frequencies should be 4 Hz. The above screening method is used to determine the best specific stimulation frequency of the left and right index fingers of each subject, which is applied to the vibration stimulation selective attention experiment.

[0036] III. Vibration stimulation selective attention experiment

[0037] In the experiment, the target stimulation position was the left and right index fingers, and the stimulation frequencies of the two channels of the vibration tactile stimulation device were set to the best specific stimulation frequencies selected. The experiment was divided into two stages of training and testing, and two tasks of selective attention left hand and selective attention right hand were designed, Figure 5The flow design of a single trial in the vibration stimulation selective attention experiment is shown. At the beginning of each trial, a cross is presented in the center of the screen for 1s, prompting the subject to be ready. After the cross disappears, the vibration tactile stimulation device applies vibration stimulation to the index fingers of the left and right hands at the same time, and at the same time, a circle appears on the screen for 1s, prompting the subject to prepare to perform the selective attention task. After the circle disappears, a selective attention direction prompt appears on the screen, represented as an arrow pointing to the left or an arrow pointing to the right, and after 1s the direction prompt disappears, presenting a black screen. The subject needs to immediately perform the corresponding task according to the prompt after seeing the direction prompt. When the direction prompt is an arrow pointing to the left, the subject needs to focus on the vibration stimulation of the left index finger, while ignoring the vibration of the right index finger; when the direction prompt is an arrow pointing to the right, the subject needs to focus on the vibration stimulation of the right index finger, ignoring the left index finger. To improve the subject's attention to the target stimulation finger, the subject needs to count the number of motor vibrations roughly in the selective attention task. The vibration stimulation lasts for 6s in the whole trial, and the subject needs to focus on the target side vibration stimulation according to the prompt direction after seeing the selective attention task direction prompt until the vibration stimulation ends. In the training phase, the interval between trials is 2-3s, and the subject needs to remain still and adjust the state during this period to wait for the next trial. The training phase collects 10 blocks of EEG data, each block contains 20 trials, and two kinds of selective attention tasks appear randomly, each with 10 trials. Each type of selective attention task collects 100 trials of data. Each block of experiment takes about 3 minutes, and there is a 5-minute rest between every two blocks to facilitate the subject to adjust the state and avoid fatigue. After the training phase, the training data is preprocessed and feature extracted to build a classification model for the left and right fingers, which is used in the test phase. The experimental paradigm in the test phase is basically the same as in the training phase, except that vibration feedback is added after the selective attention task. In the test phase, after completing a trial, the collected data is preprocessed and feature extracted, and the target of the subject's attention is identified according to the classification model built in the training phase, and the classification result is fed back to the subject within 1.5s, in the form of vibration stimulation on the corresponding finger, and the subject can adjust his state according to the feedback information.

[0038] A total of 25 subjects participated in the experiment, of which 14 were female, aged 19-26, all right-handed healthy students. A 10x10 cross-validation strategy was used to extract features and classify the 100 trials of EEG data in each selective attention task in the training phase. As shown in Table 1, the average two-classification accuracy between selective attention left hand and selective attention right hand tasks is 76.29%, which is better than the average classification level of 60%-72% in the related literature of SSSEP-BCI system.

[0039] Table 1 Classification results of training stage (%)

[0040]

[0041] IV. Data acquisition and processing

[0042] The electroencephalogram acquisition part in the embodiment of the application adopts a 64-channel electroencephalogram acquisition system of SynAmps2 produced by Neuroscan Company. The system mainly comprises a SynAmps2 electro-physiological electroencephalogram amplifier and a Scan4.5 electroencephalogram acquisition software. A 64-electrode cap is used in the experiment, the electrodes used are silver / silver chloride alloy electrodes, and the electrode position distribution adopts the standard positioning of the international 10-20 system. The forehead and the tip of the nose are respectively used as the ground and reference electrodes. The sampling frequency is set to 1000 Hz, and the 50 Hz power frequency interference is filtered out.

[0043] In the vibration stimulation frequency screening experiment, the pre-processing of the electroencephalogram signal comprises: removing M1, M2, CB1 and CB2 leads, re-referencing to common average reference, performing 0.5-40 Hz band-pass filtering, down-sampling to 200 Hz, taking the data segment of-2 s to 4 s with the stimulation starting time as 0 time, and calibrating with-2 s to 0 s as the baseline. Then, the 0.5-4 s data segment of the pre-processed electroencephalogram data is taken for FFT, and the best stimulation frequency is screened out by comparing the amplitude differences of the SSSEP induced in different stimulation frequencies in the related leads. The FFT algorithm converts the electroencephalogram sequence from the time domain to the frequency domain, and calculates the frequency components contained in the electroencephalogram sequence. The calculation formula is as follows:

[0044]

[0045] In the above formula, N represents the sampling frequency, e represents the natural logarithm base, x[n] represents the time domain finite length discrete electroencephalogram sequence, and in the calculation result, is a complex number, which can be denoted as The amplitude response corresponding to the frequency can be obtained by calculating the modulus of the first N / 2 points.

[0046] In the vibration stimulation selective attention experiment, the pre-processing of the electroencephalogram signal comprises: removing M1, M2, CB1 and CB2 leads, re-referencing to common average reference, performing 8-40 Hz band-pass filtering, and down-sampling to 200 Hz. Then, the data segment of 1-5 s during the selective attention task is taken with the direction prompt starting time as 0 time, feature extraction and classification recognition are performed.

[0047] The feature extraction method adopts a common spatial pattern (CSP) algorithm, the basic principle of the algorithm is that through designing a spatial filter, multi-dimensional data is mapped to low dimension, the variance of one kind of electroencephalogram signal is maximized, and the variance of another kind of electroencephalogram signal is minimized, so that the difference between two kinds of target electroencephalogram signals to be classified is maximized, and the calculation formula is as follows:

[0048] X CSP =W T X

[0049] Wherein, X is an original electroencephalogram signal, the dimension is N*T, wherein N is the number of leads, and T is the sample point number. W is a spatial filter matrix to be solved, each column vector w j ∈W N*N (j=1, 2...N) is a filter. X CSP is the filtered electroencephalogram signal.

[0050] The classification recognition method adopts a support vector machine (SVM) algorithm. The SVM is a commonly used two-class algorithm, the basic idea of the algorithm is to find a classification hyperplane, so that the two kinds of target data can be distinguished, and the distance between the plane and the two kinds of data should be maximized. Compared with other classification algorithms, the SVM is especially suitable for the case that the sample size is small and the data dimension is high. In the embodiment of the application, the function in the LIBSVM tool package based on MATLAB is used to realize the specific classification operation.

[0051] The application can be applied to the fields of clinical rehabilitation, thought control, game entertainment and the like, and a perfect system can be constructed through further research, and considerable social and economic benefits can be expected.

[0052] In the embodiment of the application, the model of each device is not limited, as long as the device can complete the above functions.

[0053] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred embodiment, and the above embodiment numbers are only for description, not representing the advantages and disadvantages of the embodiments.

[0054] The above is only a preferred embodiment of the application, and does not limit the application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A brain-computer interface system based on selective attention to vibration stimulation, characterized in that, The system includes: a vibration tactile stimulation device, an electroencephalogram (EEG) acquisition system, and a computer for stimulation interface presentation and data processing; Subjects sat on a chair 70cm in front of the stimulation interface. Vibration stimulation was applied to the index fingers of both hands according to the experimental paradigm. The stimulation frequency was set to the optimal specific stimulation frequency that had been pre-selected. Subjects selectively focused their attention on the target side according to the prompts on the stimulation interface. The EEG acquisition system collects the subject's EEG data. The collected data undergoes subsequent preprocessing, feature extraction, and classification to identify the target stimulus that the subject is interested in, and the identification results are fed back to the subject through vibration stimulation. In the experiment, the target stimulus location was the index finger of the left or right hand. The frequency selection process of both index fingers was carried out separately. The duration of vibration stimulation in a single trial was 4 s, the interval between stimulations was 2 s, and the stimulation frequency was a random frequency with a step size of 2 Hz in the range of 17 to 35 Hz. First, maintain a 4-second black screen silence. Then, a circle appears in the center of the screen for 3.5 seconds to prompt the subject to focus their attention. After the circle disappears, addition and subtraction numbers appear in the center of the screen. The subject is asked to focus their attention on the addition and subtraction calculations and answer the calculation results during the black screen rest period after the numbers disappear. The experiment continues after the circle prompts reappear. The system also includes: preprocessing the EEG data collected in the experiment, calculating the average FFT result of 30 trials at the same stimulation frequency, analyzing the EEG response spectrum of the contralateral anterior and middle parietal regions with obvious activation phenomena in leads F, FC, C, and CP, and selecting the stimulation frequencies that can induce the highest SSSEP amplitude at 10 screening frequencies in all leads as the optimal specific stimulation frequencies for the target limbs of the subjects.

2. The brain-computer interface system based on selective attention to vibration stimulation according to claim 1, characterized in that, In the experiment, each group consisted of 10 trials, covering 10 random stimulation frequencies from 17 to 35 Hz, and each block consisted of 5 groups. The left and right index fingers were used for 6 blocks each to improve the testing accuracy. Each index finger contained data from 30 trials at the same stimulation frequency. Between blocks, the subjects rested and adjusted according to their fatigue levels.

3. The brain-computer interface system based on selective attention to vibration stimulation according to claim 1, characterized in that, The minimum difference between the two frequencies must be 4 Hz. The optimal specific stimulation frequency for the index fingers of each subject's left and right hands is determined, and the optimal specific stimulation frequency is applied to the vibration stimulation selective attention experiment.

4. A brain-computer interface system based on selective attention to vibration stimulation according to claim 3, characterized in that, In the experiment, the target stimulation locations were the index fingers of the left and right hands, and the stimulation frequencies of the two channels of the vibration tactile stimulation device were set to the selected optimal specific stimulation frequencies.

5. A brain-computer interface system based on selective attention to vibration stimulation according to claim 3, characterized in that, At the start of each trial, a cross appears in the center of the screen for 1 second to prompt the subject to prepare. After the cross disappears, the vibratory tactile stimulation device applies vibration stimulation to the index fingers of both hands simultaneously, and a circle appears on the screen for 1 second to prompt the subject to prepare to perform a selective attention task. After the circle disappears, a selective attention direction prompt appears on the screen, which is represented by an arrow pointing to the left or to the right. After 1 second, the direction prompt disappears and the screen goes black. After seeing the directional prompts appear, the participants immediately performed the corresponding tasks according to the prompts; When the directional cue is an arrow pointing to the left, participants are required to focus their attention on the vibration of their left index finger while ignoring the vibration of their right index finger. When the directional cue is an arrow pointing to the right, participants are required to focus on the vibration of their right index finger while ignoring the vibration of their left index finger.

6. A brain-computer interface system based on selective attention to vibration stimulation according to claim 3, characterized in that, In the selective attention task, participants were required to roughly count the number of vibrations of the motor. The vibration stimulus lasted for 6 seconds throughout the trial. After seeing the directional cue for the selective attention task, participants focused on the vibration stimulus on the target side according to the direction of the cue until the vibration stimulus ended. During the training phase, the interval between trials was 2-3 seconds. During this period, the subjects remained resting and adjusted their state to wait for the next trial. A total of 10 blocks of EEG data were collected during the training phase.

Images