Brain-computer interface system construction method and device for neural feedback training

Abstract

The invention discloses a brain-computer interface system construction method and device for neural feedback training, and relates to the field of neural feedback training.The method comprises the steps that T-shaped channel layout of near-infrared light brain imaging equipment is designed, specifically, light sources and receivers are combined in pairs to carry out T-shaped layout on a plane, and the light sources and the receivers are arranged in pairs; the adjacent light sources and receivers are combined to form a data acquisition channel, and a plurality of combined data acquisition channels can be mapped to the whole brain; a BCI communication module, a band-pass filtering module, a baseline calculation and self-adaption module, a feedback mode selection and calculation module and an interactive display module are configured in a brain-computer interface computer connected with the near-infrared light brain imaging device. According to the invention, the adjusting effect of the resting brain state in the multi-task feedback training process can be known.

Description

Technical Field

[0001] The present invention relates to the field of neurofeedback training, and in particular to a method and device for constructing a brain-computer interface system for neurofeedback training. Background Art

[0002] Traditional neurofeedback training methods improve cognitive functions related to specific brain regions by closed-loop regulation of the activation level of that region, but traditional neurointervention methods have limited effects on improving cognitive functions. The functions of the brain, especially advanced cognitive functions, are supported by a wider range of brain regions. These brain regions are interconnected to form a complex brain network, and cognitive functions emerge on the basis of this complex brain network. The construction of a closed-loop neurofeedback training method based on brain-computer interface technology can effectively regulate the overall topological properties of the brain network instead of just limited intervention in individual brain regions. It more directly reflects specific cognitive functions and is of great significance for the rehabilitation treatment of mental illness and cognitive intervention for special populations. In addition, the resting brain reflects important intrinsic characteristics of the individual, which can effectively identify mental illness.

[0003] In the prior art, electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI) are used as imaging modalities to construct brain-computer interfaces to achieve closed-loop neural regulation. EEG records the brain's electrical activity through electrodes placed on the scalp. Its high temporal resolution can monitor brain activity in real time, but its disadvantages are low signal-to-noise ratio and low spatial resolution. fMRI uses magnetic fields and radio frequency pulses to measure changes in the brain's blood oxygen level to monitor neural activity. Its high spatial resolution can accurately locate brain activity areas and has good imaging quality. Its disadvantages are that it is sensitive to head movements and is expensive.

[0004] However, the above-mentioned technology does not take into account the resting-state regulation effect of the brain in the feedback training of the brain network, and does not fully consider the cognitive function of the brain, so it cannot fully regulate the overall topological properties of the brain network affected by feedback training. Summary of the invention

[0005] The embodiments of the present invention provide a method and device for constructing a brain-computer interface system for neural feedback training, which can solve the problem in the prior art that training does not take the resting state of the brain into consideration and thus cannot fully adjust the overall topological properties of the brain network.

[0006] The embodiment of the present invention provides a method for constructing a brain-computer interface system for neurofeedback training, comprising the following steps: designing a T-shaped channel layout of a near-infrared light brain imaging device; the T-shaped channel layout comprises: using a combination of light sources and receivers in pairs to perform a T-shaped layout on a plane, adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain; wherein the T-shaped channel layout is used to image the function of a brain area whose brain activity is under neurofeedback training; the near-infrared light brain imaging device is used to convert the received imaging results into mental state data of the brain; a BCI communication module, a bandpass filtering module, a baseline calculation and adaptation module, a feedback mode selection and calculation module, and an interactive display module are configured in a brain-computer interface computer connected to the near-infrared light brain imaging device; wherein the BCI communication module is used to convert the mental state data into an electroencephalogram signal; the bandpass filtering module is used to extract a signal in a required frequency range from the electroencephalogram signal; the baseline calculation and adaptation module is used to determine the level of brain activity in a resting state; the feedback mode selection and calculation module is used to feedback the changes between the electroencephalogram signal and the baseline; and the interactive display module is used to display the feedback results and adjust the brain activity according to the displayed results.

[0007] Furthermore, the tasks of the neurofeedback training include: a resting state test task, a real-time feedback training task and a delayed feedback training task; the resting state test task requires observing the symbol "+" in the center of the display device and keeping the brain in a resting state without thinking for three minutes to obtain the first mental state data of the brain; the real-time feedback training task requires observing the status bar on the right side of the center of the display device, adjusting the brain's mental state to change according to the prompt information displayed by the display device, and obtaining the second mental state data of the brain; the status bar is raised by the change of the second mental state data, maintained for thirty seconds, and then rested for fifteen seconds, and repeated twenty times; the delayed feedback training task requires observing the status bar on the right side of the center of the display device and adjusting the brain's mental state to change according to the prompt information displayed by the display device, and obtaining the second mental state data of the brain; the status bar is raised according to the change of the second mental state data, maintained for thirty seconds, and then rested for fifteen seconds, and repeated twenty times; The training task requires continuous observation of the status bar and adjustment of the brain's mental state to produce changes, and obtain the brain's third mental state data; after keeping the status bar still for twenty-five seconds, adjust the status bar according to the third mental state data, and obtain the corresponding score according to the status bar after five seconds, rest for fifteen seconds, and repeat twenty times; construct a seven-day brain network adjustment task as neurofeedback training, and the brain network adjustment task performs a resting state test task before the start, a real-time feedback training task is performed on the first, third and fifth days respectively, and a delayed feedback training task is performed on the second, fourth and sixth days respectively, and a resting state test task is performed after the brain network adjustment task is completed.

[0008] Furthermore, the observer needs to sit upright in front of the display device, with the eyes 60 to 70 centimeters away from the screen.

[0009] Furthermore, the T-shaped channel layout is used to image the function of the brain area whose brain activity is under neurofeedback training. The specific steps include: imaging the function of the prefrontal lobe area of ​​the brain through the T-shaped channel layout set in the forehead part of the whole-brain nylon cap; uploading the imaging results to the brain-computer interface computer.

[0010] Furthermore, the T-shaped channel layout specifically includes: the channel layout adopts a combination of twelve pairs of light sources and receivers in a T-shaped layout on the plane, each light source and each receiver constitute a data acquisition channel, a total of 35 data acquisition channels are combined, and the distance between adjacent light sources and receivers is three centimeters.

[0011] An embodiment of the present invention provides a brain-computer interface system construction device for neural feedback training, comprising:

[0012] The device setting module is used to design the T-shaped channel layout of the near-infrared light brain imaging device; the T-shaped channel layout includes: using a combination of light sources and receivers in pairs to form a T-shaped layout on a plane, adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain; wherein the T-shaped channel layout is used to image the function of the brain area under the neurofeedback training; the near-infrared light brain imaging device is used to convert the received imaging results into the mental state data of the brain; the signal processing module is used to configure a BCI communication module, a bandpass filtering module, a baseline calculation and adaptation module, a feedback mode selection and calculation module and an interactive display module in a brain-computer interface computer connected to the near-infrared light brain imaging device; wherein the BCI communication module is used to convert the mental state data into an electroencephalogram signal; the bandpass filtering module is used to extract the signal of the required frequency range from the electroencephalogram signal; the baseline calculation and adaptation module is used to determine the resting brain activity level; the feedback mode selection and calculation module is used to feedback the changes between the electroencephalogram signal and the baseline; the interactive display module is used to display the feedback results and adjust the brain activity according to the displayed results.

[0013] The embodiment of the present invention provides a method and device for constructing a brain-computer interface system for neural feedback training. Compared with the prior art, the beneficial effects thereof are as follows:

[0014] Design a T-shaped channel layout of a near-infrared light brain imaging device; the T-shaped channel layout includes: using a combination of light sources and receivers in pairs to form a T-shaped layout on a plane, adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain; wherein the T-shaped channel layout is used to image the function of the brain area under the neurofeedback training; the near-infrared light brain imaging device is used to convert the received imaging results into the mental state data of the brain. Configure a BCI communication module, a bandpass filtering module, a baseline calculation and adaptation module, a feedback mode selection and calculation module, and an interactive display module in a brain-computer interface computer connected to the near-infrared light brain imaging device; wherein the BCI communication module is used to convert the mental state data into an electroencephalogram signal; the bandpass filtering module is used to extract the signal of the required frequency range from the electroencephalogram signal; the baseline calculation and adaptation module is used to determine the resting brain activity level; the feedback mode selection and calculation module is used to feedback the changes between the electroencephalogram signal and the baseline; and the interactive display module is used to display the feedback results and adjust the brain activity according to the displayed results.

[0015] Among them, the T-shaped channel layout of the near-infrared brain imaging device is used to image the functions of the brain areas whose brain activity is under neurofeedback training. In the brain-computer interface computer connected to the near-infrared brain imaging device, the baseline calculation and adaptive modules are used to determine the resting-state brain activity level. Therefore, resting-state brain activity is added as a reference in neurofeedback training, and a comprehensive study of the brain's mental state is conducted. The resting-state regulation effect of the brain network during neurofeedback training is taken into account. The neurofeedback training of the brain network affects the overall topological properties of the brain network, thereby achieving full regulation of the overall topological properties of the brain network. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a system architecture provided by an embodiment of the present invention;

[0017] Figure 2 A schematic diagram of channel layout and whole-brain channel positioning provided by an embodiment of the present invention;

[0018] Figure 3 A schematic diagram of a brain network feedback form provided by an embodiment of the present invention;

[0019] Figure 4 A schematic diagram of a brain network regulation task provided in an embodiment of the present invention (A is a resting state test task, B is a real-time feedback training task, C is a delayed feedback training task, and D is the entire training process);

[0020] Figure 5 A schematic diagram of improving the brain network regulation ability and inhibitory control ability provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0022] See also Figure 1 to Figure 5 The embodiment of the present invention provides a method for constructing a brain-computer interface system for neural feedback training, comprising the following steps:

[0023] Step 1: Design a T-shaped channel layout for near-infrared brain imaging equipment. The T-shaped channel layout includes: using a combination of light sources and receivers in pairs to form a T-shaped layout on a plane. Adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain. Among them, the T-shaped channel layout is used to image the functions of brain areas under neurofeedback training. The near-infrared brain imaging equipment is used to convert the received imaging results into mental state data of the brain.

[0024] Specifically, the channel layout uses a combination of twelve pairs of light sources and receivers in a T-shaped layout on the plane. Each light source and each receiver constitute a data acquisition channel, and a total of 35 data acquisition channels are combined. The distance between adjacent light sources and receivers is three centimeters.

[0025] Among them, the tasks of neurofeedback training include: resting state test task, real-time feedback training task and delayed feedback training task. The trainee observes the symbol "+" in the center of the display device and remains in a resting state without thinking for three minutes, which is used as the resting state test task. The trainee observes the status bar on the right side of the center of the display device, which rises and falls according to the changes in the mental state; the trainee adjusts the mental state according to the displayed prompt information to make the status bar rise, and lasts for thirty seconds, and rests in a resting state for fifteen seconds after completion, and repeats twenty times, which is used as the real-time feedback training task. The trainee continues to adjust the mental state according to the previous training experience to make the status bar rise, but the picture remains still in the first 25 seconds, and the trainee cannot obtain feedback information. After the last 5 seconds, the progress bar begins to change in real time, and the trainee can obtain feedback information. After a total of 30 seconds of training, the corresponding score is obtained, and after completion, the trainee rests in a resting state for 15 seconds; the cycle is repeated twenty times, which is used as the delayed feedback training task.

[0026] The resting-state test task, real-time feedback training task and delayed feedback training task were used as tasks for brain network regulation training; brain network regulation training was carried out for seven days, with a resting-state test task before the start of training, a real-time feedback training task on the first, third and fifth days respectively, a delayed feedback training task on the second, fourth and sixth days respectively, and a resting-state test task after the training.

[0027] Step 2: Configure a BCI communication module, a bandpass filter module, a baseline calculation and adaptation module, a feedback mode selection and calculation module, and an interactive display module in a brain-computer interface computer connected to a near-infrared brain imaging device. Among them, the BCI communication module is used to convert mental state data into EEG signals. The bandpass filter module is used to extract signals in the required frequency range from the EEG signals. The baseline calculation and adaptation module is used to determine the resting brain activity level. The feedback mode selection and calculation module is used to feedback the changes between the EEG signals and the baseline. The interactive display module is used to display the feedback results and adjust the brain activity according to the displayed results.

[0028] The near-infrared light brain imaging device fNIRS is connected to the brain-computer interface computer through a network switch to transmit brain imaging data to the brain-computer interface computer in real time. The brain-computer interface computer is connected to the near-infrared light brain imaging device fNIRS through a serial cable to send data tag information from the computer to the fNIRS device. The brain-computer interface computer is connected to peripherals such as a monitor, keyboard, and mouse to receive the trainee's operation of the brain-computer interface computer.

[0029] The human brain accounts for 5% of the body's weight but consumes 20% of the energy. At the same time, the energy consumption of the brain during active tasks increases by less than 5%, which shows that the brain is not only a high-energy-consuming organ, but also that it never stops working, even when deliberately avoiding its "activity". The resting state of the brain is the state of the brain when an individual is not performing any active tasks or deliberate thinking activities. Studies have shown that the resting state of the brain reflects the important inherent characteristics of the individual, which can effectively identify mental illnesses, such as attention deficit hyperactivity disorder ADHD, autism spectrum disorder ASD, antisocial personality disorder ASPD, etc. Even some important individual performance differences can be shown in the resting state of the brain. Unlike the task state, the resting state of the brain has no obvious activation, or although there is activation, these random activations will be annihilated after long-term average superposition, thus showing a steady-state characteristic. An important method for extracting network features in the resting state is independent component analysis. Through the correlation analysis of time series between brain regions, a set of brain regions with synchronous fluctuation characteristics in the resting state can be extracted. The default mode network is separated in this way. This is a brain network state that occurs when an individual is not performing a specific task or performing a self-reference task. Studies have shown that the resting state brain network contains rich information and may reflect the inherent characteristics of brain structural connections. Effective intervention in the resting state brain network is an important target for the treatment of mental illness.

[0030] Current neurofeedback technology methods generally use task adjustment methods. Although these adjustment methods under task conditions are believed to have a corresponding impact on the resting brain network, the direction and effect of adjustment are still controversial. This is mainly because the feedback itself, as a constantly changing stimulus, will have a distracting effect on the trainees. The trainees are busy processing the stimulus information and have no time to consider establishing a deep connection between brain activity and active will. The present invention uses an intermittent feedback method, which is more in line with the law of learning, so that individuals have sufficient time to verify their conjectures and improve their brain network adjustment ability through continuous attempts.

[0031] Compared with the traditional EEG method and fMRI neurofeedback method, this invented method is more suitable for promotion and application. It can be used for rehabilitation treatment of people with mental illnesses, such as attention deficit hyperactivity disorder ADHD, autism spectrum disorder ASD, antisocial personality disorder ASPD, etc. It can also be used to improve the symptoms of obsessive-compulsive disorder, addiction disorder, obesity, mood disorders, etc. This method can also be used for cognitive enhancement training of special healthy people, such as people in high-intensity operations, pilots, soldiers, etc.

[0032] The present invention also has the following beneficial effects:

[0033] (1) The overall topological properties of the brain network are used as feedback cues to achieve comprehensive regulation of the whole brain and effectively improve cognitive performance such as inhibitory control.

[0034] (2) The neurofeedback training method in the form of brain network does not depend on the activation state of brain areas, and can achieve the regulation of resting brain networks, thereby improving behavioral efficacy and maintaining effects.

[0035] (3) Use a variety of feedback presentation forms, combined with intermittent feedback strategies, to motivate trainees.

[0036] (4) The prior art uses feedback from a single brain region (bilateral dorsolateral prefrontal cortex DLPFC) to improve cognitive ability, but fails to fully consider the brain network properties of cognitive ability, resulting in limited training effects. The present invention uses a brain network-based regulation training method to regulate the overall topological properties of the brain, which is more in line with the brain's learning rules and has better cognitive improvement and disease symptom intervention effects;

[0037] (5) The prior art has not fully examined the functional changes in the brain, and the regulatory effect on the resting-state brain is unknown. The present invention adopts a training method in the form of delayed feedback, fully considering the changing laws of resting-state regulation, and has better training transfer effect and maintenance effect.

[0038] An embodiment of the present invention provides a brain-computer interface system construction device for neural feedback training, comprising:

[0039] The device setting module is used to design the T-shaped channel layout of the near-infrared brain imaging device; the T-shaped channel layout includes: using a combination of light sources and receivers in pairs to make a T-shaped layout on a plane, adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain; wherein the T-shaped channel layout is used to image the functions of brain areas whose brain activities are under neurofeedback training; the near-infrared brain imaging device is used to convert the received imaging results into mental state data of the brain.

[0040] A signal processing module is used to configure a BCI communication module, a bandpass filtering module, a baseline calculation and adaptation module, a feedback mode selection and calculation module, and an interactive display module in a brain-computer interface computer connected to a near-infrared brain imaging device; wherein the BCI communication module is used to convert mental state data into electroencephalogram signals; the bandpass filtering module is used to extract signals in a required frequency range from the electroencephalogram signals; the baseline calculation and adaptation module is used to determine the level of resting brain activity; the feedback mode selection and calculation module is used to feedback changes between the electroencephalogram signals and the baseline; and the interactive display module is used to display the feedback results and adjust brain activity according to the displayed results.

[0041] A specific embodiment is as follows:

[0042] 1. System construction

[0043] The system hardware includes near-infrared brain imaging equipment fNIRS, brain-computer interface computer and display equipment, as well as necessary data link equipment such as network switches, serial communication cables, etc. The system architecture is as follows: Figure 1 shown.

[0044] (1) Connect the fNIRS device to the BCI computer through a network switch, and correctly set the LAN IP address and subnet mask of the fNIRS device and computer. Use the ping command to debug and test the network connectivity for real-time transmission of brain imaging data from the fNIRS device to the BCI computer.

[0045] (2) Connect the computer and fNIRS device via a serial cable and set the corresponding serial port number on the computer for the computer to send data tag information to the fNIRS device.

[0046] (3) The brain-computer interface computer is connected to the monitor, keyboard, mouse and other peripherals. The subject sits in front of the screen with his eyes 60-70 cm away from the screen.

[0047] (4) Channel layout design is illustrated by taking the targeted rehabilitation treatment of ADHD patients as an example. The core cognitive defect of ADHD patients is inhibitory control, and the corresponding key functional brain area is located in the prefrontal lobe. The present invention selects this brain area as the intervention and regulation brain area. Other disease types or cognitive ability interventions can flexibly select the brain area of ​​interest according to the specific situation. The present invention adopts a combination of 12 pairs of light sources and receivers. The distance between the light sources and receivers is 3 cm. A T-shaped layout is adopted on the plane, and a total of 35 data acquisition channels are combined. These 35 data channels are mapped to the forehead of the whole-brain nylon cap, and the motion tracking and positioning system FSTRAK is used to measure the actual spatial coordinates of the channel positions and convert them into MNI standard brain coordinates of the standard spatial coordinate system for neuroimaging data, as well as the Brodmann brain area division of the cerebral cortex. The registration operation shows that the data channels can basically cover most of the frontal lobe brain areas. The channel layout and the whole-brain channel positioning are as follows: Figure 2 shown.

[0048] (5) The system software has functions such as brain-computer interface communication and network topology calculation and display, and specifically includes five modules: BCI communication module, bandpass filtering module, baseline calculation and adaptation module, feedback mode selection and calculation module, and interactive display module. The feedback mode selection and calculation module includes three feedback forms, namely the traditional single brain region feedback form, the functional connection feedback form between brain regions, and the brain network feedback form adopted by the present invention. The software design process is as follows: Figure 3 shown.

[0049] 2. Brain network neurofeedback training

[0050] The brain network regulation task includes three independent tasks: resting state test task, real-time feedback training task, and delayed feedback training task.

[0051] Task 1: Resting state test task, used to check and verify the intervention effect. The trainees are required to sit in front of the screen, keep their eyes open, and look at the "+" fixation point in the center of the screen. They should relax their body and mind as much as possible and do not need to think about anything. This task lasts for 3 minutes. Figure 4 As shown in A.

[0052] Task 2: Real-time feedback training task, used for real-time feedback regulation of brain networks. A status bar in the form of a thermometer appears in the center right of the screen. This status bar is a real-time measurement of the state of the brain network and does not represent the actual temperature. Trainees need to self-regulate their mental state according to the prompts on the screen so that the status bar is as high as possible, and gradually establish a connection between their own mental state and the topological characteristics of the brain network to master the regulation rules. Each adjustment training trial lasts 30 seconds, followed by a 15-second break. A total of 20 trials are conducted, taking 15 minutes. Figure 4 As shown in B.

[0053] Task three: Delayed feedback training task, used for regulation training of resting-state brain networks. This form of intermittent feedback can effectively transfer the regulatory ability learned in real-time feedback training to the resting state. The brain-computer interface computer continuously acquires and analyzes the brain mental state data collected by the near-infrared brain imaging device. Within 25 seconds after the start of training, the analysis results are stored in the computer but are not fed back to the trainees. The display device screen always remains static and does not change. Within the last 5 seconds of training, the analysis results are fed back to the trainees in real time through changes in the status bar as a delayed feedback training task. After completion, rest for 15 seconds and proceed to the next trial. A total of 20 trials are conducted, taking 15 minutes. Figure 4 As shown in C.

[0054] The whole training process was completed within 7 days. A resting state test task was conducted before the start of the first day of training and after the end of the seventh day of training. A real-time feedback training task was conducted every day on the 1st, 3rd and 5th days. A delayed feedback training task was conducted every day on the 2nd, 4th and 6th days. Figure 4 As shown in D.

[0055] A comparative analysis of the training methods was conducted. By setting up a pseudo-feedback group and a traditional continuous feedback group for comparison, the regulation level of the brain network of the individuals after intermittent neurofeedback training was significantly improved, but there was no significant change in the pseudo-feedback group and the continuous feedback group. The cognitive performance before and after training (using the Stroop task used to study attention, cognitive control and interference processing) was tested. Except for the pseudo-feedback group, the cognitive performance of the intermittent feedback group and the continuous feedback group was significantly improved, but the cognitive performance of the intermittent feedback group improved more. The results are as follows Figure 5 This shows that neurofeedback training can significantly improve cognitive performance, and intermittent feedback can not only effectively improve cognitive performance, but also deeply change the resting brain network.

[0056] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for constructing a brain-computer interface system for neurofeedback training, characterized in that: The following steps are involved: Design the T-shaped channel layout of the near-infrared brain imaging device; The T-shaped channel layout includes: using a combination of light sources and receivers in pairs to perform a T-shaped layout on a plane, adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain; The T-shaped channel layout is used to image the function of the brain area whose brain activity is under neurofeedback training; the near-infrared brain imaging device is used to convert the received imaging results into mental state data of the brain; A BCI communication module, a bandpass filtering module, a baseline calculation and adaptation module, a feedback mode selection and calculation module, and an interactive display module are configured in a brain-computer interface computer connected to the near-infrared brain imaging device; Among them, the BCI communication module is used to convert mental state data into EEG signals; the bandpass filtering module is used to extract signals in the required frequency range from the EEG signals; the baseline calculation and adaptation module is used to determine the resting brain activity level; the feedback mode selection and calculation module is used to feedback the changes between the EEG signals and the baseline; the interactive display module is used to display the feedback results and adjust the brain activity according to the displayed results.

2. A method for constructing a brain-computer interface system for neural feedback training as claimed in claim 1, characterized in that: The tasks of the neurofeedback training include: a resting state test task, a real-time feedback training task and a delayed feedback training task; The resting state test task requires observing the symbol "+" in the center of the display device and keeping the brain in a non-thinking resting state for three minutes to obtain the first mental state data of the brain; The real-time feedback training task requires observing the status bar on the right side of the center of the display device, adjusting the brain's mental state to change according to the prompt information displayed by the display device, and obtaining the second mental state data of the brain; the status bar is raised by the change of the second mental state data, maintained for thirty seconds, and then rested for fifteen seconds, and repeated twenty times; The delayed feedback training task requires continuous observation of the status bar and adjustment of the brain's mental state to produce changes, and acquisition of the brain's third mental state data; after keeping the status bar still for twenty-five seconds, the status bar is adjusted according to the third mental state data, and after five seconds, the corresponding score is acquired according to the status bar, and a fifteen-second rest is performed, and the cycle is repeated twenty times; A seven-day brain network regulation task was constructed as neurofeedback training, wherein a resting-state test task was performed before the start of the brain network regulation task, a real-time feedback training task was performed on the first, third and fifth days respectively, a delayed feedback training task was performed on the second, fourth and sixth days respectively, and a resting-state test task was performed after the end of the brain network regulation task.

3. A method for constructing a brain-computer interface system for neural feedback training as claimed in claim 2, characterized in that: The observer needs to sit upright in front of the display device, with the eyes 60 to 70 centimeters away from the screen.

4. A method for constructing a brain-computer interface system for neural feedback training as claimed in claim 1, characterized in that: The T-shaped channel layout is used to image the function of a brain region whose brain activity is under neurofeedback training, and the specific steps include: The function of the prefrontal region of the brain was imaged through a T-shaped channel layout set in the frontal part of the whole-brain nylon cap; Upload the imaging results to the brain-computer interface computer.

5. A method for constructing a brain-computer interface system for neural feedback training as claimed in claim 1, characterized in that: The T-shaped channel layout specifically includes: The channel layout uses a combination of twelve pairs of light sources and receivers in a T-shaped layout on the plane. Each light source and each receiver constitute a data acquisition channel, and a total of 35 data acquisition channels are combined. The distance between adjacent light sources and receivers is three centimeters.

6. A device for constructing a brain-computer interface system for neurofeedback training, characterized in that: include: The device setup module is used to design the T-shaped channel layout of the near-infrared brain imaging device; The T-shaped channel layout includes: using a combination of light sources and receivers in pairs to form a T-shaped layout on a plane, adjacent light sources and receivers are combined to form a data acquisition channel, and the combined multiple data acquisition channels can be mapped to the whole brain; wherein the T-shaped channel layout is used to image the function of the brain area under the neurofeedback training; the near-infrared brain imaging device is used to convert the received imaging results into the mental state data of the brain; A signal processing module is used to configure a BCI communication module, a bandpass filtering module, a baseline calculation and adaptation module, a feedback mode selection and calculation module, and an interactive display module in a brain-computer interface computer connected to the near-infrared brain imaging device; wherein the BCI communication module is used to convert mental state data into electroencephalogram signals; the bandpass filtering module is used to extract signals in a required frequency range from the electroencephalogram signals; the baseline calculation and adaptation module is used to determine the level of resting brain activity; the feedback mode selection and calculation module is used to feedback changes between the electroencephalogram signals and the baseline; and the interactive display module is used to display the feedback results and adjust brain activity according to the displayed results.

Images