Cognitive ability training method and system based on virtual reality technology

Abstract

The invention provides a cognitive ability training method and system based on a virtual reality technology, and the method comprises the steps: obtaining multi-data-dimension feature data used for evaluating the cognitive ability of a subject in the training process of each round; wherein the feature data comprises objective data collected by a sensor during training and subjective data counted after training; scoring the cognitive competence of the subject in each data dimension according to a preset rule, and performing weighted fusion of scores according to preset weight parameters for different data dimensions to obtain a comprehensive score; and taking the score of each data dimension and / or the comprehensive score as input, and utilizing a pre-trained artificial intelligence model to adaptively adjust the difficulty level suitable for the tested training task. According to the invention, the difficulty level suitable for the tested training task can be adaptively adjusted by using the pre-trained artificial intelligence model.

Description

Technical Field

[0001] The present invention relates to the technical field of personnel cognitive ability training, and in particular to a cognitive ability training method and system based on virtual reality technology. Background Art

[0002] Cognitive ability refers to the ability of an individual's brain to receive, process and process information, such as attention, memory, thinking, reaction and self-control. Cognitive ability has a great impact on people's life, study and work. Therefore, it is of great value and significance to conduct a comprehensive assessment of an individual's cognitive ability and to carry out targeted intervention training on the individual's cognitive ability.

[0003] However, existing cognitive ability training usually relies on PCs or mobile devices and is conducted in real environments. This method often lacks immersion and interactivity, and is still not ideal in terms of training efficiency and effectiveness, which in turn limits the training effect.

[0004] To this end, how to provide a cognitive ability assessment method and training environment with a stronger sense of immersion and interactivity is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a cognitive ability training method and system based on virtual reality technology to eliminate or improve one or more defects in the prior art.

[0006] One aspect of the present invention provides a cognitive ability training method based on virtual reality technology, which uses virtual reality technology to present each round of training tasks to subjects, and the training tasks are pre-divided into difficulty levels. The method includes the following steps: in each round of training, obtaining feature data of multiple data dimensions for evaluating the cognitive ability of the subject; wherein the feature data includes objective data collected by sensors during training and subjective data statistically collected after training; scoring the cognitive ability of the subject in each data dimension according to preset rules, and performing weighted fusion of the scores according to preset weight parameters for different data dimensions to obtain a comprehensive score; using the scores of each data dimension and / or the comprehensive score as input, using a pre-trained artificial intelligence model to adaptively adjust the difficulty level of the training task suitable for the subject.

[0007] In some embodiments of the present invention, the objective data includes multi-dimensional physiological data of the subject and multi-dimensional environmental data during training; wherein the dimensions of the physiological data include at least two of heart rate, heart rate variability, blood pressure fluctuations, galvanic skin response, eye tracking and EEG fluctuations; and the dimensions of the environmental data include at least two of light intensity, sound intensity and temperature data.

[0008] In some embodiments of the present invention, the subjective data includes task execution quality results evaluated according to preset rules and autonomous assessment results of the subjects; wherein the task execution quality results include reaction time and / or accuracy, and the autonomous assessment results include scales and / or questionnaires.

[0009] In some embodiments of the present invention, before the step of scoring the cognitive ability of the subject in each data dimension according to preset rules, the method also includes: performing data preprocessing and standardization on the feature data of multiple data dimensions; wherein the data preprocessing step includes multiple data cleaning, denoising and missing value processing; the standardization processing step includes using standardization technology to convert subjective data and objective data of different dimensions and magnitudes to the same scale, and the types of standardization technology include Z-score standardization, T-score standardization, percentile standardization and Min-Max standardization.

[0010] In some embodiments of the present invention, the type of the artificial intelligence model is a prediction model, and the method also includes a pre-training step for the artificial intelligence model, including: generating a training sample set; pre-training the artificial intelligence model based on the training sample set to obtain a pre-trained artificial intelligence model.

[0011] In some embodiments of the present invention, the sample data in the training sample set includes scores for each data dimension, a comprehensive score and an applicable difficulty level, and the scores for each data dimension and the comprehensive score are input into the artificial intelligence model to obtain an applicable difficulty level; or the sample data in the training sample set includes a comprehensive score and an applicable difficulty level, and the comprehensive score is input into the artificial intelligence model to obtain an applicable difficulty level.

[0012] In some embodiments of the present invention, the training tasks are pre-divided into difficulty levels in different data dimensions; the training sample set includes scores for each data dimension and difficulty levels for each data dimension of the training tasks applicable to the subjects.

[0013] In some embodiments of the present invention, the types of prediction models include classification models, regression models and clustering models; the method also includes: in the process of pre-training the artificial intelligence model based on the training sample set, the prediction ability of the model is improved by training through feature selection, feature construction and feature conversion, and the model parameters are optimized through one or more methods of grid search, random search and Bayesian optimization.

[0014] In some embodiments of the present invention, in the process of pre-training the artificial intelligence model based on the training sample set, the method also includes: verifying the performance of the artificial intelligence model through one or more of cross-validation, confusion matrix and performance indicator detection, wherein the performance indicators include one or more of accuracy, recall rate and F1 score.

[0015] In some embodiments of the present invention, the method further includes: statistically calculating the scores of various data dimensions and the comprehensive scores of the subjects within a historical period of preset length to construct a scoring matrix; presetting the weights of the scoring matrix in a pattern in which the weights increase incrementally over time; weighting the scoring matrix according to the preset weights of the scoring matrix; associating the weighted scoring matrix with the difficulty level of the training tasks suitable for the subjects, and recommending the difficulty level of the next round of training tasks for the subjects based on the pre-stored scoring data of the subjects within the historical period of preset length.

[0016] Another aspect of the present invention provides an electronic device, including a processor, a memory, and a computer program / instructions stored in the memory, wherein the processor is used to execute the computer program / instructions, and when the computer program / instructions are executed, the electronic device implements the steps of the method described in any one of the above embodiments.

[0017] Another aspect of the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the above embodiments.

[0018] Another aspect of the present invention provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the method described in any one of the above embodiments.

[0019] Another aspect of the present invention provides a cognitive ability training system based on virtual reality technology, the system comprising: an electronic device, a computer-readable storage medium and a computer program product for implementing the method described in any one of claims 1 to 10; a plurality of sensors for collecting objective data of multiple data dimensions for evaluating the cognitive ability of the subject; and an interactive interface for receiving subjective data of multiple data dimensions for evaluating the cognitive ability of the subject that is statistically analyzed after training.

[0020] The cognitive ability training method and system based on virtual reality technology of the present invention can integrate subjective data and objective data to obtain a comprehensive scoring result, and use the comprehensive score as a reference to adaptively adjust the difficulty level of the training task suitable for the subject using a pre-trained artificial intelligence model.

[0021] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0022] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of the present application, and do not constitute a limitation of the present invention. In the drawings:

[0024] Figure 1 This is a flow chart of a cognitive ability training method based on virtual reality technology in one embodiment of the present invention.

[0025] Figure 2 This is a flow chart of a cognitive ability training method based on virtual reality technology in another embodiment of the present invention.

[0026] Figure 3 This is a flowchart of the pre-training of an artificial intelligence model in one embodiment of the present invention.

[0027] Figure 4 FIG. 1 is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0030] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0031] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection with an intermediate.

[0032] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0033] In order to overcome the problems existing in the prior art, the present invention proposes a cognitive ability training method based on virtual reality technology, which can provide an immersive environment and an interesting interactive environment based on virtual reality technology, evaluate the cognitive ability of individuals by fusion of subjective and objective data, and adjust the difficulty of training tasks according to the evaluation results through AI adaptive adjustment algorithm to better adapt to the actual situation of the subjects and improve the training effect.

[0034] Figure 1 The flowchart of the cognitive ability training method based on virtual reality technology in one embodiment of the present invention. In the method, virtual reality technology is used to present each round of training tasks to the subjects, and the training tasks are pre-divided into difficulty levels. The method includes the following steps:

[0035] Step S110: During each round of training, feature data of multiple data dimensions for evaluating the cognitive ability of the subject is obtained; wherein the feature data includes objective data collected by sensors during training and subjective data collected after training.

[0036] In the specific implementation process, the objective data may include multi-dimensional physiological data of the subject and multi-dimensional environmental data during training. The dimensions of physiological data may include at least two of heart rate, heart rate variability, blood pressure fluctuations, galvanic skin response, eye tracking, and EEG fluctuations; the dimensions of environmental data may include at least two of light intensity, sound intensity, and temperature data.

[0037] In the specific implementation process, the subjective data includes the task execution quality results evaluated according to the preset rules and the self-assessment results of the subjects. Among them, in a specific embodiment of the present invention, the task execution quality results include reaction time and / or accuracy, and the self-assessment results include scales and / or questionnaires.

[0038] Step S120: Score the cognitive ability of the subject in each data dimension according to preset rules, and perform weighted fusion of the scores according to preset weight parameters for different data dimensions to obtain a comprehensive score.

[0039] Step S130: Using the scores and / or comprehensive scores of each data dimension as input, the pre-trained artificial intelligence model is used to adaptively adjust the difficulty level of the training task suitable for the subject.

[0040] The cognitive ability training method and system based on virtual reality technology proposed in the present invention can integrate subjective data and objective data to obtain a comprehensive scoring result. With the comprehensive score as a reference, the difficulty level of the training task suitable for the subject can be adaptively adjusted using a pre-trained artificial intelligence model.

[0041] In some embodiments of the present invention, before the step of scoring the cognitive ability of the subject in each data dimension according to preset rules in step S120, the method further includes: performing data preprocessing and standardization on the feature data of multiple data dimensions.

[0042] Among them, the data preprocessing step may include multiple data cleaning, denoising and missing value processing; the standardization processing step includes using standardization technology to convert subjective data and objective data of different dimensions and magnitudes to the same scale, and the types of standardization technology include Z-score standardization, T-score standardization, percentile standardization and Min-Max standardization.

[0043] By adopting the embodiment of the invention, the quality of data used for model training and actual testing can be improved.

[0044] In some embodiments of the present invention, the type of the artificial intelligence model is a prediction model, and the method also includes a pre-training step for the artificial intelligence model, including: (1) generating a training sample set; (2) pre-training the artificial intelligence model based on the training sample set to obtain a pre-trained artificial intelligence model.

[0045] Specifically, the sample data in the training sample set may include scores for each data dimension, a comprehensive score and an applicable difficulty level, and the scores for each data dimension and the comprehensive score are input into the artificial intelligence model to obtain an applicable difficulty level. Alternatively, the sample data in the training sample set may include a comprehensive score and an applicable difficulty level, and the comprehensive score can be input into the artificial intelligence model to obtain an applicable difficulty level.

[0046] By adopting this embodiment of the invention, a pre-trained prediction model can be used to predict the difficulty level of the next round of training tasks presented to the subjects based on the comprehensive score.

[0047] In some embodiments of the present invention, the training tasks can be further pre-divided into difficulty levels in different data dimensions, or in other words, training tasks of different difficulty levels can be generated in different data dimensions for selection. Accordingly, the training sample set includes the scores of each data dimension and the difficulty level of each data dimension of the training task applicable to the subject, and the scores of each data dimension are input into the artificial intelligence model to obtain the difficulty level of each data dimension of the training task applicable to the subject.

[0048] In some embodiments of the present invention, the types of prediction models include classification models, regression models, and clustering models. The method also includes: in the process of pre-training the artificial intelligence model based on the training sample set, the prediction ability of the model is improved by training through feature selection, feature construction, and feature conversion, and the model parameters are optimized by one or more methods of grid search, random search, and Bayesian optimization.

[0049] By adopting this embodiment of the invention, the granularity of adjusting the difficulty of training tasks can be further refined.

[0050] Furthermore, in some embodiments of the present invention, in the process of pre-training the artificial intelligence model based on the training sample set, the method also includes: verifying the performance of the artificial intelligence model through one or more of cross-validation, confusion matrix and performance indicator detection, wherein the performance indicators include one or more of accuracy, recall rate and F1 score.

[0051] By adopting the embodiment of the invention, the final effect of model training can be improved.

[0052] In some embodiments of the present invention, the method further includes: (1) statistically calculating the scores of each data dimension and the comprehensive score of the subjects within a historical period of preset length to construct a scoring matrix; (2) presetting the weights of the scoring matrix in a mode in which the weights increase incrementally over time; (3) weighting the scoring matrix according to the preset weights of the scoring matrix; (4) associating the weighted scoring matrix with the difficulty level of the training tasks suitable for the subjects, and recommending the difficulty level of the training tasks for the subjects in the next round based on the pre-stored scoring data of the subjects within the historical period of preset length.

[0053] Figure 2 This is a flow chart of a cognitive ability training method based on virtual reality technology in another embodiment of the present invention. The method comprises the following steps:

[0054] Step S210: providing an immersive training environment for users through virtual reality equipment.

[0055] Among them, virtual reality scene design includes: creating diverse virtual environments, such as space stations, deep-sea exploration, etc., to enhance user immersion.

[0056] Step S220: using the data collection module to collect the user's multi-dimensional subjective data and objective data.

[0057] Among them, subjective data include the accuracy rate and reaction time of experimental tasks. Objective data include physiological data and environmental parameters. Physiological parameters include the status of personnel in the recognition task, such as fatigue and load; eye movement data. Subjective data collection includes: (1) User performance in the training task, such as the accuracy rate and completion time of the task. (2) User self-assessment, including the perception of training difficulty, experience feedback during the training process (difficulty, scene satisfaction), etc. Objective data collection includes: (1) Physiological indicators, such as heart rate, skin galvanic response, eye tracking data, etc., which can reflect the user's physiological state and concentration during the training process. For example: ① Heart rate variability HRV: reflects the user's stress level and concentration. ② Eye tracking data: records the path, dwell time and gaze point of the user's line of sight, and analyzes attention distribution. (2) Environmental parameters, such as light intensity and sound intensity during training, which may affect the user's performance.

[0058] Step S230: The subjective data and the objective data are integrated and processed to obtain a comprehensive score.

[0059] The steps of data fusion processing include: (1) Data preprocessing: including data cleaning, denoising, missing value processing, etc. (2) Normalization / standardization: convert subjective and objective data of different dimensions and magnitudes to the same scale for comparison and fusion. Z-score standardization, T-score standardization, percentile standardization or Min-Max standardization can be used according to needs. (3) Rating calculation: score calculation for each data dimension according to the preset scoring criteria. (4) Weighted fusion: assign different weights according to the importance of each data dimension, and then perform weighted summation to obtain a comprehensive score.

[0060] Step S240: Using the pre-trained artificial intelligence model to obtain the difficulty level of the training task suitable for the subject. Specifically, the AI ​​algorithm can be used to analyze the scoring results and adaptively adjust the difficulty of the next training task.

[0061] Among them, there are at least two mechanisms for adjusting task difficulty: (1) Real-time difficulty adjustment: adjusting the difficulty of subsequent tasks in real time based on the user's performance in the current task. (2) Personalized recommendation system: using a recommendation algorithm to recommend suitable training tasks to users based on their historical data and performance.

[0062] Step S250: modifying the difficulty level of the training task presented to the subject. Specifically, the task generation module can be used to generate the training task according to the user's ability and training goal.

[0063] Figure 3 The following is a flow chart of pre-training of an artificial intelligence model in one embodiment of the present invention. The training process includes the following steps:

[0064] Step S310: constructing a training set, including data preprocessing and feature engineering, presetting scoring rules, and manually assessing (labeling) difficulty level labels.

[0065] Data preprocessing and feature engineering include but are not limited to: (1) Data cleaning: removing or filling missing values, processing outliers or erroneous data. (2) Feature selection: identifying the features most relevant to the training task, reducing feature dimensions, and improving model performance. (3) Feature construction: generating new features based on existing data, such as calculating moving averages or rates of change from time series data. (4) Feature transformation: applying methods such as polynomial expansion and logarithmic transformation to improve the learning ability of the model.

[0066] The specific steps of training set construction include: (1) Multi-source data integration: Integrate data from different users and different training scenarios to form a diversified training set. (2) Difficulty stratification: According to the user's baseline ability, the training set is divided into different difficulty levels to meet the needs of personalized training. This can be achieved by manually labeling the difficulty level labels.

[0067] Among them, feature engineering includes feature selection, feature construction and feature transformation to enhance the predictive ability of the model.

[0068] Step S320: Select an initial model and perform training.

[0069] Among them, model selection includes selecting the appropriate model architecture according to the task type (classification, regression, clustering), such as deep learning models such as decision trees, support vector machines, convolutional neural networks (CNNs) or recurrent neural networks (RNNs).

[0070] During model training, you can choose to perform hyperparameter tuning, which includes using grid search, random search, or Bayesian optimization to find the optimal model parameters.

[0071] Step S330: Application of test set and refinement of result output. This step includes but is not limited to: (1) Cross-validation: Use K-fold cross-validation to evaluate the stability and generalization ability of the model. (2) Confusion matrix: Analyze the performance of the model on different categories and identify the strengths and weaknesses of the model. (3) Performance indicators: Use indicators such as accuracy, recall, and F1 score to evaluate the accuracy of model prediction.

[0072] Step S340: Model deployment and application. This step includes but is not limited to: (1) Model deployment: deploying the trained model to the production environment to provide prediction services in real time or batch processing. (2) Online learning: after the model is deployed, online learning or model update is performed based on the newly collected data to adapt to changes in data distribution.

[0073] Step S350: Model monitoring and maintenance. This step includes but is not limited to: (1) Performance monitoring: Continuously monitor the prediction performance of the model to ensure its stability and accuracy.

[0074] (2) Data drift detection: monitor the distribution changes of input data and adjust the model in time to cope with data drift. Optionally, the content of the training task can be: (1) Memory-memory recognition task: the user memorizes pictures of multiple objects before the task starts, and then searches and finds these objects in the scene to interact. The system records the task completion time and accuracy, and can adjust the ratio of target objects to non-target objects and the maximum task completion time. (2) Attention-Schulte grid task: the user needs to click the numbers on the n*n squares in sequence. The system records the completion time and can adjust the value of n and the maximum task completion time. (3) Spatial perception-mental rotation task: the user needs to judge whether two cubes with different orientations are the same. The user can observe the cube from different angles. The system records the accuracy and reaction time, and can set the rotation angle, shape, and the same and different ratio of the cube. The present invention is not limited to this, and the above training tasks are only examples.

[0075] Optionally, innovations can be made in the form of training tasks, including but not limited to: (1) Virtual reality scene design: creating a variety of virtual environments, such as space stations and deep-sea exploration, to enhance the user's sense of immersion. (2) Interactive task design: designing tasks that require users to perform physical operations or make decisions, such as object manipulation or path planning in virtual reality.

[0076] As mentioned above, the cognitive ability training method and system based on virtual reality technology proposed in the present invention can fuse subjective data and objective data to obtain a comprehensive scoring result, and use the pre-trained artificial intelligence model to adaptively adjust the difficulty level of the training task applicable to the subject with the comprehensive score as a reference. Compared with the prior art, the system and method of the present invention can provide a more immersive and interactive training environment, effectively improve the training effect through the fusion of subjective and objective data and the adaptive adjustment of AI algorithms, and are suitable for rehabilitation training for cognitive impairment and personnel cognitive ability improvement training, such as aerospace, transportation human factors, sports, military industry and national defense and other fields.

[0077] The key points of the present invention are:

[0078] (1) Use subjective and objective data fusion technology to evaluate the individual's cognitive ability. This enables objective evaluation of the individual's performance in different data dimensions in the training task and comprehensive evaluation results based on multiple dimensions.

[0079] (2) Use the pre-trained AI model to adaptively adjust the difficulty of the training task. By adaptively adjusting the difficulty, the training task can be more closely aligned with the actual performance of the individual subject, neither too easy nor too difficult.

[0080] (3) Adjust the difficulty of different dimensions of the training task according to the subjective data or objective data of different dimensions.

[0081] Corresponding to the above method, the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the electronic device implements the steps of the above method. This electronic device is the computing and processing core of the cognitive ability training system of virtual reality technology.

[0082] See also Figure 4 The electronic device 00 includes: a processor 01, a memory 02, and a computer program stored in the memory 02 and executable on the processor 01. When the processor 01 executes the computer program, the human factor data server access control method provided in the above method embodiment is implemented.

[0083] Among them, the processor 01 is connected to the memory 02, such as through a bus 03. The processor 01 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in combination with the contents disclosed in this application. The processor 01 can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of DSP and microprocessors, etc. The bus 03 may include a path to transmit information between the above components. The bus 03 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 130 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4Only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The memory 02 is used to store a computer program corresponding to the human factor data server access control method of the above embodiment of the present application, and the computer program is controlled and executed by the processor 01. The processor 01 is used to execute the computer program stored in the memory 02 to implement the content shown in the above method embodiment.

[0084] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned edge computing server deployment method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0085] In one embodiment of the present invention, a computer program product is further provided, including a computer program / instruction, which implements the contents shown in the above-mentioned embodiment when executed by a processor.

[0086] Corresponding to the above method, the present invention also provides a cognitive ability training system based on virtual reality technology, the system comprising: an electronic device, a computer-readable storage medium and a computer program product for implementing the above embodiments; a plurality of sensors for collecting objective data of multiple data dimensions for evaluating the cognitive ability of the subject; and an interactive interface for receiving subjective data of multiple data dimensions for evaluating the cognitive ability of the subject that is statistically analyzed after training.

[0087] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0088] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0089] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0090] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A cognitive ability training method based on virtual reality technology, characterized in that: Using virtual reality technology to present each round of training tasks to the subject, the training tasks are pre-divided into difficulty levels, and the method includes: During each round of training, feature data of multiple data dimensions for evaluating the cognitive ability of the subject are obtained; wherein the feature data includes objective data collected by sensors during training and subjective data collected after training; The cognitive ability of the subject is scored in each data dimension according to the preset rules, and the scores are weighted and integrated according to the preset weight parameters for different data dimensions to obtain a comprehensive score; Using the scores and / or comprehensive scores of various data dimensions as input, the pre-trained artificial intelligence model is used to adaptively adjust the difficulty level of the training task suitable for the subject.

2. The method according to claim 1, characterized in that: The objective data includes multi-dimensional physiological data of the subject and multi-dimensional environmental data during training; wherein the dimensions of the physiological data include at least two of heart rate, heart rate variability, blood pressure fluctuations, galvanic skin response, eye tracking and EEG fluctuations; and the dimensions of the environmental data include at least two of light intensity, sound intensity and temperature data.

3. The method according to claim 1, characterized in that The subjective data include task execution quality results evaluated according to preset rules and autonomous assessment results of the subjects; wherein the task execution quality results include reaction time and / or accuracy, and the autonomous assessment results include scales and / or questionnaires.

4. The method according to claim 1, characterized in that: Before the step of scoring the cognitive ability of the subject in each data dimension according to preset rules, the method further includes: Perform data preprocessing and standardization on feature data of multiple data dimensions; Among them, the data preprocessing step includes multiple data cleaning, denoising and missing value processing; the standardization processing step includes using standardization technology to convert subjective data and objective data of different dimensions and magnitudes to the same scale, and the types of standardization technology include Z-score standardization, T-score standardization, percentile standardization and Min-Max standardization.

5. The method according to claim 1, characterized in that The type of the artificial intelligence model is a prediction model. The method further includes a step of pre-training the artificial intelligence model, including: Generate a training sample set; The artificial intelligence model is pre-trained based on the training sample set to obtain a pre-trained artificial intelligence model.

6. The method according to claim 5, characterized in that The sample data in the training sample set includes scores of various data dimensions, comprehensive scores and applicable difficulty levels, and the scores of various data dimensions and comprehensive scores are input into the artificial intelligence model to obtain applicable difficulty levels; or The sample data in the training sample set includes a comprehensive score and an applicable difficulty level. The comprehensive score is input into the artificial intelligence model to obtain an applicable difficulty level.

7. The method according to claim 5, characterized in that The training tasks are pre-divided into difficulty levels in different data dimensions; The training sample set includes scores for each data dimension and difficulty levels for each data dimension of a training task suitable for the subject.

8. The method according to claim 5, characterized in that The types of prediction models include classification models, regression models and clustering models; The method also includes: in the process of pre-training the artificial intelligence model based on the training sample set, the prediction ability of the model is improved by training through feature selection, feature construction and feature conversion, and the model parameters are optimized through one or more methods of grid search, random search and Bayesian optimization.

9. The method according to claim 8, characterized in that In the process of pre-training the artificial intelligence model based on the training sample set, the method also includes: verifying the performance of the artificial intelligence model through one or more of cross-validation, confusion matrix and performance indicator detection, wherein the performance indicator includes one or more of accuracy, recall rate and F1 score.

10. The method according to claim 1, characterized in that The method further comprises: Count the scores of each data dimension and comprehensive score of the subjects in the historical period of preset length, and construct a score matrix; Preset the weights of the scoring matrix in a pattern where the weights increase incrementally over time; Weighting the scoring matrix according to the preset weight of the scoring matrix; The weighted scoring matrix is ​​associated with the difficulty level of the training task suitable for the subject, and the difficulty level of the training task for the subject in the next round is recommended based on the scoring data of the subject within a historical period of preset length stored in advance.

11. An electronic device comprising a processor, a memory and a computer program / instruction stored in the memory, characterized in that: The processor is used to execute the computer program / instructions. When the computer program / instructions are executed, the electronic device implements the steps of the method according to any one of claims 1 to 10.

12. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

14. A cognitive ability training system based on virtual reality technology, characterized in that: The system comprises: An electronic device, a computer-readable storage medium, and a computer program product for implementing the method according to any one of claims 1 to 10; A variety of sensors are used to collect objective data in multiple data dimensions to evaluate the cognitive abilities of the subjects; The interactive interface is used to receive the subjective data of multiple data dimensions for evaluating the cognitive ability of the subjects after training.

Images