Closed-loop brain function training method based on brain-computer interface system

Abstract

The application discloses a closed-loop brain function training method based on a brain-computer interface system and particularly relates to the technical field of brain-computer interface clinical application; the user state value is obtained by comprehensively analyzing the electroencephalogram physiological parameters and the behavior function parameters, the task optimization value is calculated by combining the task completion degree, the task accuracy and the task time according to the parameter reference values of different age stages, and the problems that the training effect evaluation depends on subjective scoring or single behavior index, lacks unified quantitative standard and has no hierarchical reference values of different age stages, thereby leading to low parameter quantitative reliability are solved.

Description

Technical Field

[0001] This invention relates to the field of clinical application technology of brain-computer interfaces, and more specifically, to a closed-loop brain function training method based on a brain-computer interface system. Background Technology

[0002] With the increasing aging of the population and the high incidence of neurological diseases, the demand for brain function rehabilitation continues to grow. Brain function training, as a means to improve cognitive and motor function deficits, has become a research hotspot in the field of neurorehabilitation.

[0003] However, existing closed-loop brain function training methods based on brain-computer interface systems still have the following shortcomings: First, existing assessments rely heavily on subjective ratings or single behavioral indicators, failing to integrate EEG physiological parameters with behavioral function parameters and lacking multi-dimensional quantitative support. Furthermore, the lack of a stratified reference value system for different age groups results in low reliability of parameter quantification results and makes it difficult to objectively correlate with the user's cognitive function status. Secondly, existing training methods mostly use general task templates and do not combine user age, clinical diagnosis results and EEG physiological characteristics for targeted matching; at the same time, they lack a structured labeling system, the task selection process is subjective, and it is impossible to accurately locate the user's core functional deficits, resulting in insufficient training targeting and difficulty in meeting the personalized rehabilitation needs of different users. Finally, once the training plan was determined, it was implemented in a fixed manner. There were no pre-set multi-task review points during the training cycle to continuously collect data such as EEG physiological parameters, task completion time, and task excellence value. There was also no comprehensive formation of a confidence estimate of the plan by calculating quantitative indicators such as the change ratio of task excellence value and time series coefficient. Furthermore, there was no mechanism to screen low-confidence plans based on the confidence estimate and make targeted adjustments. As a result, the training could not be optimized and adjusted according to the dynamic progress of the user's functional recovery, and the overall rehabilitation efficiency was limited. Therefore, a closed-loop brain function training method based on a brain-computer interface system was developed to solve the above problems. Summary of the Invention

[0004] To overcome the above-mentioned deficiencies of the prior art, embodiments of the present invention provide a closed-loop brain function training method based on a brain-computer interface system.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A closed-loop brain function training method based on a brain-computer interface system includes the following steps: Parameter acquisition: Collect the target user's electroencephalographic parameters and basic clinical parameters; Develop a training program: After comprehensive analysis of EEG physiological data, user syndrome values ​​are obtained. Based on clinical diagnostic labels in clinical baseline parameters and target user age, target user deficits and task scenarios are matched. User training task sets are selected from the standardized task library of daily activities using target user deficits and task scenarios. The task difficulty is determined by comparing syndrome values ​​with ideal syndrome values. Standardized tasks are determined by combining target user task scenarios and matched task difficulty. After the target user performs a standardized task, the completion rate, accuracy, and time taken of the standardized task are statistically analyzed to obtain the task excellence value, and the performance level is determined based on the excellence value; based on the performance level, a training scheme that matches the current user's task performance level is matched in the standardized training method library. Training method evaluation: Users carry out brain function training according to the training plan, and the user training data during the training cycle is stored in real time. After the training is completed, the user training data is comprehensively analyzed to obtain the confidence value of the plan. The user training data includes the change ratio of task merit value of improved review points and the proportion of improved review points. Strategy optimization: Preset the confidence value threshold of the scheme corresponding to the scheme confidence value, remove schemes with confidence values ​​lower than the corresponding threshold and send them to the corresponding medical staff terminal, and rematch the training scheme of the current user's task performance level in the standardized training method library.

[0006] Specifically, the process of comprehensively analyzing brain electrophysiological parameters is as follows: The electroencephalometric parameters include the mean β wave, the mean α wave, and the prefrontal lobe ratio. A preset time evaluation window is used to extract β-wave data and α-wave data from each time acquisition node. At each time node, the data segments corresponding to the focused state and the data segments corresponding to the relaxed state are collected. The mean of each time node segment is calculated. Then, the arithmetic mean of the mean of β-wave data and the mean of α-wave data from each acquisition node are calculated to obtain the mean of β-wave and the mean of α-wave. Extract the theta wave and beta wave data of the prefrontal cortex of each acquisition node within the time evaluation window. For each acquisition node, first calculate the arithmetic mean of the theta wave energy and the arithmetic mean of the beta wave energy of that node, and then calculate the ratio of the node's theta wave mean to the node's beta wave mean to obtain the prefrontal cortex ratio of each acquisition node. The user syndrome value is obtained by comprehensively processing the mean of the β wave, the mean of the α wave, and the prefrontal ratio.

[0007] Specifically, the process for determining the standardization tasks for target users is as follows: Basic clinical parameters include age and clinical diagnostic labels; Identify the user's clinical diagnosis label, and pre-set each type of diagnosis label corresponds to a defect; after matching the clinical diagnosis label with the defect, filter out the user training task set, where the task library stores the user training tasks corresponding to different defects; The target users are categorized by age into pre-defined age ranges, and each age range corresponds to a specific task scenario. Identify the pre-labeled task difficulty tags for each group of tasks in the user's training task set. The task difficulty tags include both easy and complex. A preset ideal value for the status quo is set. The status quo value is compared with the ideal value. If the status quo value is lower than the ideal value, a simple task is selected from the task set. If the status quo value is higher than the ideal value, a complex task is selected from the task set. The target user's status quo value is matched with the ideal value to determine the task difficulty. Select tasks from the standardized task library for daily activities that can assess the target user's shortcomings, and determine standardized tasks by combining the target user's task scenario and the matching task difficulty.

[0008] Specifically, the calculation logic for the completion rate, accuracy, and time taken for standardized tasks is as follows: After the target user executes the matched standardized task, the execution data of the task is statistically analyzed, the steps of the standardized task are determined, the number of steps actually completed by the user is counted, and the ratio of the number of steps actually completed to the total number of steps of the task is calculated to obtain the task completion rate. The total number of operations for the standardized task and the number of operations that the user completed correctly are counted. The ratio of the number of correct operations to the total number of operations is calculated to obtain the task accuracy rate. The start time is the time when the user is informed of the standardized task requirements and the execution is initiated, and the end time is the time when the user indicates that the operation is completed. The operation time is recorded as the task time.

[0009] Specifically, the process for determining the task's excellence value is as follows: The task excellence value of the target user performing the standardized task is obtained by comprehensively analyzing the task completion rate, task accuracy, and task time.

[0010] Specifically, the process of determining the matching training scheme is as follows: The task performance level is determined based on the preset task performance value corresponding to each group of task performance value range and the preset mapping rules. Based on different performance levels, a training scheme corresponding to the current user's task performance level is matched in the standardized training method library; the standardized training method library stores each group of training schemes, and each group of training schemes is pre-labeled with the applicable task performance level and defects.

[0011] Specifically, the process of analyzing user training data is as follows: During the training cycle, a task review point is preset. When the task review point is reached, the standardized task is executed again and user training data is collected to calculate the task excellence value. The task excellence value of each review point is calculated. Starting from the user's initial task excellence value, a data sequence is constructed by combining the excellence values ​​of each review point. Two adjacent sets of data are extracted from the data sequence. The ratio of adjacent review points is calculated with the left side of the adjacent data point as the denominator and the right side as the numerator to obtain the change ratio of the task excellence value of the current review point. Calculate the ratio of adjacent review points. If the ratio is greater than 1, mark the corresponding review point as an improved review point. The average change ratio of the task excellence value at each review point is obtained by statistically analyzing the changes in the task excellence value at each review point. The number of points marked as improvement review points during the statistical training period is counted, and the ratio of the number of improvement review points to the total number of review points is calculated to obtain the proportion of improvement review points.

[0012] Specifically, the process for determining the confidence valuation of the proposed solution is as follows: A comprehensive analysis of the average change ratio of the merit value of the review point task and the proportion of the review point improvement is used to obtain the confidence estimate of the solution.

[0013] The technical effects and advantages of this invention are as follows: (1) This invention obtains the user status value by integrating the comprehensive analysis of EEG physiological parameters and behavioral function parameters, and calculates the task excellence value by combining the reference values ​​of parameters for different age groups with the task completion rate, task accuracy rate and task time. This solves the problem that the evaluation of training effect depends on subjective scoring or single behavioral indicators, lacks unified quantitative standards, and has low reliability of parameter quantification due to the lack of stratified reference values ​​for different age groups.

[0014] (2) This invention constructs a task library with structured tags, first matches suitable tasks for users, and then further determines the training method based on the calculated task excellence value. This solves the problem that existing training uses general task templates, unstructured tag systems, and subjective task selection, which leads to the inability to target and adapt to the user's core functional deficiencies.

[0015] (3) By pre-setting multiple task review points during the training period, the average change ratio of the task merit value of each review point and the proportion of the review point are collected. The time series coefficient is determined based on the growth rate of the change ratio of the task merit value of each review point. The scheme confidence value is obtained after comprehensive analysis, and the training effect is comprehensively evaluated. Finally, by screening low confidence schemes below the confidence value threshold, medical staff can remove or adjust them based on user training data. This makes up for the defects that only scattered feedback can be obtained after training and the trend of user function recovery cannot be tracked. At the same time, it solves the dilemma that the training scheme is fixed and cannot be dynamically adjusted based on multi-dimensional quantitative parameters to adapt to the user function recovery process. Attached Figure Description

[0016] Figure 1 This is a flowchart of the closed-loop brain function training method based on a brain-computer interface system according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1 As shown, the closed-loop brain function training method based on the brain-computer interface system is as follows: Parameter acquisition: Collect the target user's electroencephalographic parameters and basic clinical parameters, and perform preprocessing; The electroencephalographic parameters include mean beta wave, mean alpha wave, and prefrontal lobe ratio; the basic clinical parameters include age and clinical diagnostic labels. Additional explanation, 1) Parameter Acquisition

[0019] Electroencephalometric parameters: Using a 16-channel or higher EEG acquisition device (sampling rate 250-500Hz, electrode impedance <30kΩ), in a quiet, light-protected environment, the prefrontal β wave (14-30Hz), occipital α wave (8-13Hz), and prefrontal θ / β ratio were collected under the user's resting focused state (3 minutes), resting relaxed state (3 minutes), and target stimulus task state (30 effective stimuli). Behavioral function parameters: Under the guidance of professionals, functional status scores were collected using the Fugl-Meyer Motion Scale / MMSE Cognitive Scale, and the time taken by the user to complete typical daily tasks such as "independent dressing / digit memory" was tested (the average value was taken after 3 repetitions). Clinical baseline parameters: The user's precise age in one year and clinical diagnostic results related to brain function are collected through medical record verification and user confirmation. 2) Parameter preprocessing EEG data preprocessing: The raw EEG signal is filtered at 50Hz power frequency and ICA is used to remove electrooculography artifacts. Then, the target frequency band energy is extracted by fast Fourier transform. Behavioral / clinical parameter preprocessing: Remove outliers (data deviating from the mean ± 30%) from daily task time, map clinical diagnosis results to quantitative values, and ensure that all parameters are in a uniform format and free from interference; Develop a training program: Construct a unique user identifier based on basic clinical parameters, obtain the user's syndrome value after comprehensive analysis of electroencephalographic parameters, preset the ideal syndrome value corresponding to the user's syndrome value, and determine the training method for the target user based on the ideal syndrome value. Specifically: Extract the target user's age (accurate to one year) and clinical diagnosis tags from the user's clinical information, and convert the clinical diagnosis tags into standardized codes (e.g., no relevant diagnosis is marked as 00, mild cognitive impairment is marked as 01, stroke is marked as 02, attention deficit hyperactivity disorder is marked as 03, etc.). Identify the assessment date and batch serial number. The assessment date is in the format of "year, month, day" and corresponds to the specific date on which the user completed the assessment of EEG physiological parameters and behavioral function parameters. The batch serial number is the sequential number of the user assessed on the same day or in the same batch, in the format of numbers (such as 001, 002, etc.). Generate an identifier by splicing information according to a fixed format. Additional explanation: The "age + diagnosis code + assessment date + serial number" will be combined in a unified structure to form a unique user identifier. For example, if a user is 35 years old, has no relevant diagnosis code, and completes the assessment on November 20, 2025, and is the first user on that day, their corresponding unique identifier will be 35-00-20251120-001; thus completing the binding between the identifier and user data. A preset time evaluation window is used to extract β-wave data and α-wave data from each time acquisition node. At each time node, the data segments corresponding to the focused state of β-wave data and the data segments corresponding to the relaxed state of α-wave data are collected. After calculating the mean of the segments at each time node, the arithmetic mean of the mean of β-wave data and the mean of α-wave data at the acquisition node is calculated to obtain the mean of β-wave data Vai and the mean of α-wave data Tce. Extract the theta and beta wave data of the prefrontal cortex at each acquisition point within the time evaluation window. For each acquisition node, first calculate the arithmetic mean of the theta wave energy and the arithmetic mean of the beta wave energy of that node, and then calculate the ratio of the node's theta wave mean to the node's beta wave mean to obtain the prefrontal cortex ratio Glp of that acquisition node. Based on the differences in brain development maturity among different age groups, the target users were divided into several age groups, with each age group corresponding to a set of reference values. Matching the target user age groups with the reference value sets yielded reference values ​​for the mean β-wave, mean α-wave, and prefrontal ratio, which were labeled as follows: , , ; After normalizing the mean of the beta wave, the mean of the alpha wave, and the prefrontal ratio, the results were entered into the formula. After weighted calculation, the user syndrome value TsT is obtained, where s1, s2, and s3 are the preset weighting factors corresponding to the mean of the β wave, the mean of the α wave, and the prefrontal ratio, respectively. The system identifies users' clinical diagnostic labels, with each label pre-defined as corresponding to a specific deficit (e.g., mild cognitive impairment corresponds to deficits in episodic memory and executive function, while stroke corresponds to deficits in motor coordination and balance). After matching the clinical diagnostic labels with the deficits, the system filters user training task sets from a standardized daily activity task library. The standardized daily activity task library stores user training tasks corresponding to different deficits. The user training task set is obtained by integrating the user training tasks matched with the current defect point. Additional information: The clinical diagnostic label includes: Diagnostic codes: 01 = mild cognitive impairment, 02 = stroke; Defects: Identify the specific defect (e.g., episodic memory, motor coordination, etc.). Age range: Children 6-12 years old, teenagers 13-18 years old, adults 19-59 years old, seniors 60+; Task difficulty: Corresponding syndrome value (easy / complex); Classify target users by age (children / teenagers / adults / elderly) and select task scenarios that match the life experience of the corresponding age group; (e.g., gamified tasks for children, simple tasks such as sorting building blocks, and complex tasks such as building with scenario-based blocks; integrate learning scenarios for teenagers, simple tasks such as memorizing words, and complex tasks such as coordinating homework for multiple subjects). Identify the pre-labeled task difficulty tags for each group of user training tasks within the user training task set; where task difficulty tags include easy and complex. Extract the target user's statistic value, preset an ideal statistic value, and compare the statistic value with the ideal statistic value. If the statistic value is lower than the ideal statistic value, select a simple task (≤2 steps, no interference and prompts) from the user's training task set. If the statistic value is higher than the ideal statistic value, select a complex task (≥5 steps, moderate interference and no prompts) from the user's training task set. The difficulty of the task is determined by comparing the target user's statistic value with the ideal statistic value. Select tasks from the standardized daily activity task library that can assess the target user's shortcomings, and determine the standardized tasks for the target user based on the matching task difficulty. Additional information: Assume the user's information is that they are a 68-year-old with mild cognitive decline (01) and a syndrome value of 0.25 (corresponding to a simple task).

[0020] Defect: Core = Contextual Memory + Executive Function; Screening steps: ① Diagnostic fit: Screen for "MCI01" tags and assess "memory / executive" tasks; ② Age-related scenarios: Retain the "60+ years old" label and "daily self-care scenario" tasks (such as remembering medicine names and organizing items); ③ Difficulty matching: Select "Easy task" (steps 1-2, no distractions and prompts); Candidate tasks: "Remember 2 medicine names + take medicine in order of time" (execution function), "Remember 3 fruits + take them out of the fruit basket" (contextual memory); Verification and adjustment: The user has no contraindications, and the family confirms that "the medication scenario is consistent with daily life", so two tasks are finally determined; Evaluation metrics: For Task 1, assess the accuracy of sorting; for Task 2, assess the accuracy of item retrieval. After the target user performs the matched standardized task, the data of the user's standardized task execution is statistically analyzed, the steps of the standardized task are determined (e.g., memorizing 2 drug names + sorting by time, the steps are memorizing drug names → sorting by time), the number of steps actually completed by the user is statistically analyzed, and the ratio of the number of steps actually completed by the user to the total number of steps of the task is calculated to obtain the task completion degree ta1. The total number of operations for the standardized task is counted (e.g., counting 2 drug names is 2 operations, sorting by time is 1 operation, for a total of 3 operations). The number of operations that the user correctly completed is counted, and the ratio of the number of correct operations to the total number of operations is calculated to obtain the task accuracy rate ta2. The start time is the time when the user is informed of the standardized task requirements and begins execution, and the end time is the time when the user indicates that the operation is completed. The operation time is recorded as the task time ta3. Based on clinical goals and task characteristics, technicians preset ideal values ​​for task completion, task accuracy, and task time corresponding to standardized tasks for target users, and marked them accordingly. , , ; Normalize the task completion rate, task accuracy, and task time before substituting them into the formula. The task excellence value is obtained after weighted calculation, where , , These are the preset weighting factors for task completion rate, task accuracy, and task time, respectively. Additional explanation: Taking a 68-year-old patient with mild cognitive decline completing the task of memorizing two medication names and sorting them by time as an example: The clinical target was preset with ideal values ​​(completion rate 0.8, accuracy rate 0.7, time taken 2.4 minutes); the user's actual completion rate was 0.75, accuracy rate was 0.7, and time taken was 2.5 minutes; after substituting into the formula, the task's optimal value was approximately 0.77. The system presets task excellence values ​​for each group of task excellence value intervals, and each group of task excellence value intervals corresponds to a task performance level. It identifies the task excellence values ​​of the target user and compares the task excellence values ​​with the intervals to determine the task performance level. Based on different performance levels, a training scheme is matched with the current user's task performance level in the standardized training method library; the standardized training method library stores each group of training schemes, and each group of training schemes is pre-labeled with the applicable task performance level, defects, matching training frequency, single training duration, and training cycle. Additional notes: The training cycle is set by technical personnel based on the improvement goals of the training method, such as 10 training sessions per 2 weeks. For example, taking a user with "diagnosis code 01 (mild cognitive decline), 65 years old, deficit in episodic memory and executive function" as an example: their task excellence score is 0.77. By referring to the preset range (0.7-0.9 corresponds to level 2), the task performance level is determined to be 2. Then, from the standardized training method library, the program marked "applicable to level 2, elderly 60+, deficit in episodic memory and executive function" is retrieved (training frequency 5 times a week, 20 minutes per session, 10 times / 2 weeks). This is the training program that the user needs to execute. Training method evaluation: Users perform brain function training according to the training plan, and user training data is stored in real time during the training period. After the training is completed, the user training data is comprehensively analyzed to obtain the confidence value of the plan. The user training data includes the average change ratio of the task excellence value of the improvement review points and the percentage of improvement review points. Specifically: Pre-set task verification points during the training cycle (a total of 5 verification points: the starting verification point is the first time the target user executes the standardized task during the training cycle, the task before the starting verification point is the first time the target user executes the standardized task, and the ending verification point is the target user executes the standardized task after training is completed). Upon reaching the task review point, the standardized task is executed again, and user training data is collected to calculate the task performance value. The task excellence value of each review point is calculated. Starting from the user's initial task excellence value, a data sequence is constructed by combining the excellence values ​​of each review point. Two adjacent sets of data are extracted from the data sequence. The ratio of adjacent review points is calculated with the left side of the adjacent data point as the denominator and the right side as the numerator to obtain the change ratio of the task excellence value of the current review point. Calculate the ratio between adjacent review points. If the ratio is greater than 1, mark the review point as an improved review point. The average value of the change ratio of the task's merit value at each review point is calculated to obtain the average value of the change ratio of the task's merit value at the review point, Tec. The number of points marked as improvement review points during the training period is counted, and the ratio of the number of improvement review points to the total number of review points is calculated to obtain the improvement review point ratio (Bin). After normalizing the mean of the change ratio of the quality values ​​of the review points and the proportion of the review points improved, the results were entered into the formula. After weighted calculation, the confidence estimate Hraf of the proposed solution is obtained, where , These are the preset weighting factors corresponding to the average change ratio of the review point task excellence value and the percentage of review points improved, respectively; Strategy optimization: Preset the confidence value threshold of the scheme corresponding to the scheme confidence value, remove schemes with confidence values ​​lower than the corresponding threshold and send them to the corresponding medical staff terminal, and rematch the training scheme of the current user's task performance level in the standardized training method library.

[0021] The above formulas are all dimensionless calculations. Dimensionless calculations can be performed using various methods such as standardization, which will not be elaborated here. The formulas are derived from software simulations using a large amount of collected data, and are the closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0022] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, ATA hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state ATA hard disk.

[0023] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0024] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0025] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0026] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0027] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0028] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable ATA hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0029] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A closed-loop brain function training method based on a brain-computer interface system, characterized in that, Includes the following steps: Parameter acquisition: Collect the target user's electroencephalographic parameters and basic clinical parameters; Develop a training program: After comprehensive analysis of EEG physiological data, user syndrome values ​​are obtained. Based on clinical diagnostic labels in clinical baseline parameters and target user age, target user deficits and task scenarios are matched. User training task sets are selected from the standardized task library of daily activities using target user deficits and task scenarios. The task difficulty is determined by comparing syndrome values ​​with ideal syndrome values. Standardized tasks are determined by combining target user task scenarios and matched task difficulty. After the target user performs a standardized task, the completion rate, accuracy, and time taken for the standardized task are statistically analyzed to obtain the task excellence value, and the performance level is determined based on the excellence value. Based on the performance level, a training scheme is matched with the current user's task performance level from a standardized training method library; Training method evaluation: Users carry out brain function training according to the training plan, and the user training data during the training cycle is stored in real time. After the training is completed, the user training data is comprehensively analyzed to obtain the confidence value of the plan. The user training data includes the change ratio of task merit value of improved review points and the proportion of improved review points. Strategy optimization: Preset the confidence value threshold of the scheme corresponding to the scheme confidence value, remove schemes with confidence values ​​lower than the corresponding threshold and send them to the corresponding medical staff terminal, and rematch the training scheme of the current user's task performance level in the standardized training method library.

2. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific process of comprehensively analyzing electroencephalographic parameters is as follows: The electroencephalometric parameters include the mean β wave, the mean α wave, and the prefrontal lobe ratio. A preset time evaluation window is used to extract β-wave data and α-wave data from each time acquisition node. At each time node, the data segments corresponding to the focused state and the data segments corresponding to the relaxed state are collected. The mean of each time node segment is calculated. Then, the arithmetic mean of the mean of β-wave data and the mean of α-wave data from each acquisition node are calculated to obtain the mean of β-wave and the mean of α-wave. Extract the theta wave and beta wave data of the prefrontal cortex of each acquisition node within the time evaluation window. For each acquisition node, first calculate the arithmetic mean of the theta wave energy and the arithmetic mean of the beta wave energy, and then calculate the ratio of the node's theta wave mean to the node's beta wave mean to obtain the prefrontal cortex ratio of each acquisition node. The user syndrome value is obtained by comprehensively processing the mean of the β wave, the mean of the α wave, and the prefrontal ratio.

3. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific process for determining the standardization tasks for target users is as follows: Basic clinical parameters include age and clinical diagnostic labels; Identify the user's clinical diagnosis label, and pre-set each type of diagnosis label corresponds to a defect; after matching the clinical diagnosis label with the defect, filter out the user training task set, where the task library stores the user training tasks corresponding to different defects; The target users are categorized by age into pre-defined age ranges, and each age range corresponds to a specific task scenario. Identify the pre-labeled task difficulty tags for each group of tasks in the user's training task set. The task difficulty tags include both easy and complex. A preset ideal value for the status quo is set. The status quo value is compared with the ideal value. If the status quo value is lower than the ideal value, a simple task is selected from the task set. If the status quo value is higher than the ideal value, a complex task is selected from the task set. The target user's status quo value is matched with the ideal value to determine the task difficulty. Select tasks from the standardized task library for daily activities that can assess the target user's shortcomings, and determine standardized tasks by combining the target user's task scenario and the matching task difficulty.

4. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific calculation logic for the completion rate, accuracy, and time taken for standardized tasks: After the target user executes the matched standardized task, the execution data of the task is statistically analyzed, the steps of the standardized task are determined, the number of steps actually completed by the user is counted, and the ratio of the number of steps actually completed to the total number of steps of the task is calculated to obtain the task completion rate. The total number of operations for the standardized task and the number of operations that the user completed correctly are counted. The ratio of the number of correct operations to the total number of operations is calculated to obtain the task accuracy rate. The start time is the time when the user is informed of the standardized task requirements and the execution is initiated, and the end time is the time when the user indicates that the operation is completed. The operation time is recorded as the task time.

5. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific process for determining the task excellence value is as follows: The task excellence value of the target user performing the standardized task is obtained by comprehensively analyzing the task completion rate, task accuracy, and task time.

6. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific process for determining the matching training scheme is as follows: The task performance level is determined based on the preset task performance value corresponding to each group of task performance value range and the preset mapping rules. Based on different performance levels, a training scheme corresponding to the current user's task performance level is matched in the standardized training method library; the standardized training method library stores each group of training schemes, and each group of training schemes is pre-labeled with the applicable task performance level and defects.

7. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific process of analyzing user training data is as follows: During the training cycle, a task review point is preset. When the task review point is reached, the standardized task is executed again and user training data is collected to calculate the task excellence value. The task excellence value of each review point is calculated. Starting from the user's initial task excellence value, a data sequence is constructed by combining the excellence values ​​of each review point. Two adjacent sets of data are extracted from the data sequence. The ratio of adjacent review points is calculated with the left side of the adjacent data point as the denominator and the right side as the numerator to obtain the change ratio of the task excellence value of the current review point. Calculate the ratio of adjacent review points. If the ratio is greater than 1, mark the corresponding review point as an improved review point. The average change ratio of the task excellence value at each review point is obtained by statistically analyzing the changes in the task excellence value at each review point. The number of points marked as improvement review points during the statistical training period is counted, and the ratio of the number of improvement review points to the total number of review points is calculated to obtain the proportion of improvement review points.

8. The closed-loop brain function training method based on a brain-computer interface system according to claim 1, characterized in that, The specific process for determining the confidence valuation of the proposed solution is as follows: A comprehensive analysis of the average change ratio of the merit value of the review point task and the proportion of the review point improvement is used to obtain the confidence estimate of the solution.

Images