Method and system for collaborative regulation of concentration and fatigue based on electroencephalogram dual-modal feedback

Abstract

The present invention discloses a method and system for collaborative regulation of concentration and fatigue based on electroencephalogram (EEG) bimodal feedback. Before the training starts, the first EEG data of the trainees in the resting state is collected, and the mean value of the ratio of theta waves to alpha waves is calculated as the resting state baseline value of fatigue. After the training starts, the second EEG data of the trainees is collected. The concentration index is calculated according to the ratio of beta waves to the sum of beta waves, theta waves and alpha waves, and the fatigue index is calculated according to the ratio of theta waves to alpha waves. The dynamic threshold is calculated according to the resting state baseline value of fatigue. Corresponding risk prompts are generated according to whether the concentration index and the fatigue index deviate from their thresholds. The present invention breaks through the limitations of traditional single-dimensional biofeedback, significantly improves the safety and efficiency of complex task training, breaks through the limitations of traditional single-dimensional biofeedback, and provides a quantitative neurocognitive enhancement scheme for special operation training.

Description

Technical Field

[0001] The present invention relates to the technical field of safety training, and in particular to a method and system for collaborative regulation of concentration-fatigue based on electroencephalogram (EEG) bimodal feedback. Background Art

[0002] In safety training in high-risk industries (such as power, chemical, aviation, etc.), the attention and fatigue state of trainees directly affect the training effect and operation safety. Traditional training methods mainly rely on empirical prompts and post-event feedback, and it is difficult to monitor the concentration and fatigue degree of trainees in real time, resulting in poor training effects and trainees being prone to making mistakes in actual operations. In addition, in the prior art, the method of monitoring fatigue or attention alone has a high misjudgment rate. For example, the increase in theta waves during fatigue may be misinterpreted as a decrease in concentration, resulting in inaccurate feedback. At the same time, there are individual differences among trainees, which also lead to a high misjudgment rate. Summary of the Invention

[0003] Object of the Invention: The object of the present invention is to provide a method and system for collaborative regulation of concentration-fatigue based on EEG bimodal feedback, so as to solve the problems of single monitoring dimension and high misjudgment rate in the prior art.

[0004] Technical Solution: The method for collaborative regulation of concentration-fatigue based on EEG bimodal feedback according to the present invention includes the following steps:

[0005] Before the training starts, collect the first EEG data of the trainee in the resting state, and calculate the mean value of the ratio of theta waves to alpha waves as the baseline value of fatigue degree in the resting state. ;

[0006] After the training starts, collect the second EEG data of the trainee, calculate the concentration index according to the ratio of beta waves to the sum of beta waves, theta waves and alpha waves, and calculate the fatigue degree index according to the ratio of theta waves to alpha waves.

[0007] According to the baseline value of fatigue degree in the resting state Calculate the dynamic fatigue degree threshold;

[0008] Set the static threshold of concentration;

[0009] Generate corresponding risk warnings according to whether the concentration index deviates from the static threshold of concentration and whether the fatigue degree index deviates from the dynamic fatigue degree threshold.

[0010] Further, the calculation of the dynamic fatigue degree threshold according to the baseline value of fatigue degree in the resting state includes:

[0011] The dynamic fatigue degree threshold ;

[0012] wherein, is the trend change value of the ratio of theta waves to alpha waves within the sliding window, , where N is the length of the sliding window.

[0013] Further, when the training start time is less than the time of one sliding window, the value is taken as 0.

[0014] Further, generating corresponding risk warnings according to whether the concentration index and the fatigue index exceed their corresponding thresholds includes:

[0015] When the concentration index is less than its threshold and the fatigue index is greater than its threshold, trigger the highest intervention;

[0016] When the concentration index is not less than its threshold or the fatigue index is greater than its threshold, trigger the intermediate intervention;

[0017] When the concentration index is not less than its threshold and the fatigue index is not greater than its threshold, trigger the low-intensity feedback.

[0018] Further, when the fatigue index deviates from its corresponding threshold by ±5% in three consecutive cycles, trigger the automatic recalibration process; the triggering of the automatic recalibration process includes:

[0019] Resample the first electroencephalogram data in the resting state, calculate and update the resting state reference value of the fatigue degree , and based on the updated resting state reference value of the fatigue degree recalculate the dynamic fatigue threshold.

[0020] The concentration-fatigue collaborative regulation system based on electroencephalogram bimodal feedback according to the present invention includes:

[0021] A resting state baseline calculation unit, configured to collect the first electroencephalogram data of the trainee in the resting state before the start of the training, and calculate the mean value of the ratio of theta waves to alpha waves as the resting state reference value of the fatigue degree ;

[0022] A concentration-fatigue collaborative regulation unit, configured to, after the start of the training, collect the second electroencephalogram data of the trainee, calculate the concentration index according to the ratio of beta waves to the sum of beta waves, theta waves and alpha waves, and calculate the fatigue index according to the ratio of theta waves to alpha waves;

[0023] Calculate the dynamic fatigue threshold according to the resting state reference value of the fatigue degree ;

[0024] Set the static concentration threshold;

[0025] Generate corresponding risk warnings according to whether the concentration index deviates from the static concentration threshold and whether the fatigue index deviates from the dynamic fatigue threshold.

[0026] Further, in the concentration-fatigue co-regulation unit, the static fatigue reference value according to the fatigue degree Calculating the dynamic fatigue threshold includes:

[0027] Dynamic fatigue threshold ;

[0028] Wherein, is the trend change value of the ratio of theta wave to alpha wave within the sliding window, , and N is the length of the sliding window.

[0029] Further, in the concentration-fatigue co-regulation unit, generating corresponding risk warnings according to whether the concentration index and the fatigue index exceed their corresponding thresholds includes:

[0030] When the concentration index is less than its threshold and the fatigue index is greater than its threshold, trigger the highest intervention;

[0031] When the concentration index is not less than its threshold or the fatigue index is greater than its threshold, trigger the intermediate intervention;

[0032] When the concentration index is not less than its threshold and the fatigue index is not greater than its threshold, trigger the low-intensity feedback;

[0033] In the concentration-fatigue co-regulation unit, when the fatigue index deviates from its corresponding threshold by ±5% in three consecutive cycles, trigger the automatic recalibration process; the triggering of the automatic recalibration process includes:

[0034] Resample the first electroencephalogram data in the resting state, calculate and update the static fatigue reference value of the fatigue degree , and recalculate the dynamic fatigue threshold according to the updated static fatigue reference value of the fatigue degree

[0035] The electronic device according to the present invention includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the computer program is loaded into the processor, the concentration-fatigue co-regulation method based on electroencephalogram bimodal feedback described above is implemented.

[0036] The computer-readable storage medium according to the present invention stores a computer program. When the computer program is executed by a processor, the concentration-fatigue co-regulation method based on electroencephalogram bimodal feedback described above is implemented.

[0037] ​Beneficial effects: Compared with the prior art, the advantages of the present invention are as follows: The present invention integrates the dual-modal physiological indexes of prefrontal β waves and θ / α ratio to realize the real-time monitoring and dynamic intervention of the operator's concentration and fatigue state during high-risk operation training; combined with the preset safety threshold, a three-level intervention strategy (voice warning, sound and light stimulation, operation simplification) is generated, significantly improving the safety and efficiency of complex task training. Brief Description of the Drawings

[0038] Figure 1 It is a flowchart of the concentration-fatigue collaborative regulation method based on electroencephalogram (EEG) dual-modal feedback of the present invention.

[0039] Figure 2 It is a flowchart of the automatic recalibration of the present invention. Detailed Embodiment

[0040] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0041] As Figure 1 shown, the concentration-fatigue collaborative regulation method based on EEG dual-modal feedback includes the following steps.

[0042] (1) Electroencephalogram (EEG) data acquisition and processing.

[0043] Acquisition device: In this embodiment, a single-channel dry electrode EEG headband is used, and an open BCI (OpenBCI CytonBoard) single-channel dry electrode system is adopted. The electrode positioning refers to the Fp1 position in the international 10-20 system standard, and the sampling rate is ≥200 Hz to meet the requirements of the high-frequency band of β waves.

[0044] Signal preprocessing: Enable the built-in band-pass filter (0.5 - 40 Hz) of the device to eliminate power frequency interference and electromyographic noise. Further clean the signal through software filtering.

[0045] Power spectral density calculation: Adopt the fast Fourier transform (FFT), and calculate the power of each frequency band (θ / α / β) every 5-second window, with the unit of μV² / Hz.

[0046] θ wave: The frequency range is [4 Hz, 8 Hz). It is generally present in the state of people's mental trance or hypnosis.

[0047] α wave: The frequency range is [8 Hz, 13 Hz). It is the basic rhythm of the normal human brain wave. If there is no external stimulus, its frequency is quite constant. This rhythm is most obvious when a person is awake, quiet and with eyes closed.

[0048] β wave: The frequency range is [13 Hz, 30 Hz]. It is the most common high-frequency wave in people's waking state. It is the dominant brain wave of alert people.

[0049] (2)Bimodal index calculation.

[0050] Concentration index (β-wave ratio): , where , and represent the power of theta wave, alpha wave and beta wave respectively.

[0051] Fatigue index (θ / α ratio): .

[0052] (3)Collaborative judgment logic

[0053] (3.1)Threshold definition

[0054] Static concentration threshold : Set to 20% according to experimental data and not adjusted according to individual differences. When , it is determined that the concentration is insufficient.

[0055] Dynamic fatigue threshold : Calibrated based on the resting state limit value . The dynamic fatigue threshold is the critical value for judging whether the fatigue enters the dangerous area. The present invention is based on a dual mechanism of individualized reference value correction and real-time adaptive adjustment, and its calculation process includes the following key elements:

[0056] a) Physiological signal basis: θ / α ratio, reflecting the degree of fatigue accumulation;

[0057] b) Individual adaptability: Eliminate individual differences through the correction of the fatigue resting state reference value, and establish a personalized reference using the electroencephalogram activity of the trainee in the task-free state;

[0058] c) Nonlinear amplification factor: Use an empirical coefficient of 1.2 to compensate for the nonlinear change trend of physiological signals.

[0059] (3.2)The calculation method of the dynamic fatigue threshold includes the following steps:

[0060] Before training, let the trainee close their eyes and sit still for 2 minutes, and record the mean value of the θ / α ratio at this time ; The calculation formula of the dynamic fatigue threshold:

[0061] ;

[0062] Where: Resting state reference value (unit: standardized score), the average physiological index in the task-free state; 1.2 is the amplification factor, that is, the non-linear compensation factor determined by experience; is the lower limit constraint to prevent threshold collapse caused by excessive correction. It is the rate of change of the trend based on a sliding window (120 seconds in this embodiment), which is used to dynamically amplify or suppress the threshold to avoid insufficient individual adaptability caused by static coefficients. Assume that the sequence of θ / α ratio within the window is , and the corresponding time points are . By linearly regressing and fitting the data within the window, the trend slope k is obtained, and the slope k is converted into a proportional change rate relative to the current mean value, .

[0063] (3.3) Dual-threshold fusion strategy.

[0064] Level III risk (high fatigue and low concentration): and trigger the highest intervention (such as forced pause + nerve recovery stimulation);

[0065] Level II risk (abnormality in a single index, normal concentration, high fatigue): or trigger medium-level intervention (such as acoustic and optical warnings);

[0066] Level I risk (low risk): and trigger low-intensity feedback (such as operation accuracy prompts).

[0067] (4) Dynamic calibration and state determination.

[0068] When the threshold deviation degree for 3 consecutive cycles is detected to exceed ±5%, trigger the automatic recalibration process. As Figure 2 shown, when the fatigue index is detected to deviate from the dynamic fatigue threshold by ±5% for 3 consecutive cycles, start 30 seconds of resting-state resampling and recalculate the dynamic fatigue threshold.

[0069] Next, the method described in the present invention is verified through specific experiments.

[0070] The power company conducts safety training for high-voltage equipment operators, monitors the concentration and fatigue of trainees in real time, and ensures that the threshold adapts to individual differences and physiological state changes through a dynamic calibration mechanism.

[0071] Step 1, collection of resting-state reference values.

[0072] The trainee closes their eyes and sits still for 2 minutes, and the system records the mean value of their θ / α ratio as the resting-state reference value of fatigue .

[0073] Trainee A: (mean value of θ / α).

[0074] Step 2, calculation of dynamic threshold and initial calibration.

[0075] When the training of the trainee starts, the system needs to set the initial threshold based on the baseline value of the resting state. At this time, there is not enough real-time data to calculate the trend change rate, so it is temporarily set to 0;

[0076] Initial calibration of trainee A: 。

[0077] Step 3, real-time data monitoring and dynamic adjustment.

[0078] Trainee A conducts high-voltage cable operation training, and the EEG data is updated every 5 seconds. The monitoring and calculation results are shown in Table 1.

[0079] Table 1

[0080]

[0081] Adjustment logic: When continuously rises, it is amplified along with the trend (such as rising from 1.2 to 1.58) to avoid underestimation of the threshold due to cumulative fatigue. If it drops (such as due to a short break), it will automatically callback.

[0082] Step 4, abnormal data filtering and calibration trigger.

[0083] For trainee A, the is 1.58 at 240 seconds, but then due to the recovery of the physiological state, the θ / α ratio drops back to 1.3, and the deviation reaches -17.7% (>5%). The system re-collects data for nearly 5 minutes, updates and resets 。

[0084] Step 5, collaborative judgment and hierarchical intervention strategy.

[0085] When trainee A strips the cable, (<20%), (>1.2), triggering a forced pause operation + nerve recovery stimulation (10Hz strobe light + white noise).

[0086] The comparison of the monitoring and regulation effects of the fatigue index is shown in Table 2.

[0087] Table 2

[0088]

[0089] Among them, the intervention misjudgment rate: the situation where the system gives a wrong judgment for intervention, the number of wrong intervention times / the total number of trigger intervention times.

[0090] During the high-voltage cable operation training, the misoperations caused by trainees' fatigue are reduced by 60%, and the system response delay is <100 ms, meeting the real-time requirements of high-risk scenarios. The method described in the present invention can be extended to high-risk industries such as chemical industry and aviation, providing a standardized neurocognitive enhancement solution for special operation training.

Claims

1. A method for coordinated regulation of concentration and fatigue based on EEG dual-modal feedback, characterized in that: The steps include: Before the training begins, the first EEG data of each trainee in the resting state is collected, and the average of the ratio of the θ wave to the α wave is calculated as the resting state baseline value of the trainee's fatigue ; After the training started, the participants’ second EEG data were collected, and the concentration index was calculated based on the ratio of beta waves to the sum of beta waves, theta waves, and alpha waves, and the fatigue index was calculated based on the ratio of theta waves to alpha waves; According to the fatigue resting state reference value Calculate dynamic fatigue threshold; Dynamic fatigue threshold ; Among them, the ratio sequence of theta wave to alpha wave in the sliding window is fitted by linear regression to obtain the slope trend k, and k is converted into the trend change value of the ratio of theta wave to alpha wave in the sliding window relative to the mean. , , N is the sliding window length; 1.2 is the amplification factor, that is, the nonlinear compensation factor determined empirically; It is a lower limit constraint to prevent threshold collapse caused by over-correction; Set a static concentration threshold; Corresponding risk warnings are generated based on whether the concentration index deviates from the static concentration threshold and whether the fatigue index deviates from the dynamic fatigue threshold.

2. The method for coordinated control of concentration and fatigue based on EEG dual-modal feedback according to claim 1, characterized in that: When the training start time is less than one sliding window time, The value is 0.

3. The method for coordinated control of concentration and fatigue based on EEG dual-modal feedback according to claim 1, characterized in that: The generating of corresponding risk warnings according to whether the concentration index and the fatigue index exceed their corresponding thresholds includes: When the concentration index is less than its threshold and the fatigue index is greater than its threshold, the highest intervention is triggered; When the concentration index is not less than its threshold, or the fatigue index is greater than its threshold, intermediate intervention is triggered; When the concentration index is not less than its threshold and the fatigue index is not greater than its threshold, low-intensity feedback is triggered.

4. The method for coordinated control of concentration and fatigue based on EEG dual-modal feedback according to claim 1, characterized in that: When the fatigue index deviates from its corresponding threshold by ±5% in three consecutive cycles, the automatic recalibration process is triggered; the triggering of the automatic recalibration process includes: Resample the first EEG data in a resting state, and calculate and update the resting state reference value of fatigue , based on the updated resting fatigue baseline Recalculate dynamic fatigue threshold.

5. A concentration-fatigue coordinated control system based on EEG dual-modal feedback, characterized in that: include: The fatigue resting state baseline value calculation unit is used to collect the first EEG data of each trainee participating in the high-risk operation training in the resting state before the training begins, and calculate the average of the ratio of the θ wave to the α wave as the fatigue resting state baseline value of the trainee ; The concentration-fatigue coordinated regulation unit is used to collect the trainees' second EEG data after the training begins, calculate the concentration index according to the ratio of beta waves to the sum of beta waves, theta waves and alpha waves, and calculate the fatigue index according to the ratio of theta waves to alpha waves; According to the fatigue resting state reference value Calculate dynamic fatigue threshold; Dynamic fatigue threshold ; Among them, the ratio sequence of theta wave to alpha wave in the sliding window is fitted by linear regression to obtain the slope trend k, and k is converted into the trend change value of the ratio of theta wave to alpha wave in the sliding window relative to the mean. , , N is the sliding window length; 1.2 is the amplification factor, that is, the nonlinear compensation factor determined empirically; It is a lower limit constraint to prevent threshold collapse caused by over-correction; Set a static concentration threshold; Corresponding risk warnings are generated based on whether the concentration index deviates from the static concentration threshold and whether the fatigue index deviates from the dynamic fatigue threshold.

6. The concentration-fatigue coordinated control system based on EEG dual-modal feedback according to claim 5 is characterized in that: In the concentration-fatigue coordinated control unit, the generating of corresponding risk warnings according to whether the concentration index and the fatigue index exceed their corresponding thresholds includes: When the concentration index is less than its threshold and the fatigue index is greater than its threshold, the highest intervention is triggered; When the concentration index is not less than its threshold, or the fatigue index is greater than its threshold, intermediate intervention is triggered; When the concentration index is not less than its threshold and the fatigue index is not greater than its threshold, low-intensity feedback is triggered; In the concentration-fatigue coordinated control unit, when the fatigue index deviates from its corresponding threshold by ±5% in three consecutive cycles, an automatic recalibration process is triggered; the triggering of the automatic recalibration process includes: Resample the first EEG data in a resting state, and calculate and update the resting state reference value of fatigue , based on the updated resting fatigue baseline Recalculate dynamic fatigue threshold.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into the processor, the concentration-fatigue coordinated regulation method based on EEG dual-modal feedback is implemented according to any one of claims 1-4.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the concentration-fatigue coordinated regulation method based on EEG dual-modal feedback is implemented according to any one of claims 1 to 4.

Images