Adaptive cognitive training data evaluation method and system based on VR method

Abstract

The present application relates to the technical field of virtual reality medical rehabilitation, in particular to an adaptive cognitive training data evaluation method and system based on a VR method. Through the fusion of multi-modal data acquisition, adaptive algorithms and intermittent low-power strategies, the present application can accurately evaluate the cognitive state of patients within a limited time. Based on the real-time comparison of the preset threshold and the cognitive bias index, subtle changes such as patient attention dispersion and reaction delay can be quickly captured in a single training, and the task complexity can be automatically adjusted to ensure that the training intensity always adapts to the current state of the patient, thereby achieving the purpose of dynamically adjusting the training difficulty. At the same time, short-term and high-frequency training is supported, which can quickly respond to patient switching needs. By monitoring the training interval time of outpatients, it is determined whether to turn off non-core sensors to achieve circuit load balancing when multiple devices run in parallel. The present application is especially suitable for outpatient high-frequency start-stop scenarios and helps medical institutions achieve double optimization of rehabilitation effect and operating cost.

Description

Technical Field

[0001] This invention relates to the field of virtual reality medical rehabilitation technology, and more specifically, to an adaptive cognitive training data evaluation method and system based on VR methods. Background Technology

[0002] Virtual reality (VR) is a computer-generated simulated environment that allows users to immerse themselves in a fully or partially virtual three-dimensional world through multiple sensory experiences, including sight, hearing, and touch. Users can interact with objects, scenes, or characters in the virtual environment in real time by wearing specialized devices (such as headsets, controllers, and sensors), creating a sense of being there.

[0003] Currently, the application of virtual reality technology in the field of medical rehabilitation is mostly concentrated on intensive training for long-term hospitalized patients. However, its suitability for short-term, high-frequency rehabilitation scenarios for outpatients (single training session lasting 15-20 minutes, used multiple times a day) is significantly insufficient, specifically manifested in the following problems: 1. Rigid training content: The existing system uses fixed difficulty modules, which cannot be dynamically adjusted according to the individual rehabilitation progress of outpatients (such as the speed of attention recovery and memory fluctuations), resulting in a disconnect between training intensity and patient needs; 2. Delayed real-time assessment: Relying on manual observation by rehabilitation therapists or periodic scale assessments makes it difficult to capture subtle cognitive changes (such as delayed response and spatial perception deviation) in real time during a single training session, affecting the dynamic optimization of the treatment plan; 3. Excessive equipment power consumption: VR devices in outpatient settings need to be started and stopped frequently. The existing system lacks an intermittent power consumption strategy, which leads to increased power consumption during idle and increased hardware wear and tear. 4. Sensitive to environmental interference: Environmental factors such as lighting and noise in the examination room can easily interfere with patients' cognitive performance. The existing system does not integrate an environmental parameter correction mechanism, which leads to distorted assessment results.

[0004] In addition, the parallel operation of multiple devices in outpatient settings can easily cause fluctuations in circuit load, and existing technologies do not combine cognitive training with energy management, making it difficult to achieve the dual goals of optimizing rehabilitation effects and operating costs. Summary of the Invention

[0005] The purpose of this invention is to provide an adaptive cognitive training data evaluation method and system based on VR, so as to solve the problems of rigid training modes, low real-time evaluation efficiency and high energy consumption of existing VR systems in outpatient settings as mentioned in the background art.

[0006] To address the aforementioned technical problems, the present invention provides a method for evaluating adaptive cognitive training data based on VR, comprising the following steps: S1. Real-time collection of cognitive training data from outpatients, including electroencephalogram (EEG) signals. Eye-tracking data Reaction time Task completion rate Light intensity and noise level ; S2, according to the formula The cognitive bias index of patients was calculated by combining the collected cognitive training data. ,in, , and These are the weighting coefficients. The baseline EEG signal; S3, Cognitive Bias Index With preset threshold Compare and run the training difficulty adjustment strategy. > This generates a training difficulty adjustment signal, dynamically increasing the complexity of the VR training content. ≤ If so, maintain or reduce the difficulty of training; S4. Generate personalized rehabilitation paths for patients based on historical training data. At the same time, set an intermittent low-power mode according to the frequency of device usage in the outpatient setting. When patients are not training, the mode will automatically switch to standby mode.

[0007] As a further improvement to this technical solution, the cognitive bias index in step S2... Further calculations after enabling eye-tracking include: According to the formula And combined with the collected eye-tracking data Calculate attention concentration ; Updated Cognitive Bias Index for: ,in, This represents the attention weighting coefficient.

[0008] As a further improvement to this technical solution, in step S2, when there is significant light or noise interference in the examination room environment, according to the formula... And combined with the collected light intensity and noise level Corrected Cognitive Bias Index ,in, and As the baseline environmental parameters, and This is a correction factor.

[0009] As a further improvement to this technical solution, the specific operation of the training difficulty adjustment strategy in step S3 is as follows: when > At that time, according to the formula Increase task complexity level ,in, For adjustment coefficients; when ≤ At that time, according to the formula Reduce task complexity level .

[0010] As a further improvement to this technical solution, the specific operation of generating a personalized rehabilitation path for the patient based on historical training data in step S4 is as follows: Extracting the patient's past Cognitive bias index sequence of training Calculate the trend slope ; like A value greater than 0 indicates that the patient's cognitive bias index is generally on the rise and the rehabilitation effect is not as expected, so the training difficulty should be continuously increased. like ≤0 indicates that the patient's cognitive bias index is stable or has decreased, and the rehabilitation effect has reached or improved. In this case, the level is reset to the initial difficulty level.

[0011] As a further improvement to this technical solution, the trend slope The calculation formula is: in, This represents the sequence number of the training iterations. Corresponding to the 1st to the 2nd Training session; For the first Cognitive bias index of the training session; This represents the total number of historical training sessions.

[0012] As a further improvement to this technical solution, the specific operation for implementing the intermittent low-power mode in step S4 is as follows: Monitoring the training interval of outpatients and the training interval With preset threshold Compare; like ≥ If so, non-core sensors will be turned off and the VR device refresh rate will be reduced to the preset minimum value; like < If not, low-power operation will not be triggered.

[0013] To address the aforementioned technical problems, another technical solution provided by this invention is: an adaptive cognitive training data evaluation system based on VR methods, the system comprising a data acquisition module, an intelligent analysis module, an adaptive control module, and an energy consumption optimization module; The data acquisition module is used to collect cognitive training data from outpatients in real time. The intelligent analysis module is used to calculate the cognitive bias index based on the collected cognitive training data, judge the patient's cognitive status through formulaic threshold comparison, and generate quantitative assessment results. The adaptive control module is used to dynamically adjust the complexity of VR training content based on the cognitive bias index. The energy consumption optimization module is used to generate personalized rehabilitation paths for patients based on historical training data. At the same time, it sets an intermittent low-power mode according to the frequency of equipment use in the outpatient scenario, and automatically switches to standby mode when the patient is not training.

[0014] As a further improvement to this technical solution, the data acquisition module is internally equipped with an EEG sensor, an eye tracker, a light intensity sensor, and a noise level sensor. The EEG sensor is used to collect EEG signals from outpatients in real time. The eye tracker is used to collect eye-tracking data from outpatients in real time. The light intensity sensor is used to acquire the light intensity of the clinic in real time. The noise level sensor is used to acquire the noise level of the examination room in real time. .

[0015] As a further improvement to this technical solution, the system further includes a processor, a memory and instructions stored therein, wherein the processor executes the instructions to implement the steps of the above method.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention collects multimodal information such as EEG signals and eye-tracking data in real time, and dynamically adjusts the training difficulty in combination with the cognitive bias index to achieve a personalized rehabilitation plan. This avoids the problem of the intensity being out of touch with the actual needs of patients caused by traditional fixed-difficulty training, and significantly improves rehabilitation efficiency. At the same time, based on the real-time comparison of preset thresholds and the cognitive bias index, it can quickly capture subtle changes such as the patient's attention being distracted and reaction delays in a single training session, and automatically adjust the task complexity to ensure that the training intensity is always adapted to the patient's current state.

[0017] 2. This invention effectively reduces unnecessary energy consumption and extends equipment lifespan by setting up an energy consumption optimization module and adopting an intermittent low-power mode. It also reduces electricity costs for medical institutions, making it particularly suitable for high-frequency start-stop scenarios in outpatient clinics. This helps medical institutions achieve dual goals of optimizing rehabilitation effectiveness and operating costs. It also supports short-term, high-frequency training, can quickly respond to patients' switching needs, and can determine whether to shut down non-core sensors by monitoring the training interval of outpatient patients, thereby achieving circuit load balancing when multiple devices are running in parallel and ensuring system stability.

[0018] 3. This invention integrates a light intensity sensor and a noise level sensor to dynamically correct the calculation results of the cognitive bias index, eliminate the interference of fluctuations in the clinic environment on the assessment results, ensure data accuracy, and avoid misjudgments caused by environmental factors.

[0019] 4. This invention can generate personalized rehabilitation paths by combining historical training data, and analyze the long-term recovery progress of patients through trend slope analysis, providing rehabilitation therapists with objective decision-making basis and avoiding reliance on subjective experience judgment. Attached Figure Description

[0020] Figure 1 This is a schematic diagram illustrating the steps of the adaptive cognitive training data evaluation method based on VR method of the present invention.

[0021] Figure 2 This is a logical block diagram of the adaptive cognitive training data evaluation system based on the VR method of the present invention.

[0022] Figure 3 This is a flowchart illustrating the logic of calculating and adjusting the cognitive bias index in this invention. Detailed Implementation

[0023] 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.

[0024] like Figure 1 and Figure 3 As shown, this invention provides an adaptive cognitive training data evaluation method based on VR, comprising the following steps: The first step is to collect cognitive training data from outpatients in real time, including electroencephalogram (EEG) signals. Eye-tracking data Reaction time Task completion rate Light intensity and noise level This provides multi-dimensional input for subsequent analysis.

[0025] Step 2: According to the formula The cognitive bias index of patients was calculated by combining the collected cognitive training data. ,in, , and These are the weighting coefficients. The reference EEG signal.

[0026] In scenarios requiring precise assessment of a patient's attention span, enabling eye-tracking necessitates additional calculations of attention concentration. At this point, according to the formula And combined with the collected eye-tracking data Calculate attention concentration .

[0027] Then update the cognitive bias index. for: ,in, This represents the attention weighting coefficient.

[0028] When there is significant light or noise interference in the examination room environment, the interference must be eliminated (regardless of whether eye tracking is enabled), according to the formula. And combined with the collected light intensity and noise level Corrected Cognitive Bias Index ,in, and As the baseline environmental parameters, and This is a correction factor.

[0029] Step 3: Cognitive Bias Index With preset threshold Compare and implement training difficulty adjustment strategies.

[0030] like > This generates a training difficulty adjustment signal, dynamically increasing the complexity of the VR training content, specifically according to the formula: Increase task complexity level ,in, This is for adjusting the coefficient.

[0031] like ≤ If the training difficulty is maintained or reduced, then according to the formula: Reduce task complexity level .

[0032] The fourth step is to generate a personalized rehabilitation path for the patient based on historical training data, specifically as follows: Extracting the patient's past Cognitive bias index sequence of training Calculate the trend slope ; like A value greater than 0 indicates that the patient's cognitive bias index is generally on the rise and the rehabilitation effect is not as expected, so the training difficulty should be continuously increased. like ≤0 indicates that the patient's cognitive bias index is stable or has decreased, and the rehabilitation effect has reached or improved. In this case, the level is reset to the initial difficulty level.

[0033] Trend slope The calculation formula is: in, This represents the sequence number of the training iterations. Corresponding to the 1st to the 2nd Training session; For the first Cognitive bias index of the training session; This represents the total number of historical training sessions.

[0034] At the same time, based on the frequency of device usage in outpatient settings, an intermittent low-power mode is set, specifically as follows: Monitoring the training interval of outpatients and the training interval With preset threshold Compare; like ≥ If so, non-core sensors will be turned off and the VR device refresh rate will be reduced to the preset minimum value; like < If not, low-power operation will not be triggered.

[0035] And it automatically switches to standby mode when the patient is not undergoing training.

[0036] like Figure 2 As shown, the present invention also provides an adaptive cognitive training data evaluation system based on VR method, the system including a data acquisition module, an intelligent analysis module, an adaptive control module and an energy consumption optimization module.

[0037] The data acquisition module performs the specific operations described in step one above, and is used to collect cognitive training data from outpatients in real time. The data acquisition module is internally equipped with an electroencephalogram (EEG) sensor, an eye tracker, a light intensity sensor, and a noise level sensor, each used to collect corresponding data.

[0038] The intelligent analysis module can perform the specific operations of step two above. It is used to calculate the cognitive bias index based on the collected cognitive training data, judge the patient's cognitive status through formulaic threshold comparison, and generate quantitative assessment results.

[0039] The adaptive control module can perform the specific operations in step three above, and is used to dynamically adjust the complexity of VR training content according to the cognitive bias index.

[0040] The energy consumption optimization module can perform the specific operations in step four above. It is used to generate personalized rehabilitation paths for patients based on historical training data. At the same time, it sets an intermittent low power consumption mode according to the frequency of device use in the outpatient scenario. When the patient is not training, it automatically switches to standby mode.

[0041] The system also includes a processor, a memory and the instructions stored therein, and the steps of the above method are implemented when the processor executes the instructions.

[0042] The beneficial effects of the present invention will be illustrated below through specific embodiments: Example 1: The complete process of short-term high-frequency training for outpatients.

[0043] Step 1: Data Acquisition and Initialization.

[0044] The data acquisition module is activated, and the patient wears a VR device that integrates EEG sensors, eye trackers, and environmental sensors.

[0045] Real-time data collection: EEG signals : Obtain the patient's frontal lobe area through electrode cap Wave intensity (8-12Hz) =20 benchmark value =20 .

[0046] Eye-tracking data Record the percentage of time spent in the effective fixation area (e.g., the patient fixates on the virtual task target for 80% of the time).

[0047] reaction time The average time for patients to complete a spatial memory task was 3.2 seconds.

[0048] Task completion rate The current task completion rate is 70%.

[0049] Environmental parameters: Light intensity in the examination room =300lx (benchmark) =250lx), noise level =55dB (baseline) =50dB).

[0050] Step 2: Calculation and correction of the cognitive bias index.

[0051] The intelligent analysis module calculates the cognitive bias index based on the formula. : Enable environment correction ( , ): Step 3: Dynamically Adjust Training Difficulty ).

[0052] The adaptive control module will correct the =0.134375 and the preset threshold Comparing with =0.2, because < According to the formula Reduce task complexity level (For example, reducing the number of virtual obstacles).

[0053] Step 4: Energy consumption optimization management.

[0054] Energy consumption optimization module monitors equipment idle time =3 minutes (threshold) =5 minutes), because < To maintain full functionality, the VR refresh rate is kept at 90Hz to ensure rapid access for the next patient.

[0055] In summary, this invention achieves real-time data acquisition. , The task complexity is dynamically adjusted based on the data, avoiding a one-size-fits-all approach to training. Simultaneously, the revised... It eliminates light / noise interference, ensuring accurate assessment, and does not trigger power derating during short periods of idle time, thus guaranteeing device response speed.

[0056] Example 2: Generation and difficulty reset of long-term rehabilitation pathways.

[0057] Step 1: Historical data analysis.

[0058] Extract the cognitive bias index sequence from the patient's past 5 training sessions: .

[0059] Calculate the trend slope : .

[0060] Step 2: Adjustment of the rehabilitation pathway.

[0061] because =-0.036<0, indicating that the patient's cognitive ability is continuously improving, and the system automatically resets the task difficulty to the initial level (such as restoring the basic virtual scene complexity).

[0062] In summary, this invention objectively assesses long-term progress through slope analysis, avoiding reliance on subjective judgment, while also automatically resetting the difficulty level to prevent patients from stagnating due to over-adaptation to the training content.

[0063] Example 3: Parallel operation and energy consumption management of multiple devices in outpatient clinics.

[0064] Scenario Description: A clinic sees 10 patients in the morning, with approximately 8 minutes between each patient's training session. The equipment is idle during this time. =8 minutes (threshold) =5 minutes).

[0065] because > Therefore, the following operations are performed: The power optimization module triggers an intermittent low-power mode: Turn off non-core sensors (such as backup eye trackers); Reduce the VR refresh rate to 60Hz to decrease power consumption; The device power consumption is significantly reduced in standby mode.

[0066] In summary, the intermittent mode of the present invention can significantly reduce idle power consumption, while load balancing when multiple devices are running in parallel can avoid circuit overload.

[0067] Example 4: Practical application of environmental interference correction.

[0068] Scene description: A sudden burst of bright light and noise pollution in the examination room reduces the intensity of light inside the room. =600lx, noise level =70dB, the patient is undergoing spatial memory training. Step 1: Data Acquisition and Environmental Parameter Input.

[0069] The data acquisition module acquires the following data in real time: EEG signals =22 (benchmark value) =20 ), reaction time =4 seconds, task completion rate =60%, Environmental parameters: Light intensity in the examination room =600lx (baseline) =250lx), noise level =70dB (reference) =50dB).

[0070] Step 2: Cognitive Bias Index calculate.

[0071] The intelligent analysis module calculates the basic cognitive bias index. .

[0072] Step 3: Environmental Modification calculate( , ), preset threshold =0.25, . Difficulty adjustment: due to =0.3025> =0.25, according to the formula Increase the task complexity level (e.g., increasing the density of virtual obstacles).

[0073] Before revision =0.25≤threshold =0.25, the system should have maintained the difficulty level. In the uncorrected scenario, the system mistakenly believed that the patient's ability met the standard and maintained the original difficulty level. In reality, the patient's performance declined due to environmental interference. Revised =0.30 accurately reflects the increased cognitive load caused by environmental interference, triggering an upgrade in difficulty, avoiding misjudging the patient's ability due to strong light and noise. In the corrected scenario, the system's ability to recognize the true state of the problem decreases, so the difficulty is increased in time to maintain the effectiveness of training.

[0074] In summary, the revised index of this invention accurately reflects the patient's true ability, avoids misjudging cognitive decline due to environmental interference, and dynamically adjusts to ensure that the training intensity always matches the patient's true level.

[0075] This invention integrates multimodal data acquisition, adaptive algorithms, and intermittent low-power strategies to accurately assess a patient's cognitive state within a limited time, dynamically adjust training difficulty, and simultaneously optimize device energy consumption, thereby improving rehabilitation efficiency and reducing the operating costs of medical institutions.

[0076] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An adaptive cognitive training data evaluation method based on VR, characterized in that, Includes the following steps: S1. Real-time collection of cognitive training data from outpatients, including electroencephalogram (EEG) signals. Eye-tracking data Reaction time Task completion rate Light intensity and noise level ; S2, according to the formula The cognitive bias index of patients was calculated by combining the collected cognitive training data. ,in, , and These are the weighting coefficients. The reference EEG signal; S3, Cognitive Bias Index With preset threshold Compare and implement training difficulty adjustment strategies, if > This generates a signal to adjust and increase the training difficulty, dynamically increasing the complexity of the VR training content. ≤ If so, maintain or reduce the difficulty of training; S4. Generate personalized rehabilitation paths for patients based on historical training data. At the same time, based on the frequency of equipment use in outpatient settings, the energy consumption optimization module sets an intermittent low-power mode. When patients are not training, the VR device automatically switches to standby mode.

2. The adaptive cognitive training data evaluation method based on VR method according to claim 1, characterized in that, The cognitive bias index in step S2 Further calculations after enabling eye-tracking include: According to the formula And combined with the collected eye-tracking data Calculate attention concentration ; Updated Cognitive Bias Index for: ,in, These are the weighting coefficients.

3. The adaptive cognitive training data evaluation method based on VR method according to claim 1, characterized in that, In step S2, when there is significant light or noise interference in the examination room environment, according to the formula... And combined with the collected light intensity and noise level Corrected Cognitive Bias Index ,in, and As the baseline environmental parameters, and This is a correction factor.

4. The adaptive cognitive training data evaluation method based on VR method according to claim 1, characterized in that, The specific operation of the training difficulty adjustment strategy in step S3 is as follows: when > At that time, according to the formula Increase task complexity level ,in, For adjustment coefficients; when ≤ At that time, according to the formula Reduce task complexity level .

5. The adaptive cognitive training data evaluation method based on VR method according to claim 1, characterized in that, The specific operation of generating a personalized rehabilitation path for the patient based on historical training data in step S4 is as follows: Extracting the patient's past Cognitive bias index sequence of training Calculate the trend slope ; like A value greater than 0 indicates that the patient's cognitive bias index is generally on the rise and the rehabilitation effect is not as expected, so the training difficulty should be continuously increased. like ≤0 indicates that the patient's cognitive bias index is stable or has decreased, and the rehabilitation effect has reached or improved. In this case, the level is reset to the initial difficulty level.

6. The adaptive cognitive training data evaluation method based on VR method according to claim 5, characterized in that, The slope of the trend The calculation formula is: in, This represents the sequence number of the training iterations. Corresponding to the 1st to the 2nd Training session; For the first Cognitive bias index of the training session; This represents the total number of historical training sessions.

7. The adaptive cognitive training data evaluation method based on VR method according to claim 1, characterized in that, The specific operation for implementing the intermittent low-power mode in step S4 is as follows: Monitoring the training interval of outpatients and the training interval With preset threshold Compare; like ≥ If so, non-core sensors will be turned off and the VR device refresh rate will be reduced to the preset minimum value; like < If not, low-power operation will not be triggered.

Images