Intelligent nerve regulation and control method and system based on attention state
By collecting EEG signals in real time and using shallow convolutional neural network model to evaluate attention state, and personalized regulation is carried out in combination with transcranial electrical stimulation, the accuracy and sensitivity of attention state regulation in the existing technology are solved, and attention is achieved Efficient evaluation and adjustment of force state.
Patent Information
- Application Number
- CN202410995783.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-24
AI Technical Summary
The existing attention state regulation methods have problems with low accuracy and sensitivity, especially when behavioral performance assessment is easily subjective, and the sensitivity of sensory cues is also poor when EEG signal analysis is poor when the attention state is poor.
Real-time acquisition of EEG signals is used, attention status is evaluated through shallow convolutional neural network model, and neural regulation strategies are generated based on the evaluation results, and personalized regulation is used using transcranial electrical stimulation, including Gaussian random current, γ wave, β wave and transverse Cranial DC stimulation.
It achieves high accuracy and sensitivity assessment of attention state, can monitor and adjust attention state in real time, avoid interference with behavioral performance and external environmental impact, and provide more stable and reliable data.
Smart Images

Figure CN118925061A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of attention state analysis and regulation, and specifically relates to an intelligent neural regulation method and system based on attention state. Background Art
[0002] The state of human attention plays a key role in many jobs and tasks. Whether the attention is focused or not directly affects the work effect and safety. For example, when driving a car or an airplane, the driver needs to focus on the equipment and the environment so that he can respond quickly to emergencies. Therefore, some feasible technologies are used to adjust and enhance the state of attention to avoid serious accidents caused by distraction during tasks or work.
[0003] Behavior-based attention state analysis is a method to infer an individual's attention state by observing and analyzing their external behavior. This method can combine different behavioral characteristics, such as eye movements, facial expressions, body posture, and activity levels, to assess an individual's degree of concentration. However, factors such as environmental dependence, subjective performance characteristics, and individual differences pose great challenges to the accuracy of the data.
[0004] It is a common method to adjust the state of attention through sensory feedback prompts, such as using sounds, visual signals, tactile vibrations, etc. to remind individuals to maintain or restore their attention. However, when using this method of feedback prompts, if the user is concentrating on the task, the sensory feedback itself becomes a source of interference, causing the user's attention state to decrease; if other factors cause the user's attention to decrease during the task, their own sensory attention will also decrease, and the sensory prompts may also be ignored due to the user's lack of concentration, making it difficult to improve the state of attention through self-regulation.
[0005] In the existing methods, the attention state is generally evaluated by observing the user's behavior through video, such as whether yawning, whether the eyes are tired (Chinese patent with publication number CN112401887A), whether there are actions unrelated to the current behavior (eating actions / drinking actions / making phone calls / entertainment actions while driving, etc.) (Chinese patent with publication number CN109937152A), and based on this, only a two-level evaluation of the user's attention state is performed. In addition, there are also a small number of methods that use EEG signals as the basis for analysis of attention state monitoring (Chinese patent with publication number CN112957049A). Among these existing methods, on the one hand, the recognition rate is easily affected by subjective performance when evaluating the user's behavior performance, and the monitoring equipment is also affected by the environment (video occlusion) to reduce the recognition rate; on the other hand, in the method of analyzing the attention state through EEG signals, when the user's attention state is low, it is often fed back by sensory prompts, but when the attention state is poor, the sensitivity of the senses is also poor. Summary of the invention
[0006] In order to solve the problems of low accuracy and sensitivity in existing attention state regulation methods, the present invention provides an intelligent neural regulation method and system based on attention state.
[0007] The technical solution adopted by the present invention to solve the technical problem is as follows:
[0008] The present invention provides an intelligent neural control method based on attention state, which mainly includes the following steps:
[0009] (1) Collect the user's EEG signals in real time and capture brain waves related to attention status;
[0010] (2) Preprocessing the collected raw EEG signals and extracting attention-related features from the EEG signals;
[0011] (3) When the user initially uses the device, collect EEG signals of the user in different attention states according to the usage task requirements to establish baseline data; use intelligent learning algorithms to train the attention state assessment model based on the baseline data;
[0012] (4) When the user uses the device, the system will collect and monitor EEG signals in real time, and use the trained attention state evaluation model to analyze and evaluate the real-time collected EEG signals to determine the user's current attention state;
[0013] (5) Generate neural regulation strategies;
[0014] Based on the real-time attention state evaluation results, determine whether neural regulation needs to be performed; when neural regulation needs to be performed, the system will generate a neural regulation strategy to select appropriate neural regulation methods and neural regulation parameters;
[0015] (6) Implement neuromodulatory strategies.
[0016] Furthermore, in step (2), the collected original EEG signal is filtered.
[0017] Furthermore, in step (2), the EEG signal in each time window is subjected to spectrum analysis, and the power value of each frequency band is calculated using fast Fourier transform.
[0018] Furthermore, in step (3), the attention state evaluation model includes: a data preprocessing module and a shallow convolutional neural network module;
[0019] The data preprocessing module is used to organize and convert EEG signal data into time and frequency: first, the data is cleaned to remove redundant and missing data to reduce the data operation burden of the model; then the EEG signal in each time window is subjected to spectrum analysis, and the power value of each frequency band is calculated using fast Fourier transform; then the principal component analysis method is used to reduce the number of features; finally, the data is normalized to accelerate the convergence of the model;
[0020] The shallow convolutional neural network module is used to classify and evaluate the preprocessed data; the shallow convolutional neural network module includes: an input layer, two convolutional layers, two pooling layers, a fully connected layer and an output layer; the input layer obtains the preprocessed signal according to the perception used by the individual user and the corresponding brain area or multi-perception; the convolutional layer adopts a double-layer structure, extracts local features through two convolutional layers, and a maximum pooling layer is set after each convolutional layer to reduce the feature size; the fully connected layer further optimizes the results to reduce overfitting; the output layer adopts the Softmax activation function to realize the classification and evaluation of attention status.
[0021] Furthermore, in step (3), the specific training process of the attention state evaluation model is as follows:
[0022] The training process of the attention state assessment model relies on the EEG signal data of different attention states of individual users for training; the tasks completed by the users for the first time are complex math problem tasks with high cognitive load, reading tasks with medium cognitive load, image recognition tasks with low cognitive load, and resting with eyes closed in a relaxed state. These tasks are divided into high attention state, medium attention state, low attention state, and distracted attention state in turn. EEG signal data are collected under different attention states as the basis for classification, wherein the ratios of training set and test set are 80% and 20% respectively, and the shallow convolutional neural network module is trained by cross-validation method.
[0023] Furthermore, in step (4), the threshold and softmax classification algorithm are used to evaluate the user's current attention state. The specific operation process is as follows:
[0024] After the data is processed by the fully connected layer, the optimized classification result is sent to the softmax output layer and the sigmoid output layer respectively, and the data category is judged by probability and threshold respectively. The softmax output layer outputs a multi-classification probability distribution, and then obtains the category through Argmax, where the maximum probability category is taken, and then the category is predicted through softmax, and the A category is output; the sigmoid output layer outputs the probability of each category, and then judges the category through the threshold, and the value greater than the threshold is 1. Finally, the B category is output after the threshold is judged; finally, the A category and the B category are integrated to obtain the C category, and the final classification result is output.
[0025] Furthermore, in step (5), the optimal neural regulation strategy is screened adaptively through the EEG real-time monitoring and attention state assessment results and the transcranial electrical stimulation strategy. The specific operation process is as follows:
[0026] After real-time EEG monitoring and attention state evaluation, the user state level is obtained. For the evaluated high attention state, medium attention state, low attention state or distracted attention state, the corresponding transcranial electrical stimulation strategies with different parameters are set: when in the high attention state, a Gaussian random current stimulation scheme in the range of 0.5-1mA is used to maintain the attention state and optimize the allocation of attention resources; when in the medium attention state, two types of mixed transcranial electrical stimulation schemes in the range of 0.5-1mA are used, which are Gaussian random current stimulation scheme and γ wave (30-45Hz) co / sinusoidal alternating current stimulation scheme respectively; when in the low attention state, two types of mixed transcranial electrical stimulation schemes in the range of 1.0-1.5mA are used, which are Gaussian random current stimulation scheme and β wave (13-30Hz) co / sinusoidal alternating current stimulation scheme respectively; when in the distracted attention state, three types of mixed transcranial electrical stimulation schemes in the range of 1.5-2mA are used, which are Gaussian random current stimulation scheme, β wave co / sinusoidal alternating current stimulation scheme and transcranial direct current stimulation scheme respectively.
[0027] Furthermore, after each stimulation cycle, the attention state will be evaluated again, and the transcranial electrical stimulation strategy will be optimized based on the result of the evaluation threshold. When the first evaluation result is a high attention state, real-time periodic monitoring and evaluation will be performed, and Gaussian random current stimulation will be performed after 1, 3, 5... cycles in turn. After the attention state level is evaluated again as a medium attention state, a low attention state, or a distracted attention state, the following corresponding transcranial electrical stimulation strategies will be performed: when the threshold after stimulation tends to be a high attention state, the last transcranial electrical stimulation scheme will be maintained, and the transcranial electrical stimulation scheme will also be the optimal neural regulation scheme. When the threshold after stimulation tends to remain unchanged or decrease, the parameters of the last transcranial electrical stimulation scheme will be adjusted, and the current and frequency will be increased by 0.1 mA and 1 Hz, and the adjustment will be performed once every cycle until the optimal neural regulation scheme is adjusted.
[0028] The time proportions corresponding to the Gaussian random current stimulation scheme in the high, medium, low attention states and the distracted attention state are 100%, 50%, 30% and 10% respectively; the time proportion corresponding to the transcranial direct current stimulation scheme in the distracted attention state is 30%, and the remaining time is the corresponding co / sinusoidal alternating current stimulation scheme; at the same time, the stimulation time period proportion will remain unchanged as the real-time monitoring threshold tends to the high attention state, or the threshold corresponding to the attention state remains unchanged or tends to the low attention state, and the co / sinusoidal alternating current stimulation scheme proportion time is increased to obtain the optimal neural regulation scheme; and when the threshold remains unchanged or decreases in the distracted attention state, the time proportion corresponding to the transcranial direct current stimulation scheme is increased.
[0029] The present invention provides an intelligent neural regulation system based on attention state, which is used to implement the intelligent neural regulation method based on attention state. The system mainly includes: an EEG acquisition module, a main control module and a neural regulation module; the main control module sends a start instruction for EEG signal acquisition to the EEG acquisition module, the EEG acquisition module starts to collect EEG signals and sends them to the main control module, the main control module performs signal processing and feature extraction, and simultaneously performs model training and real-time attention state evaluation, and judges whether neural regulation needs to be performed according to the real-time attention state evaluation result; when neural regulation needs to be performed, a neural regulation strategy will be generated in the neural regulation module, the main control module sends a neural regulation parameter instruction to the neural regulation module, the neural regulation module will execute the neural regulation strategy according to the neural regulation parameter instruction, and feedback the control end signal to the main control module.
[0030] Furthermore, when it is necessary to execute a neural regulation strategy for the user's attention state, the EEG acquisition module and the neural regulation module alternately start and stop the operation of the cyclic regulation scheme.
[0031] The beneficial effects of the present invention are:
[0032] The present invention uses a personalized assessment level of attention status based on EEG signals, and intelligently selects the optimal non-invasive transcranial electrical stimulation neural regulation method based on the assessment results to improve attention status. The present invention implements a graded assessment of attention status, and uses a neural regulation method to improve attention when the attention status is poor.
[0033] The present invention mainly solves the problem of real-time monitoring of EEG signals for attention state level evaluation algorithm, and adopts the optimal mode of transcranial electrical stimulation according to the evaluation level results to improve the user's attention state, and can meet diversified needs according to the user's environmental perception needs (such as hearing, vision, touch, etc.).
[0034] Compared with the attention state analysis method of behavioral performance, the present invention adopts electroencephalogram (EEG) signal to analyze to achieve higher effectiveness (EEG measurement time resolution can reach milliseconds), and reflect brain activity changes in real time; while directly measuring brain activity, it can accurately reflect the physiological basis of attention state, avoid interference and errors that may be introduced by behavioral performance, and EEG is very sensitive to small changes in attention, and can detect subtle changes that cannot be captured by behavioral performance, so that it has higher accuracy and sensitivity; behavioral performance analysis depends on external performance and is easily affected by factors such as external environment and individual emotions, while EEG is not disturbed by these factors and can provide more stable and reliable data; most importantly, EEG can identify the hidden attention state that the user does not show in behavior, such as distraction but no obvious changes in behavior. It can be seen that compared with the attention state analysis method of behavioral performance, the use of EEG physiological signal analysis has higher timeliness, accuracy and sensitivity, does not depend on external behavioral performance, can identify hidden attention state, and has broad application prospects and practical advantages. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 A flow chart of an attention state-based intelligent neural regulation method provided by the present invention.
[0036] Figure 2 A block diagram of an attention state-based intelligent neural regulation system provided by the present invention.
[0037] Figure 3 Schematic diagram of the structural composition of a shallow convolutional neural network module.
[0038] Figure 4 The specific operation flow chart of using threshold and softmax classification algorithm to evaluate the user's current attention state. DETAILED DESCRIPTION
[0039] The present invention is further described in detail below in conjunction with the accompanying drawings.
[0040] In the first aspect, the present invention provides an intelligent neural control method based on attention state. The present invention mainly includes two important parts:
[0041] (1) Personalized attention state level assessment method
[0042] The user's electroencephalogram (EEG) signal is collected in real time and preprocessed, the spectrum of the EEG signal in each time window is analyzed, and the power value of each frequency band is calculated using fast Fourier transform (FTT).
[0043] During the initial stage of the user's use of the device, the user's EEG signals in different attention states will be collected, and an attention state assessment model will be established, including relaxation, light concentration, and high concentration.
[0044] When the user uses the device, the system will collect and monitor EEG signals in real time and compare them with the attention state evaluation model, and use the threshold and softmax classification algorithm to evaluate the user's current attention state. The evaluation results will be fed back to the user in real time through the user interface. At the same time, the current attention state will be converted into appropriate instructions and the user's attention state will be adjusted using intelligent neural regulation methods.
[0045] (2) Intelligent and personalized neural regulation methods
[0046] The method chosen for neural regulation is transcranial electrical stimulation, which stimulates the cerebral cortex through low-intensity electric current to affect neural activity, provide real-time feedback on the user's EEG activity and differentiate their attention states. At the same time, based on the difference in attention state levels compared with the benchmark, the neural regulation method is intelligently executed, especially the setting of neural regulation parameters of transcranial electrical stimulation, to help users adjust their attention states.
[0047] According to the initialized user attention state setting, the neuroregulatory parameters of transcranial electrical stimulation are intervened according to the user's personalized neuroregulatory scheme, including the transcranial electrical stimulation current scheme (tDCS, tRCS, and tACS, etc.), current intensity, stimulation cycle, and stimulation duration that set the perception needs, and achieve optimal stimulation through personalized customization. At the same time, the neuroregulatory method takes into account individual differences of users, provides immediate intervention and adjustment, and especially continuously optimizes and improves the attention state regulation effect based on user feedback data. In these transcranial electrical stimulation methods, a closed-loop regulation based on a human-intelligent neuroregulatory system is formed through the evaluation of the attention state level.
[0048] Transcranial Electrical Stimulation (tES) neurofeedback regulation method has significant advantages over the regulation method of sensory cues such as audio-visual. First, tES acts directly on the brain, regulating neuronal activity through microcurrent, thereby changing the physiological state of the brain. This direct way of regulating brain activity is more effective than indirectly affecting the brain through sensory cues such as audio-visual. Second, tES can produce obvious effects in a short time, while methods such as audio-visual cues require multiple steps such as sensory input and information processing, and tES acts directly on neurons, which can change the state of attention more quickly. The effect of tES can last for a long time, and even after the stimulation ends, the activity of neurons is still affected, while the effect of audio-visual cues is usually short-lived, and tES can accurately act on specific areas of the brain, such as the frontal lobe, parietal lobe, etc., which are related to attention, thereby achieving more efficient regulation. Finally, the tES method does not require users to actively participate or respond to sensory cues, which reduces the occupation of users' attention. Sensory stimuli such as audio-visual cues may be affected by external factors such as environmental noise and light. tES is performed in a quieter environment and is not affected by these interferences. Therefore, compared with sensory cues such as audio and video, the transcranial electrical stimulation neurofeedback regulation method has many advantages, such as more direct and effective ability to regulate brain neural activity, faster immediate effects and longer duration, higher accuracy, and no need for active user participation and sensory coordination. This makes tES have significant advantages and application potential in other neural regulation aspects such as attention state regulation.
[0049] like Figure 1 As shown, the present invention provides an intelligent neural regulation method based on attention state, which specifically includes the following steps:
[0050] (1) Physiological signal collection;
[0051] The EEG sensor collects the user's EEG signals in real time to capture brain waves related to the state of attention. For example, when visual and auditory needs are met, the EEG signals of the visual and auditory areas will be focused on.
[0052] (2) Signal processing and feature extraction;
[0053] (2.1) Preprocessing the collected raw EEG signals;
[0054] Among them, preprocessing mainly includes filtering.
[0055] Among them, the filtering methods mainly include: frequency filtering and principal component analysis, but are not limited thereto.
[0056] (2.2) Extract attention-related features from EEG signals; specifically, perform spectral analysis on the EEG signal within each time window and use fast Fourier transform (FTT) to calculate the power value of each frequency band.
[0057] The specific operation process of using fast Fourier transform (FTT) to calculate the power value of each frequency band is as follows:
[0058] The user sets or adjusts the time period, specifically collecting EEG signals in a time period of 10 to 20 minutes. After signal filtering, the power value of each frequency band (including δ, θ, α, β, γ) is calculated using fast Fourier transform for the EEG signal within this time window.
[0059] (3) Personalized model training;
[0060] During the initial stage of using the device, the user's EEG signals in different attention states are collected according to the task requirements (vision, hearing, touch, etc.) to establish personalized baseline data. Then, an intelligent learning algorithm is used to train a personalized attention state assessment model based on the baseline data, including relaxation, light concentration and high concentration.
[0061] The attention state evaluation model in the present invention mainly includes: a data preprocessing module and a shallow convolutional neural network module.
[0062] The data preprocessing module mainly realizes the sorting and time-frequency conversion of EEG signal data. Specifically: first, the data preprocessing module cleans the data, removes redundant and missing data, and reduces the data operation burden of the model; then, the EEG signal in each time window is analyzed by spectrum, and the power value of each frequency band is calculated by fast Fourier transform; then the principal component analysis (PCA) method is used to reduce the number of features; finally, the data is normalized to speed up the convergence of the model.
[0063] The shallow convolutional neural network module is used to classify and evaluate the preprocessed data. Figure 3 As shown in the figure, the shallow convolutional neural network module mainly includes: input layer, two convolutional layers, two pooling layers, fully connected layer and output layer. Among them, the input layer obtains the preprocessed signal according to the perception used by the individual user and its corresponding brain area (vision corresponds to the occipital lobe, hearing corresponds to the temporal lobe and touch corresponds to the parietal lobe, etc.) or multi-perception; the convolution layer adopts a double-layer structure, extracts local features through two convolutional layers, and sets a maximum pooling layer after each convolutional layer to reduce the feature size; the fully connected layer further optimizes the results to reduce overfitting; the output layer uses the Softmax activation function to realize the classification and evaluation of attention state.
[0064] The hardware environment used by the attention state evaluation model in the present invention is based on a portable mobile hardware system, so a lightweight evaluation model is adopted. At the same time, the attention state evaluation model is based on a personalized evaluation scheme, which makes the attention state evaluation model not need to process a large amount of data, but only requires individual attention state difference data, which also reduces the requirements for the computational complexity of the attention state evaluation model and maintains the evaluation performance of the attention state evaluation model.
[0065] Among them, the specific training process of the attention state evaluation model is as follows:
[0066] The specific training process of the attention state assessment model relies on the EEG signal data of different attention states of individual users for training. When users use the system for the first time, they will be required to complete different tasks for a specified number of times. These tasks can regulate users in different attention states. The EEG acquisition module obtains the EEG data set of the corresponding state of the user. The attention state assessment model uses the data in these EEG data sets for training and verification, so as to obtain a personalized and stable attention state assessment model. Among them, the tasks completed by the user for the first time are complex mathematical problem tasks with high cognitive load, reading tasks with medium cognitive load, image recognition tasks with low cognitive load, and resting with eyes closed in a relaxed state. These tasks are divided into four attention states: high attention state, medium attention state, low attention state, and distracted attention state. The EEG signal data collected under different attention states will be used as the basis for classification. In the four different attention state data sets collected, the proportions of training set and test set are 80% and 20% respectively, and the shallow convolutional neural network module is trained by cross-validation method to further improve the generalization ability of individual user data and reduce the deviation caused by data division.
[0067] (4) Real-time attention state assessment;
[0068] When the user uses the device, the system will collect and monitor EEG signals in real time, and use the trained attention state evaluation model to analyze and evaluate the real-time collected EEG signals to determine the user's current attention state. Specifically, the real-time collected EEG signals are compared with the trained attention state evaluation model, and the threshold and softmax classification algorithm are used to evaluate the user's current attention state.
[0069] The specific operation process of using the threshold and softmax classification algorithm to evaluate the user's current attention state is as follows:
[0070] In order to further improve the accuracy of the assessment and classification of the attention state, the present invention optimizes its classification method and adopts the threshold and softmax classification algorithm to assess the attention state. Figure 4As shown. After the data of the model is processed by the fully connected layer, the optimized classification result is sent to the softmax output layer and the sigmoid output layer respectively. The probability and threshold are used to judge the category of the data respectively. The softmax output layer outputs the multi-classification probability distribution, and then obtains the category through Argmax, in which the maximum probability category is taken, and then the category is predicted through softmax, and the A category is output; the sigmoid output layer outputs the probability of each category, and then judges the category through the threshold, and the value greater than the threshold is 1. Finally, the B category is output after judging the category through the threshold; finally, the above A category and B category are integrated to obtain the C category, and the final classification result is output, and the classification accuracy is improved by integrating the results.
[0071] (5) Generation of neural regulatory strategies;
[0072] According to the real-time attention state evaluation results, it is determined whether neuroregulation needs to be performed. When neuroregulation needs to be performed, the system will generate a personalized neuroregulation strategy to select appropriate neuroregulation methods and neuroregulation parameters. Among them, the user's personalized neuroregulation strategy will be used in the previous neuroregulation processes to adaptively select the optimal neuroregulation strategy through EEG real-time monitoring and attention state evaluation results and transcranial electrical stimulation strategies, and finally realize personalized customization of the neuroregulation strategy for the user's attention state.
[0073] Among them, the specific screening process of adaptive optimal neural regulation strategy based on EEG real-time monitoring and attention state assessment results and transcranial electrical stimulation strategy is as follows:
[0074] The present invention obtains the user state level after EEG real-time monitoring and attention state evaluation, and sets the transcranial electrical stimulation strategy corresponding to different parameters for the evaluated high attention state, medium attention state, low attention state or attention distraction state. When in the high attention state, a Gaussian random current stimulation scheme in the range of 0.5 to 1 mA is used to maintain the attention state and optimize the allocation of attention resources; when in the medium attention state, two types of mixed transcranial electrical stimulation schemes in the range of 0.5 to 1 mA are used, respectively using a Gaussian random current stimulation scheme and a γ wave (30 to 45 Hz) co / sine alternating current stimulation scheme; when in the low attention state, two types of mixed transcranial electrical stimulation schemes in the range of 1.0 to 1.5 mA are used, respectively using a Gaussian random current stimulation scheme and a β wave (13 to 30 Hz) co / sine alternating current stimulation scheme; when in the attention distraction state, three types of mixed transcranial electrical stimulation schemes in the range of 1.5 to 2 mA are used, respectively using a Gaussian random current stimulation scheme, a β wave co / sine alternating current stimulation scheme and a transcranial direct current stimulation scheme.
[0075] After each stimulation cycle, the attention state will be evaluated again, and the transcranial electrical stimulation strategy will be optimized based on the result of the evaluation threshold. When the first evaluation result is a high attention state, real-time periodic monitoring and evaluation will be performed, and Gaussian random current stimulation will be performed after 1, 3, 5... cycles. After the attention state level is evaluated again as a medium attention state, a low attention state, or a distracted attention state, the following corresponding transcranial electrical stimulation strategy will be executed: when the threshold after stimulation is biased towards a high attention state, the last transcranial electrical stimulation scheme will be maintained, and this transcranial electrical stimulation scheme will also be the optimal neuroregulation scheme; when the threshold after stimulation is biased towards a constant or decreased state, the parameters of the last transcranial electrical stimulation scheme will be adjusted, and the current and frequency will be increased by 0.1 mA and 1 Hz, and the adjustment will be performed once every cycle until the optimal neuroregulation scheme is reached. During the regulation of the transcranial electrical stimulation scheme, these two parameters (current and frequency) do not exceed the limit values of the corresponding transcranial electrical stimulation scheme. For example, the maximum stimulation current in the medium attention state is 1 mA, and the maximum stimulation frequency is 45 Hz. Only when the attention state changes (such as the low attention state changes to the medium attention state), the corresponding transcranial electrical stimulation scheme will be adjusted.
[0076] Among them, the time proportion of the hybrid transcranial electrical stimulation scheme also changes dynamically. The corresponding time proportions of the Gaussian random current stimulation scheme in high, medium, low attention states and distracted attention states are 100%, 50%, 30% and 10% respectively. The time proportion of the transcranial direct current stimulation scheme in the distracted attention state is 30%, and the remaining time is the corresponding co / sinusoidal alternating current stimulation scheme. At the same time, the proportion of the stimulation time cycle will remain unchanged as the real-time monitoring threshold tends to the high attention state, or the threshold corresponding to the attention state remains unchanged or tends to the low attention state, and the co / sinusoidal alternating current stimulation scheme proportion time is increased to obtain the optimal neural regulation scheme. In the distracted attention state, if the threshold remains unchanged or decreases, the corresponding time proportion of the transcranial direct current stimulation scheme is increased. The above adjustment time proportion is adjusted in real time by 5% of the time proportion each time.
[0077] (6) neuromodulatory strategy execution;
[0078] Start the neural control module C to implement the neural control strategy. During the neural control process, the user's EEG signal is monitored in real time, and the EEG signal (attention state analysis and rating) is collected in real time using the EEG acquisition module A in a time window. When it is necessary to execute the neural control strategy of the user's attention state, in order to avoid the influence of transcranial electrical stimulation on the EEG signal, the EEG acquisition module A and the neural control module C alternately start and stop the cycle control scheme, that is, the EEG acquisition module A (attention state analysis and rating) - neural control module C - EEG acquisition module A (attention state analysis and rating). Only when the attention state rating of the EEG acquisition module A meets the requirements, the EEG acquisition module A (attention state analysis and rating) will be executed. Once the attention state rating of the EEG acquisition module A is low, the system will execute the start and stop cycle control scheme to adjust the user's attention state.
[0079] In a second aspect, the present invention provides an intelligent neural control system based on attention state. The system is mainly used to implement an intelligent neural control method based on attention state described in the first aspect of the present invention.
[0080] like Figure 2 As shown, an intelligent neural regulation system based on attention state of the present invention mainly includes: an EEG acquisition module A, a main control module B and a neural regulation module C. The user interacts with the EEG acquisition module A and the neural regulation module C respectively, and the main control module B interacts with the EEG acquisition module A and the neural regulation module C respectively.
[0081] Specifically, the main functions and effects of the EEG acquisition module A are: the EEG signal acquisition of specific sensory organs (vision / auditory / touch / olfaction / taste) can be customized.
[0082] Specifically, the main functions and roles of the main control module B are: 1. Evaluate the attention state level based on the collected EEG signals; 2. Send the start / stop command of EEG acquisition to the EEG acquisition module A; 3. Send the optimal control command parameters according to the attention state evaluation level.
[0083] Specifically, the main functions and effects of the neural regulation module C are: 1. Initiating the stimulation mode according to the received instructions; 2. Feedback the end instruction after the stimulation is completed.
[0084] The present invention is an intelligent neural regulation system based on attention state, and its specific working principle is as follows:
[0085] The main control module B sends a start instruction for EEG signal acquisition to the EEG acquisition module A, and the EEG acquisition module A starts to collect EEG signals and sends them to the main control module B. The main control module B performs signal processing and feature extraction, and at the same time performs personalized model training and real-time attention state evaluation, and judges whether neural regulation needs to be executed based on the real-time attention state evaluation results; when neural regulation needs to be executed, a personalized neural regulation strategy will be generated in the neural regulation module C, and the main control module B will send a neural regulation parameter instruction to the neural regulation module C. The neural regulation module C will execute the neural regulation strategy based on the neural regulation parameter instruction, and feedback the control end signal to the main control module B.
[0086] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. An intelligent neural control method based on attention state, It is characterized in that The following steps are involved: (1) Collect the user's EEG signals in real time and capture brain waves related to attention status; (2) Preprocessing the collected raw EEG signals and extracting attention-related features from the EEG signals; (3) When the user initially uses the device, collect EEG signals of the user in different attention states according to the usage task requirements to establish baseline data; use intelligent learning algorithms to train the attention state assessment model based on the baseline data; (4) When the user uses the device, the system will collect and monitor EEG signals in real time, and use the trained attention state evaluation model to analyze and evaluate the real-time collected EEG signals to determine the user's current attention state; (5) Generate neural regulation strategies; Based on the real-time attention state assessment results, determine whether neural regulation needs to be performed; When neuromodulation needs to be performed, the system will generate a neuromodulation strategy to select appropriate neuromodulation methods and neuromodulation parameters; (6) Implement neuromodulatory strategies.
2. The intelligent neural control method based on attention state according to claim 1, It is characterized in that In step (2), the collected original EEG signal is filtered.
3. The intelligent neural control method based on attention state according to claim 1, It is characterized in that In step (2), the EEG signal in each time window is subjected to spectrum analysis, and the power value of each frequency band is calculated using fast Fourier transform.
4. The intelligent neural control method based on attention state according to claim 1, It is characterized in that In step (3), the attention state evaluation model includes: a data preprocessing module and a shallow convolutional neural network module; The data preprocessing module is used to organize and convert EEG signal data into time and frequency: first, the data is cleaned to remove redundant and missing data to reduce the data operation burden of the model; then the EEG signal in each time window is subjected to spectrum analysis, and the power value of each frequency band is calculated using fast Fourier transform; then the principal component analysis method is used to reduce the number of features; finally, the data is normalized to accelerate the convergence of the model; The shallow convolutional neural network module is used to classify and evaluate the preprocessed data; the shallow convolutional neural network module includes: an input layer, two convolutional layers, two pooling layers, a fully connected layer and an output layer; the input layer obtains the preprocessed signal according to the perception used by the individual user and the corresponding brain area or multi-perception; the convolutional layer adopts a double-layer structure, extracts local features through two convolutional layers, and a maximum pooling layer is set after each convolutional layer to reduce the feature size; the fully connected layer further optimizes the results to reduce overfitting; the output layer adopts the Softmax activation function to realize the classification and evaluation of attention status.
5. The intelligent neural control method based on attention state according to claim 4, It is characterized in that In step (3), the specific training process of the attention state evaluation model is as follows: The training process of the attention state assessment model relies on the EEG signal data of different attention states of individual users for training; the tasks completed by the users for the first time are complex math problem tasks with high cognitive load, reading tasks with medium cognitive load, image recognition tasks with low cognitive load, and resting with eyes closed in a relaxed state. These tasks are divided into high attention state, medium attention state, low attention state, and distracted attention state in turn. EEG signal data are collected under different attention states as the basis for classification, wherein the ratios of training set and test set are 80% and 20% respectively, and the shallow convolutional neural network module is trained by cross-validation method.
6. The intelligent neural control method based on attention state according to claim 4, It is characterized in that In step (4), the threshold and softmax classification algorithm are used to evaluate the user's current attention state. The specific operation process is as follows: After the data is processed by the fully connected layer, the optimized classification result is sent to the softmax output layer and the sigmoid output layer respectively, and the data category is judged by probability and threshold respectively. The softmax output layer outputs a multi-classification probability distribution, and then obtains the category through Argmax, where the maximum probability category is taken, and then the category is predicted through softmax, and the A category is output; the sigmoid output layer outputs the probability of each category, and then judges the category through the threshold, and the value greater than the threshold is 1. Finally, the B category is output after the threshold is judged; finally, the A category and the B category are integrated to obtain the C category, and the final classification result is output.
7. The intelligent neural control method based on attention state according to claim 1, It is characterized in that In step (5), the optimal neural regulation strategy is screened adaptively through the results of EEG real-time monitoring and attention state assessment and the transcranial electrical stimulation strategy. The specific operation process is as follows: After real-time EEG monitoring and attention state evaluation, the user state level is obtained. For the evaluated high attention state, medium attention state, low attention state or distracted attention state, the corresponding transcranial electrical stimulation strategies with different parameters are set: when in the high attention state, a Gaussian random current stimulation scheme in the range of 0.5-1mA is used to maintain the attention state and optimize the allocation of attention resources; when in the medium attention state, two types of mixed transcranial electrical stimulation schemes in the range of 0.5-1mA are used, which are Gaussian random current stimulation scheme and γ wave (30-45Hz) co / sinusoidal alternating current stimulation scheme respectively; when in the low attention state, two types of mixed transcranial electrical stimulation schemes in the range of 1.0-1.5mA are used, which are Gaussian random current stimulation scheme and β wave (13-30Hz) co / sinusoidal alternating current stimulation scheme respectively; when in the distracted attention state, three types of mixed transcranial electrical stimulation schemes in the range of 1.5-2mA are used, which are Gaussian random current stimulation scheme, β wave co / sinusoidal alternating current stimulation scheme and transcranial direct current stimulation scheme respectively.
8. The intelligent neural control method based on attention state according to claim 7, It is characterized in that After each stimulation cycle, the attention state will be evaluated again, and the transcranial electrical stimulation strategy will be optimized based on the result of the evaluation threshold. When the first evaluation result is a high attention state, real-time periodic monitoring and evaluation will be performed, and Gaussian random current stimulation will be performed after 1, 3, 5... cycles in turn. After the attention state level is evaluated as a medium attention state, a low attention state, or a distracted attention state, the following corresponding transcranial electrical stimulation strategies will be executed: When the threshold after stimulation tends to be a high attention state, the last transcranial electrical stimulation scheme will be maintained, and this transcranial electrical stimulation scheme will also be the optimal neural regulation scheme. When the threshold after stimulation tends to remain unchanged or decrease, the parameters of the last transcranial electrical stimulation scheme will be adjusted, and the current and frequency will be increased by 0.1 mA and 1 Hz, and the adjustment will be performed once every cycle until the optimal neural regulation scheme is adjusted. The time proportions corresponding to the Gaussian random current stimulation scheme in the high, medium, low attention states and the distracted attention state are 100%, 50%, 30% and 10% respectively; the time proportion corresponding to the transcranial direct current stimulation scheme in the distracted attention state is 30%, and the remaining time is the corresponding co / sinusoidal alternating current stimulation scheme; at the same time, the stimulation time period proportion will remain unchanged as the real-time monitoring threshold tends to the high attention state, or the threshold corresponding to the attention state remains unchanged or tends to the low attention state, and the co / sinusoidal alternating current stimulation scheme proportion time is increased to obtain the optimal neural regulation scheme; and when the threshold remains unchanged or decreases in the distracted attention state, the time proportion corresponding to the transcranial direct current stimulation scheme is increased.
9. An attention state-based intelligent neural control system, used to implement an attention state-based intelligent neural control method according to any one of claims 1 to 8, It is characterized in that The system includes: an EEG acquisition module, a main control module and a neural regulation module; the main control module sends a start instruction for EEG signal acquisition to the EEG acquisition module, the EEG acquisition module starts to acquire EEG signals and sends them to the main control module, the main control module performs signal processing and feature extraction, and simultaneously performs model training and real-time attention state evaluation, and determines whether neural regulation needs to be performed based on the real-time attention state evaluation result; when neural regulation needs to be performed, a neural regulation strategy will be generated in the neural regulation module, the main control module sends a neural regulation parameter instruction to the neural regulation module, the neural regulation module will execute the neural regulation strategy based on the neural regulation parameter instruction, and feedback a control end signal to the main control module.
10. The intelligent neural regulation system based on attention state according to claim 9, It is characterized in that When it is necessary to execute the neural regulation strategy of the user's attention state, the EEG acquisition module and the neural regulation module alternately start and stop the cyclic regulation scheme operation.
Citation Information
Patent Citations
Control system
CN108837304A
Attention regulating method based on transcranial direct current stimulation
CN109589493A
Industry multi-label noise reduction method based on cyclic relabeling self-service method
CN110705607A
Method for improving attention breadth through transcranial direct current stimulation and device thereof
CN111281379A
Image recognition method and system, computer equipment and storage medium
CN112613581A
Cited By
Intelligent auxiliary training brain-computer system for improving attention
CN119587833A
Real-time evaluation method and system for transcranial electrical stimulation effect
CN119989823A
Transcranial electrical stimulation physiotherapy system based on EEG-fNIRS feedback
CN120189632A
Nerve detection method
CN120753675A
Transcranial electrical stimulation control method and system for regulating mental fatigue
CN120919533A
