Online game obstacle recognition, prediction and intervention effect evaluation method and system based on electroencephalogram characteristics and machine learning model

Abstract

The invention relates to the technical field of medical artificial intelligence, and discloses an online game barrier recognition, prediction and intervention effect evaluation method and system based on electroencephalogram characteristics and a machine learning model, and the method comprises the steps: building an online game barrier recognition model through obtaining electroencephalogram data of a target group in a resting state and a game clue reaction task; an objective judgment basis is provided, and the limitation of a traditional subjective scale is overcome; an online game craving degree prediction model is constructed by obtaining electroencephalogram data of an addict under a task and craving score data after the task, the craving degree is predicted in real time, and personalized intervention is supported; an online game addiction intervention effect evaluation model is constructed by obtaining electroencephalogram data and desire score data of addicts before and after transcranial direct current stimulation intervention, the intervention effect is quantified, and the durability of the intervention effect is evaluated. According to the method, a whole-process closed loop from identification, prediction to evaluation can be realized, the judgment objectivity, the prediction accuracy and the intervention pertinence are remarkably improved, and the method has important scientific value and application prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of medical artificial intelligence, and particularly to a method and system for identifying, predicting, and evaluating the intervention effect of Internet gaming disorder based on electroencephalogram features and machine learning models. Background Art

[0002] In the prior art, the determination of Internet Gaming Disorder (IGD) mainly relies on subjective scales, such as the Internet Addiction Test (IAT) or the Internet Gaming Disorder Scale (IGDS). Although these scales can reflect the individual's addiction level to a certain extent, their results are easily affected by the subject's subjective emotions, cognitive biases, and social desirability effects, lacking objectivity and stability. In addition, subjective scales cannot capture neurophysiological changes related to IGD, such as electroencephalogram (EEG) features, functional magnetic resonance imaging (fMRI) signals, etc., resulting in limitations in the scientificity and accuracy of the determination results.

[0003] Among the intervention means for IGD, transcranial direct current stimulation (tDCS), as a non-invasive brain stimulation technique, shows certain potential, but its intervention effect lacks specific targets. Existing tDCS intervention programs are mostly based on general brain region stimulation, such as the prefrontal cortex, without personalized regulation for the specific neural circuit abnormalities of IGD patients. This non-specific intervention method may lead to unstable treatment effects and even adverse reactions in some patients. In addition, existing research mainly focuses on short-term effect evaluation, lacking long-term follow-up data, and it is difficult to comprehensively evaluate the persistence and effectiveness of the intervention.

[0004] In addition, the prior art has not realized a full-process closed-loop from the determination and prediction of Internet gaming addiction to the evaluation of the intervention effect.

[0005] Therefore, there is an urgent need for a method and system for identifying, predicting, and evaluating the intervention effect of Internet gaming disorder based on electroencephalogram features and machine learning models to improve the accuracy of IGD identification, the reliability of prediction, and the pertinence of intervention. Summary of the Invention

[0006] The present invention provides a method and system for identifying, predicting, and evaluating the intervention effect of Internet gaming disorder based on electroencephalogram features and machine learning models to solve the defect that the prior art cannot accurately identify, predict, and evaluate the intervention effect of Internet gaming disorder.

[0007] A method for constructing an Internet gaming disorder identification model provided by the present invention includes:

[0008] Obtain the electroencephalogram (EEG) data of the target group in the resting state and the EEG data in the game cue response task, where the target group includes the online game addiction group, the recreational user group, and the healthy control group;

[0009] According to the EEG data of the target group in the resting state and the EEG data in the game cue response task, using the first machine learning algorithm, let the model learn the EEG data of the online game addiction group, the recreational user group, and the healthy control group in the resting state and in the game cue response task, and construct an online game disorder recognition model.

[0010] According to a method for constructing an online game disorder recognition model provided by the present invention, the EEG data of the target group in the resting state includes the absolute power spectral density values of Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 13 Hz), and Beta (13 - 30 Hz) frequency bands at 31 electrode points of the whole brain of the target group in the resting state.

[0011] According to a method for constructing an online game disorder recognition model provided by the present invention, the EEG data of the target group in the game cue response task includes the power spectral density values of Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 12 Hz), and Beta (12 - 30 Hz) frequency bands in the time window of the P300 component of the EEG at 8 electrode points in the parieto-occipital lobe of the target group in the game cue response task (300 - 500 ms).

[0012] According to a method for constructing an online game disorder recognition model provided by the present invention, the game cue response task is: the purpose is to test the craving level of the subject for game-related cues. In the task, a series of pictures are displayed on the screen. The content of the pictures involves online games or other online activities. A scene description of the picture will appear below the picture. The subject is required to imagine how much they want to play the game when the content in the picture appears in front of them in real life after observing the picture and reading the description. After the pictures are displayed for a certain period of time, it enters the scoring session. The subject needs to perform a visual analogue scale scoring of game craving (the score ranges from low to high, indicating the increasing degree of the desire to play the game) for each picture to represent their craving level for playing the game after browsing the pictures.

[0013] According to a method for constructing an online game disorder recognition model provided by the present invention, the step of constructing an online game disorder recognition model by using the first machine learning algorithm to let the model learn the EEG data of the online game addiction group, the recreational user group, and the healthy control group in the resting state and in the game cue response task according to the EEG data of the target group in the resting state and the EEG data in the game cue response task includes:

[0014] When training the online gaming disorder recognition model, the linear kernel support vector machine (SVM) is used as the first machine learning algorithm, and the leave-one-out cross-validation method is used for model training and evaluation. In each round of cross-validation, feature selection is performed by recursive feature elimination method, and the EEG data of the top 80% of the electrode points with the highest contribution rate are retained. At the same time, the regularization parameter C is optimized by the grid search method to optimize the model performance and avoid overfitting, and finally a high-precision online gaming disorder recognition model is obtained.

[0015] The present invention also provides a method for constructing an online game desire prediction model, comprising:

[0016] Obtaining EEG data of the online game addicts during a game cue response task and online game craving score data after a preset period of time after the game cue response task;

[0017] Based on the EEG data of the online game addict group under the game cue response task and the online game craving score data after a preset time period after the game cue response task, a second machine learning algorithm is used to enable the model to learn the relationship between the EEG data of the online game addict group under the game cue response task and the online game craving score data after a preset time period after the game cue response task, so as to construct a prediction model for the degree of online game craving.

[0018] According to a method for constructing a network game craving prediction model provided by the present invention, the EEG data of a group of network game addicts under a game cue response task includes the power spectral density value of the Alpha frequency band of the brain CP2 electrode point of the group of network game addicts under the game cue response task.

[0019] According to a method for constructing a network game craving prediction model provided by the present invention, the method uses a second machine learning algorithm to learn the relationship between the EEG data of the network game addicts under the game clue reaction task and the network game craving score data after a preset time period after the game clue reaction task, so as to construct a network game craving prediction model, including:

[0020] When training the online game desire prediction model, support vector regression (SVR) is used as the second machine learning algorithm, elastic network is used for regularization, leave-one-out cross-validation method is used for model training and evaluation, and the regularization parameter C is optimized through grid search method to optimize model performance and improve generalization ability, finally obtaining a high-precision online game desire prediction model.

[0021] The present invention also provides a method for constructing an evaluation model for the intervention effect of online game addiction, including:

[0022] Obtaining the electroencephalogram data of the online game addiction group before and after transcranial direct current stimulation (tDCS) intervention under the game cue response task, and the online game craving score data of the online game addiction group after a preset time period after the tDCS intervention;

[0023] According to the electroencephalogram data of the online game addiction group before and after tDCS intervention under the game cue response task and the online game craving score data of the online game addiction group after a preset time period after the tDCS intervention, using a third machine learning algorithm, enabling the model to learn the relationship between the electroencephalogram data of the online game addiction group before and after tDCS intervention under the game cue response task and the online game craving score data of the online game addiction group after a preset time period after the tDCS intervention, and constructing an evaluation model for the intervention effect of online game addiction.

[0024] According to the method for constructing an evaluation model for the intervention effect of online game addiction provided by the present invention, the electroencephalogram data of the online game addiction group before and after tDCS intervention under the game cue response task includes the average amplitude difference of the P300 electroencephalogram component at the Pz electrode point of the brain of the online game addiction group before and after tDCS intervention under the game cue response task.

[0025] According to the method for constructing an evaluation model for the intervention effect of online game addiction provided by the present invention, the step of using a third machine learning algorithm according to the electroencephalogram data of the online game addiction group before and after tDCS intervention under the game cue response task and the online game craving score data of the online game addiction group after a preset time period after the tDCS intervention, enabling the model to learn the relationship between the electroencephalogram data of the online game addiction group before and after tDCS intervention under the game cue response task and the online game craving score data of the online game addiction group after a preset time period after the tDCS intervention, and constructing an evaluation model for the intervention effect of online game addiction includes:

[0026] When training the evaluation model for the intervention effect of online game addiction, the support vector machine regression method is used as the third machine learning algorithm, elastic net is used for regularization, the leave-one-out cross-validation method is used for model training and evaluation, and at the same time, the grid search method is used to optimize the regularization parameters, and finally a prediction model for the degree of online game craving is obtained.

[0027] A method for constructing an evaluation model for the intervention effect of online game addiction according to the present invention, obtaining electroencephalogram data of an online game addict group before and after transcranial direct current stimulation intervention in a game cue response task and online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention, including:

[0028] Receiving electroencephalogram data of an online game addict group before and after transcranial direct current stimulation intervention in a game cue response task from an electroencephalogram device (such as an electroencephalogram cap);

[0029] Receiving online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention from at least one terminal.

[0030] The present invention also provides an online game disorder identification system, including:

[0031] A first receiving module, configured to: receive electroencephalogram data of a person to be tested in a resting state and electroencephalogram data of the person to be tested in a game cue response task from at least one terminal, wherein the electroencephalogram data of the person to be tested in the resting state at least includes any one of the absolute power spectral density values of Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 13 Hz), and Beta (13 - 30 Hz) frequency bands at 31 electrode points of the whole brain of the person to be tested in the resting state, and the electroencephalogram data of the person to be tested in the game cue response task at least includes any one of the power spectral density values of Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 12 Hz), and Beta (12 - 30 Hz) frequency bands in the time window of the electroencephalogram P300 component (300 - 500 ms) at 8 electrode points of the parieto-occipital lobe of the person to be tested in the game cue response task;

[0032] An identification module, configured to: obtain an online game disorder identification result of the person to be tested through the online game disorder identification model obtained by the method for constructing an online game disorder identification model described in any one of the above through the electroencephalogram data of the person to be tested in the resting state and the electroencephalogram data of the person to be tested in the game cue response task;

[0033] A first output module, configured to: output the online game disorder identification result of the person to be tested to at least one terminal.

[0034] The present invention also provides an online game craving degree prediction system, including:

[0035] A second receiving module, configured to: receive, from at least one terminal, electroencephalogram data of a subject under a game clue response task, where the subject is a network game addict, and the electroencephalogram data of the subject under the game clue response task includes the power spectral density value of the Alpha frequency band at the CP2 electrode point of the subject's brain under the game clue response task;

[0036] A prediction module, configured to: obtain a predicted result of the subject's network game craving degree after a preset time period after the game clue response task, according to the electroencephalogram data of the subject under the game clue response task, by using the network game craving degree prediction model constructed by the construction method of the network game craving degree prediction model described in any one of the above;

[0037] A second output module, configured to: output the predicted result of the subject's network game craving degree after a preset time period after the game clue response task to at least one terminal.

[0038] The present invention further provides a network game addiction intervention effect evaluation system, including:

[0039] A third receiving module, configured to: receive, from at least one terminal, electroencephalogram data of a subject under a game clue response task before and after transcranial direct current stimulation intervention, where the subject is a network game addict, and the electroencephalogram data of the subject under the game clue response task before and after transcranial direct current stimulation intervention includes the average amplitude difference of the P300 electroencephalogram component at the Pz electrode point of the subject's brain under the game clue response task before and after transcranial direct current stimulation intervention;

[0040] An evaluation module, configured to: obtain an evaluation result of the network game addiction intervention effect of the subject after a preset time period after the transcranial direct current stimulation intervention, according to the electroencephalogram data of the subject under the game clue response task before and after transcranial direct current stimulation intervention, by using the network game addiction intervention effect evaluation model constructed by the construction method of the network game addiction intervention effect evaluation model described in any one of the above;

[0041] A third output module, configured to: output the evaluation result of the network game addiction intervention effect of the subject after a preset time period after the transcranial direct current stimulation intervention to at least one terminal.

[0042] The present invention further provides a comprehensive network game addiction evaluation system, which is obtained by integrating at least two of the above network game disorder identification system, network game craving degree prediction system, and network game addiction intervention effect evaluation system.

[0043] It should be noted that a terminal refers to an input / output device connected to a computer system. According to different functions, terminals can be divided into various types: smart terminals or intelligent terminals, dumb terminals, interactive terminals or online terminals. Specifically, a terminal can be various mobile communication devices, such as mobile phones, tablets, etc. This article aims to provide users with the functions of inputting data and obtaining data output.

[0044] The present invention also provides an electronic device, including a processor and a memory storing a computer program. When the processor executes the computer program, it implements any one or any combination of the above-mentioned methods for constructing a network game disorder recognition model, a method for constructing a network game craving degree prediction model, and a method for evaluating the intervention effect of network game addiction.

[0045] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements any one or any combination of the above-mentioned methods for constructing a network game disorder recognition model, a method for constructing a network game craving degree prediction model, and a method for evaluating the intervention effect of network game addiction.

[0046] The present invention also provides a computer program product. The computer program product includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute any one or any combination of the above-mentioned methods for constructing a network game disorder recognition model, a method for constructing a network game craving degree prediction model, and a method for evaluating the intervention effect of network game addiction.

[0047] The method and system for network game disorder recognition, prediction and intervention effect evaluation based on EEG features and machine learning models provided by the present invention can at least bring the following beneficial effects:

[0048] By acquiring the EEG data of the target groups (including network game addicts group, recreational users group and healthy control group) in the resting state and the game cue response task, and using the first machine learning algorithm to construct an identification model, it can effectively overcome the problems of strong dependence on traditional subjective scales and lack of objectivity. By learning the differences in EEG features of different groups, the network game disorder recognition model can provide an objective and stable judgment basis at the neurophysiological level, significantly improving the accuracy and scientificity of network game disorder recognition.

[0049] By obtaining the electroencephalogram (EEG) data of a group of Internet game addicts during a game cue response task and the craving score data within a preset time period after the task, and using a second machine learning algorithm to construct a prediction model for the degree of Internet game craving, the correlation between EEG features and the degree of craving can be accurately captured. The prediction model for the degree of Internet game craving can predict the degree of craving of Internet game addicts in real time, providing data support for personalized intervention, making up for the lack of objective prediction indicators in the existing technology, and improving the reliability and practicality of prediction.

[0050] By obtaining the EEG data of a group of Internet game addicts before and after transcranial direct current stimulation (tDCS) intervention and the craving score data after the intervention, and using a third machine learning algorithm to construct an evaluation model for the intervention effect of Internet game addiction, the relationship between the changes in EEG features before and after the intervention and the improvement of the degree of craving can be quantified. The evaluation model for the intervention effect of Internet game addiction can not only evaluate the short-term effect of tDCS intervention, but also evaluate the persistence and effectiveness of the intervention through long-term data tracking, solve the problems of lack of specific targets and long-term data support in the evaluation of intervention effects in the existing technology, and improve the pertinence and scientificity of the intervention.

[0051] In summary, a method and system for identifying, predicting, and evaluating the intervention effect of Internet game disorder based on EEG features and machine learning models provided by the present invention realize a full-process closed loop from Internet game disorder identification, craving degree prediction to intervention effect evaluation by combining EEG features and machine learning algorithms, significantly improving the objectivity of Internet game addiction determination, the accuracy of prediction, and the pertinence of intervention, and having important scientific value and application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0053] Figure 1 It is a schematic flow chart of the construction method of the Internet game disorder identification model, the construction method of the prediction model for the degree of Internet game craving, and the construction method of the evaluation model for the intervention effect of Internet game addiction provided by the present invention.

[0054] Figure 2 It is an example diagram of a game cue response task.

[0055] Figure 3 It is a schematic flow chart of transcranial direct current stimulation intervention.

[0056] Figure 4Schematic structural diagrams of the online game disorder recognition system, online game craving degree prediction system, and online game addiction intervention effect evaluation system provided by the present invention.

[0057] Figure 5 Schematic structural diagram of the electronic device provided by the present invention. Detailed implementation manners

[0058] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention, and they should not be construed as limiting the present invention. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. In the description of the present invention, it should be understood that the terms used are only for the purpose of description and cannot be construed as indicating or implying relative importance.

[0059] See Figure 1 , a method for online game disorder recognition, prediction, and intervention effect evaluation based on electroencephalogram features and machine learning models provided by the present invention actually includes a method for constructing an online game disorder recognition model, a method for constructing an online game craving degree prediction model, and a method for constructing an online game addiction intervention effect evaluation model.

[0060] The execution subject of any one of the methods for online game disorder recognition, prediction, and intervention effect evaluation based on electroencephalogram features and machine learning models provided by the present invention can be any applicable terminal-side device or network-side device, such as an online game disorder recognition device, an online game craving degree prediction device, an online game addiction intervention effect evaluation device, etc.

[0061] A method for constructing an online game disorder recognition model provided by the present invention may include:

[0062] S1. Obtain the electroencephalogram (EEG) data of the target group in the resting state and the EEG data of the target group in the game cue response task. The target group includes an online game addiction group, a recreational user group, and a healthy control group. The EEG data of the target group in the resting state includes the absolute power spectral density values of the Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 13 Hz), and Beta (13 - 30 Hz) frequency bands at 31 electrode points in the whole brain of the target group in the resting state. The EEG data of the target group in the game cue response task includes the power spectral density values of the Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 12 Hz), and Beta (12 - 30 Hz) frequency bands in the time window of the P300 component of the EEG at 8 electrode points in the parieto-occipital lobe of the target group in the game cue response task (300 - 500 ms).

[0063] In one embodiment, the EEG data of the target group can be measured by medical staff for the target group.

[0064] For example, the EEG data can be measured by a 32-channel Smarting Pro (Mbraintrain, Serbia) EEG device, which follows the 10 - 20 system. The EEG cap uses Ag / AgCl electrodes with an electrode radius of 1 cm. Before wearing the EEG cap, the subject needs to wash and dry their hair first. After ensuring that the scalp is cooled, then wear the EEG cap. By locating the Cz electrode and bilateral mastoids, the correct position of the EEG cap is determined. Subsequently, GreenTek GT5 medical conductive paste is injected into each electrode channel to ensure good contact between the electrode and the scalp and reduce the impedance. When the impedance of all electrode channels drops below 5 kΩ, the acquisition and recording of EEG can begin. The EEG signal is amplified in the bandpass range of 0 - 125 Hz and is digitally processed online at a sampling rate of 500 Hz. The EEG acquisition is carried out in a soundproof room, and a computer screen and keyboard for the response task are equipped in front of the subject.

[0065] The EEG acquisition includes two modes: the resting state and the task state. During the resting state EEG acquisition, the subject needs to sit quietly on the chair in a comfortable position, close their eyes and rest, avoiding eye movements and large body movements to reduce artifacts. During the task state EEG acquisition, first determine the placement position of the subject's key-pressing hand to reduce muscle movement artifacts. Then conduct a practice test to train the subject's blinking pattern during the task, enabling them to learn to blink during the gaps between target stimuli to reduce eye movement artifacts.

[0066] See Figure 2, in one embodiment, the game cue response task is as follows: The purpose is to test the craving level of the subjects for game-related cues. In the task, a series of pictures are displayed on the screen. The content of the pictures involves online games or other online activities (the pictures are collected and sorted out by the researchers in advance, and a certain number of college students are invited to evaluate the picture valence). A scene description of the picture will appear below the picture. The subjects are required to imagine how much they want to play the game when the content in the picture appears in front of them in real life after observing the picture and reading the description. After the pictures are displayed for a certain period of time, the scoring session begins. The subjects need to conduct a visual analogue scale scoring of game craving (the scores range from low to high, indicating the increasing degree of the desire to play the game) for each picture to represent their craving degree for playing the game after browsing the pictures. A total of 30 game cue pictures and 30 neutral pictures (pictures of activities other than games using mobile phones) are displayed.

[0067] S2. According to the electroencephalogram data of the target group in the resting state and the electroencephalogram data in the game cue response task, using the first machine learning algorithm, let the model learn the electroencephalogram data of the online game addiction group, the recreational user group, and the healthy control group in the resting state and in the game cue response task, and construct an online game disorder recognition model.

[0068] In one embodiment, when training the online game disorder recognition model, a linear kernel support vector machine (SVM) is used as the first machine learning algorithm, and the leave-one-out cross-validation method is used for model training and evaluation. In each round of cross-validation, feature selection is performed by the recursive feature elimination method, and the electroencephalogram data of the electrode points with the top 80% contribution rate is retained. At the same time, the regularization parameter C is optimized by the grid search method to optimize the model performance and avoid overfitting, and finally a high-precision online game disorder recognition model is obtained.

[0069] A method for constructing an online game craving degree prediction model provided by the present invention may include:

[0070] S’1. Obtain the electroencephalogram data of the online game addiction group in the game cue response task and the online game craving score data after a preset time period (such as 6 months) after the game cue response task. Among them, the electroencephalogram data of the online game addiction group in the game cue response task includes the power spectral density value of the Alpha band of the CP2 electrode point of the brain of the online game addiction group in the game cue response task.

[0071] In one embodiment, the online game craving score data of online game addicts after a preset time period after the game cue response task can be obtained by having the online game addicts use a game craving degree assessment scale after a preset time period after the game cue response task. The Game Craving Degree Assessment Scale (QGU-B) is a short version scale used to evaluate an individual's craving degree for online games and is used to measure the current craving degree of the subject for online games. The QGU-B scale contains 10 items, and its intensity is represented by a score from 1 to 7. The higher the score, the higher the game craving degree. Among them, "1" represents "not craving at all", and "7" represents "craving very much". The higher the scale score, the higher the craving degree of the subject for online games.

[0072] S’2. According to the electroencephalogram data of the online game addict group under the game cue response task and the online game craving score data after a preset time period after the game cue response task, using the second machine learning algorithm, let the model learn the relationship between the electroencephalogram data of the online game addict group under the game cue response task and the online game craving score data after a preset time period after the game cue response task, and construct an online game craving degree prediction model.

[0073] In one embodiment, when training the online game craving degree prediction model, the support vector regression (SVR) method is used as the second machine learning algorithm, elastic net is used for regularization, the leave-one-out cross-validation method is used for model training and evaluation, and at the same time, the grid search method is used to optimize the regularization parameter C to optimize the model performance and improve the generalization ability, and finally an online game craving degree prediction model with high accuracy is obtained.

[0074] A method for constructing an online game addiction intervention effect evaluation model provided by the present invention may include:

[0075] S”1. Obtain the electroencephalogram data of the online game addict group under the game cue response task before and after transcranial direct current stimulation intervention and the online game craving score data of the online game addict group after a preset time period (such as 1 week) after transcranial direct current stimulation intervention. Among them, the electroencephalogram data of the online game addict group under the game cue response task before and after transcranial direct current stimulation intervention includes the average amplitude difference of the P300 electroencephalogram component at the Pz electrode point of the brain under the game cue response task before and after transcranial direct current stimulation intervention of the online game addict group.

[0076] S”2. Based on the electroencephalogram data of the online game addict group before and after transcranial direct current stimulation (tDCS) intervention in the game cue response task, and the online game craving score data of the online game addict group after a preset period of time after tDCS intervention, using the third machine learning algorithm, the model is made to learn the relationship between the electroencephalogram data of the online game addict group before and after tDCS intervention in the game cue response task and the online game craving score data of the online game addict group after a preset period of time after tDCS intervention, and an online game addiction intervention effect evaluation model is constructed.

[0077] See Figure 4 , in one embodiment, the electroencephalogram data of the online game addict group before and after tDCS intervention in the game cue response task can be measured by medical staff using a tDCS device (such as a Neurocon device) to perform tDCS intervention on the online game addict group, and making the online game addict group wear an electroencephalogram device (such as an electroencephalogram cap) and perform the game cue response task before and after the intervention.

[0078] The specific process of tDCS intervention can be:

[0079] Place a 5*5 cm square electrode patch of the tDCS device on the scalp of the subject corresponding to the central parietal lobe brain region (Pz electrode point, 10-20 electroencephalogram system), and place another 5*5 cm square electrode patch of the tDCS device on the right dorsal trapezius muscle area of the subject's scalp. Apply cathodal stimulation to the central parietal lobe brain region and anodal stimulation to the right dorsal trapezius muscle area. In the first 10 seconds of the stimulation, the current magnitude rises from 0 to 1.5 mA at a slope, then maintains a magnitude of 1.5 mA and continues to stimulate for 19 minutes and 40 seconds, and in the last 10 seconds, the current magnitude drops from 1.5 mA to 0 mA at a slope. The whole process lasts for 20 minutes. Before and after the intervention, the subject is respectively made to perform the same game cue response task. 30 pictures related to online games will be shown on the screen, each picture appears 6 times, randomly. A scene description of the picture will appear below the picture. After the picture is shown for a certain time, it will enter the scoring session. 0 to 9 points (the scores increase from low to high, indicating an increase in the "desire to play" degree of the game) represent the craving degree for playing the game after browsing the picture, so as to obtain the online game craving score data. Among them, the cue picture is shown for 3 s and the scoring time is 4 s. The intervention is carried out twice a day, and before and after each day's intervention, the subject needs to complete the game cue response task and the game craving score (the score can be obtained using the game craving scale (VAS, QGU-B)).

[0080] When training the evaluation model for the intervention effect of online game addiction, the support vector machine regression method is used as the third machine learning algorithm, elastic net is used for regularization, the leave-one-out cross-validation method is used for model training and evaluation, and the regularization parameters are optimized by the grid search method. Finally, the prediction model for the degree of online game craving is obtained.

[0081] The present invention verifies the construction methods of the online game disorder recognition model, the prediction model for the degree of online game craving, and the evaluation model for the intervention effect of online game addiction provided above through four specific embodiments.

[0082] Example 1: 75 samples were collected from the medical database in Tianjin (25 cases in the IGD group, 22 cases in the entertainment group, and 28 cases in the healthy group). During the training of the online game disorder recognition model, the results showed that the energy of each frequency band of the P300 in the parieto-occipital lobe under game cues could better identify the IGD group (accuracy rate = 81.1%).

[0083] Example 2: 46 samples were collected from the medical databases in Beijing and Tianjin (23 cases in the IGD group and 23 cases in the healthy group). During the training of the online game disorder recognition model, the results showed that the energy of each frequency band of the P300 in the parieto-occipital lobe under game cues could better identify the IGD group (accuracy rate = 78.3%).

[0084] Example 3: 25 IGD patients were collected from the medical database in Tianjin. During the training of the prediction model for the degree of online game craving, the Alpha band energy at the CP2 point of the brain under game cues was used to predict the QGU-B craving score after six months. The results showed that the predicted score was significantly correlated with the actual score (r = 0.45, p = 0.024).

[0085] Example 4: 50 IGD patients were given tDCS intervention targeting the Pz electrode point under the game cue task (25 cases in the real stimulation group). After one week of intervention, the decrease in the P300 amplitude in the real stimulation group was significantly positively correlated with the decrease in the degree of craving (r = 0.56, p = 0.006).

[0086] The present invention believes that the neural electrical activity in the parietal cortex of the brain under game cue exposure is a neuroelectrophysiological marker for distinguishing online game disorder from the control group and predicting game craving. Targeted cathodal transcranial direct current stimulation intervention on the parietal cortex of the brain of IGD patients under game cues can effectively reduce the craving degree and improve the cognitive ability of IGD patients in the real stimulation group by inhibiting the neural electrical activity in the parietal cortex induced by game cues. Moreover, the neural oscillations related to the parietal cortex of the brain induced by game cues can predict the degree of craving reduction after a period of time (such as 1 month) after transcranial direct current stimulation intervention. The following is the verification process.

[0087] 1. Subjects

[0088] (1) Inclusion criteria: 1) Age range: 18 - 24 years old; 2) Familiar with and mainly play the online game "Honor of Kings"; 3) Addiction group: Meeting at least six DSM-5 diagnostic criteria for Internet gaming disorder and playing online games at least 14 hours per week for the past 1 year until now; 4) Recreational use group: Not meeting the DSM-5 diagnostic criteria and playing online games at least 14 hours per week for the past 1 year until now; 5) Healthy control group: Not meeting the DSM-5 diagnostic criteria and playing online games less than 7 hours per week for the past 1 year until now; 6) Signing the informed consent form and voluntarily participating in this trial.

[0089] (2) Exclusion criteria: 1) Severe mental illness, neurological disorder or physical illness; 2) History of substance abuse / dependence (including alcohol); 3) Subjects who have received Internet gaming disorder-related interventions in the past 1 year; 4) Subjects who have received neuromodulation-related interventions in the past 1 year; 6) Poor compliance: Subjects who the researcher believes are unable to cooperate with the trial.

[0090] 2. Information collection

[0091] The baseline questionnaire scale includes: basic information, gaming use characteristics, severity of gaming addiction, gaming craving scale, alcohol use disorder identification test, nicotine dependence test, depression and anxiety screening scale. Before and after the intervention: gaming craving scale. The follow-up questionnaire scale includes: gaming use characteristics, gaming craving scale.

[0092] 3. EEG scanning

[0093] (1) Resting-state EEG scanning: Requiring the subjects to keep their bodies as still as possible and collecting for 5 minutes with eyes closed.

[0094] (2) Gaming cue reactivity task: Pictures of online games or other online activities (30 positive, negative gaming, and neutral mobile activities each) will be shown on the screen, and a scene description of the picture will appear below the picture. After the picture is shown, a gaming craving degree score (-3 to 3 points) will be given. Among them, the cue picture is shown for 3 s and the scoring time is 4 s.

[0095] 4. tDCS intervention

[0096] Place square electrode patches of 5*5 cm at the positions corresponding to the Pz point in the central parietal lobe and the right dorsal trapezius muscle on the scalp of the subjects, and apply cathodal stimulation to the parietal lobe. True stimulation: 10 seconds before the start of stimulation, the current magnitude rises from 0 to 1.5 mA in a ramp, then remains at 1.5 mA for 19 minutes and 40 seconds, and finally the current magnitude drops from 1.5 mA to 0 mA in a ramp during the last 10 seconds. False stimulation method: 20 seconds before the start of stimulation, the current rises from 0 mA to 1.5 mA in a ramp and then drops back to 0 mA in a ramp, maintains a current magnitude of 0 mA for 19 minutes and 20 seconds, and finally the current rises from 0 mA to 1.5 mA in a ramp and then drops back to 0 mA during the last 20 seconds. The false stimulation is only to make the subjects feel a slight tingling at the beginning and end without any intervention effect.

[0097] During the intervention, a game cue response task is carried out. 30 game pictures will be shown on the screen, and a scene description of the picture will appear below the picture. After the pictures are shown for a certain time, it will enter the scoring session. 0 to 9 points represent the degree of craving for the game after seeing the pictures. The picture display time is 3 s, and the scoring time is 4 s. The subjects will undergo the intervention for 2 days, once a day, and need to complete the craving scale for the game before and after the intervention.

[0098] 5. Experimental procedure

[0099] First, collect the electroencephalogram (EEG) changes of online game addicts, recreational game users, and healthy controls at rest and during cue exposure, and conduct follow-up on addiction severity and craving. Extract the time-domain and frequency-domain features of the EEG for machine learning to find the changes in specific cerebral cortical neural electrical activities that can distinguish IGD from the control group, and at the same time further determine the association between its specific indicators and craving. Further conduct tDCS intervention research on the found specific cerebral cortical changes, adopt a randomized controlled, double-blind intervention during cue exposure, explore whether it can reduce the craving of addicts and improve cognitive ability, and further find its impact on cerebral neuroelectrophysiological activities.

[0100] 6. Statistical methods

[0101] The questionnaire data was statistically analyzed using GraphPad prism 9 software. One-way ANOVA was used for behavioral and EEG data to determine the baseline and behavioral indicators with differences among the three groups. The collected EEG data was filtered, downsampled, and artifact-removed using EEGLAB in Matlab 2021b, and then time-domain, frequency-domain, and time-frequency analyses were performed. Support vector machine model was used for machine learning, and leave-one-out cross-validation was adopted for the verification method. Feature screening was carried out, and a model was established for the Internet gaming addiction group and the other two groups. For the electrode points with the highest contribution rate in the model, the relevant EEG indicators were extracted for Pearson correlation analysis with craving. Repeated measures two-way ANOVA was used for the between-group comparison of the data before and after intervention and during follow-up. The treatment method was used as the between-group factor, and time was used as the within-group factor. Bonferroni test was selected for post hoc test, and a two-sided p < 0.05 was used as the significance criterion for differences.

[0102] 7. Research Results

[0103] First, the game use characteristics and EEG indicators of 25 Internet gaming addicts, 22 recreational game users, and 28 healthy controls were collected. The results showed that under the positive game cue task, the craving level of IGD patients increased significantly, and an obvious EEG P300 component appeared in the parieto-occipital lobe. By analyzing the energy of each frequency band in the parieto-occipital lobe through machine learning, it was found that it could effectively distinguish IGD patients from the control group (accuracy > 80%), and the Pz electrode point in the central parietal lobe played a major role, with a significant increase in its average P300 amplitude. The energy of the Delta, Theta, and Alpha frequency bands was significantly positively correlated with the current craving of IGD patients.

[0104] Based on the above findings, further tDCS intervention was carried out on the changes in the cortical neural electrical activity of the Pz electrode point under cue exposure. A randomized controlled, double-blind experimental design was adopted. The cathode of tDCS was located at the Pz electrode point, and the anode was placed on the latissimus dorsi muscle of the back. The intervention was carried out for 2 days, 20 minutes each day. A total of 50 Internet gaming addicts were included in the study, with 25 people in the real intervention group and 25 people in the sham intervention group, and finally 23 people were included in the analysis for each group. The results showed that compared with the sham intervention group, the craving level and game use time of the real intervention group were significantly reduced within 1 - 4 weeks after intervention; under the condition of game cue exposure, the craving level of the real intervention group was also significantly reduced, and the average amplitude of the P300 component and the energy of the Delta, Theta, and Alpha frequency bands at the Pz electrode point were significantly reduced, and the effect was most significant 1 week after intervention.

[0105] In summary, the specific EEG changes under game cue exposure are important indicators for identifying IGD. The P300 component in the parietal lobe is significantly enhanced and is related to current craving, while the N400 components in the prefrontal and central regions are reduced and can predict future craving levels. Cathode intervention of tDCS targeting the Pz electrode point under cues can significantly reduce the craving and game time of IGD patients, and at the same time affect their EEG activities, providing new ideas and methods for the diagnosis and intervention of IGD.

[0106] The method and system for identifying, predicting and evaluating the intervention effect of online game disorder based on EEG features and machine learning models provided by the present invention can at least bring the following beneficial effects:

[0107] By acquiring the EEG data of the target groups (including online game addicts group, recreational users group and healthy control group) in the resting state and under the game cue response task, and using the first machine learning algorithm to construct an identification model, it can effectively overcome the problems of strong dependence on traditional subjective scales and lack of objectivity. The online game disorder identification model can provide an objective and stable judgment basis from the neurophysiological level by learning the EEG feature differences of different groups, and significantly improve the accuracy and scientificity of online game disorder identification.

[0108] By acquiring the EEG data of the online game addicts group under the game cue response task and the craving score data within a preset time period after the task, and using the second machine learning algorithm to construct an online game craving degree prediction model, it can accurately capture the correlation between EEG features and craving degree. The online game craving degree prediction model can real-time predict the craving degree of online game addicts, provide data support for personalized intervention, make up for the lack of objective prediction indicators in the existing technology, and improve the reliability and practicality of prediction.

[0109] By acquiring the EEG data of the online game addicts group before and after transcranial direct current stimulation (tDCS) intervention and the craving score data after the intervention, and using the third machine learning algorithm to construct an online game addiction intervention effect evaluation model, it can quantify the relationship between the EEG feature changes before and after the intervention and the improvement of craving degree. The online game addiction intervention effect evaluation model can not only evaluate the short-term effect of tDCS intervention, but also evaluate the persistence and effectiveness of the intervention through long-term data tracking, solve the problems of lack of specific targets and long-term data support in the evaluation of intervention effect in the existing technology, and improve the pertinence and scientificity of the intervention.

[0110] In summary, the method and system for identifying, predicting, and evaluating the intervention effect of Internet game disorder based on EEG features and machine learning models provided by the present invention achieve a full-process closed-loop from Internet game disorder identification, craving degree prediction to intervention effect evaluation by combining EEG features and machine learning algorithms, significantly improving the objectivity of Internet game addiction determination, the accuracy of prediction, and the pertinence of intervention, and having important scientific value and application prospects.

[0111] See Figure 4 , the present invention also provides an Internet game disorder identification system, an Internet game craving degree prediction system, and an Internet game addiction intervention effect evaluation system. Integrating any two or more of the three can form an Internet game addiction comprehensive evaluation system to achieve a full-process closed-loop from Internet game disorder identification, craving degree prediction to intervention effect evaluation.

[0112] Figure 5 An example of the physical structure diagram of an electronic device is shown in Figure 5 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840. Among them, the processor 810, the communication interface 820, and the memory 830 complete communication with each other through the communication bus 840. The processor 810 can call the logical instructions in the memory 830 to execute the steps in any one of the solutions A, B, C or any combination of them.

[0113] In addition, when the logical instructions in the above-mentioned memory 830 are implemented in the form of a software functional unit and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs, Read-Only Memories), random access memories (RAMs, Random Access Memories), magnetic disks, or optical discs that can store program codes.

[0114] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the steps in any one of the execution schemes A, B, and C or any combination thereof.

[0115] In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in any one of the execution schemes A, B, and C or any combination thereof are implemented.

[0116] Scheme A:

[0117] Receive the electroencephalogram data of the subject in the resting state and the electroencephalogram data of the subject in the game cue response task from at least one terminal. Among them, the electroencephalogram data of the subject in the resting state at least includes any one of the absolute power spectral density values of Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 13 Hz), and Beta (13 - 30 Hz) in 31 electrode points of the whole brain of the subject in the resting state. And the electroencephalogram data of the subject in the game cue response task at least includes any one of the power spectral density values of Delta (1 - 4 Hz), Theta (4 - 8 Hz), Alpha (8 - 12 Hz), and Beta (12 - 30 Hz) in the time window of the P300 component of the electroencephalogram of 8 electrode points in the parieto-occipital lobe of the subject in the game cue response task (300 - 500 ms);

[0118] According to the electroencephalogram data of the subject in the resting state and the electroencephalogram data of the subject in the game cue response task, obtain the online game disorder recognition result of the subject through the online game disorder recognition model construction method described in any one of the above;

[0119] Output the online game disorder recognition result of the subject to at least one terminal;

[0120] Scheme B:

[0121] Receive the electroencephalogram data of the subject in the game cue response task from at least one terminal. Among them, the subject is an online game addict, and the electroencephalogram data of the subject in the game cue response task includes the power spectral density value of the Alpha frequency band at the CP2 electrode point of the subject's brain in the game cue response task;

[0122] Based on the EEG data of the person to be tested under the game clue response task, the online game craving degree prediction model obtained by the construction method of the online game craving degree prediction model described in any one of the above, obtain the online game craving degree prediction result of the person to be tested after a preset time period after the game clue response task;

[0123] Output the online game craving degree prediction result of the person to be tested after a preset time period after the game clue response task to at least one terminal;

[0124] Solution C:

[0125] Receive the EEG data of the person to be tested under the game clue response task before and after transcranial direct current stimulation intervention from at least one terminal, where the person to be tested is an online game addict, and the EEG data of the person to be tested under the game clue response task before and after transcranial direct current stimulation intervention includes the average amplitude difference of the P300 EEG component at the Pz electrode point of the brain of the person to be tested under the game clue response task before and after transcranial direct current stimulation intervention;

[0126] Based on the EEG data of the person to be tested under the game clue response task before and after transcranial direct current stimulation intervention, the online game addiction intervention effect evaluation model obtained by the construction method of the online game addiction intervention effect evaluation model described in any one of the above, obtain the online game addiction intervention effect evaluation result of the person to be tested after a preset time period after transcranial direct current stimulation intervention;

[0127] Output the online game addiction intervention effect evaluation result of the person to be tested after a preset time period after transcranial direct current stimulation intervention to at least one terminal.

[0128] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement without creative labor.

[0129] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for constructing an online game disorder recognition model, characterized in that Including: Obtain the electroencephalogram (EEG) data of the target group in the resting state and the EEG data in the game cue response task, where the target group includes an online game addiction group, a recreational user group, and a healthy control group; According to the EEG data of the target group in the resting state and the EEG data in the game cue response task, use the first machine learning algorithm to enable the model to learn the EEG data of the online game addiction group, the recreational user group, and the healthy control group in the resting state and in the game cue response task, and construct an online game disorder recognition model.

2. The method for constructing the online game disorder recognition model according to claim 1, wherein The EEG data of the target group in the resting state includes the absolute power spectral density values of the Delta, Theta, Alpha, and Beta frequency bands at 31 electrode points of the whole brain of the target group in the resting state; The EEG data of the target group in the game cue response task includes the power spectral density values of the Delta, Theta, Alpha, and Beta frequency bands in the time window of the P300 component of the EEG at 8 electrode points in the parieto-occipital lobe of the target group in the game cue response task; The step of, according to the EEG data of the target group in the resting state and the EEG data in the game cue response task, using the first machine learning algorithm to enable the model to learn the EEG data of the online game addiction group, the recreational user group, and the healthy control group in the resting state and in the game cue response task, and constructing an online game disorder recognition model, includes: When training the online game disorder recognition model, use the linear kernel support vector machine as the first machine learning algorithm, and use the leave-one-out cross-validation method for model training and evaluation. In each round of cross-validation, perform feature selection through the recursive feature elimination method, retain the EEG data of the electrode points whose contribution rate meets the preset requirements, and at the same time optimize the regularization parameter through the grid search method to finally obtain the online game disorder recognition model.

3. A method for constructing a prediction model of the craving degree of an online game, characterized in that, Including: Obtain the EEG data of the online game addiction group in the game cue response task and the online game craving score data after a preset time period after the game cue response task; According to the EEG data of the online game addiction group in the game cue response task and the online game craving score data after a preset time period after the game cue response task, use the second machine learning algorithm to enable the model to learn the relationship between the EEG data of the online game addiction group in the game cue response task and the online game craving score data after a preset time period after the game cue response task, and construct an online game craving degree prediction model.

4. The method for constructing a prediction model for the craving degree of online games according to claim 3, wherein The EEG data of the online game addiction group in the game cue response task includes the power spectral density value of the Alpha frequency band at the CP2 electrode point of the brain of the online game addiction group in the game cue response task; Based on the electroencephalogram data of the online game addict group during the game cue response task and the online game craving score data after a preset time period after the game cue response task, using the second machine learning algorithm, making the model learn the relationship between the electroencephalogram data of the online game addict group during the game cue response task and the online game craving score data after a preset time period after the game cue response task, a prediction model for the degree of online game craving is constructed, including: When training the prediction model for the degree of online game craving, the support vector machine regression method is used as the second machine learning algorithm, elastic net is used for regularization, the leave-one-out cross-validation method is used for model training and evaluation, and at the same time, the grid search method is used to optimize the regularization parameters, and finally a prediction model for the degree of online game craving is obtained.

5. A method for constructing an evaluation model for the intervention effect of online game addiction, characterized in that, Including: Obtain the electroencephalogram data of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention and the online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention; Based on the electroencephalogram data of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention and the online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention, using the third machine learning algorithm, making the model learn the relationship between the electroencephalogram data of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention and the online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention, a model for evaluating the effect of online game addiction intervention is constructed.

6. The method for constructing an evaluation model for the intervention effect of online game addiction, as claimed in claim 5, is characterized in that The electroencephalogram data of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention includes the average amplitude difference of the P300 electroencephalogram component at the Pz electrode point of the brain of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention; Based on the electroencephalogram data of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention and the online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention, using the third machine learning algorithm, making the model learn the relationship between the electroencephalogram data of the online game addict group during the game cue response task before and after transcranial direct current stimulation intervention and the online game craving score data of the online game addict group after a preset time period after transcranial direct current stimulation intervention, a model for evaluating the effect of online game addiction intervention is constructed, including: When training the model for evaluating the effect of online game addiction intervention, the support vector machine regression method is used as the third machine learning algorithm, elastic net is used for regularization, the leave-one-out cross-validation method is used for model training and evaluation, and at the same time, the grid search method is used to optimize the regularization parameters, and finally a prediction model for the degree of online game craving is obtained.

7. An online game disorder recognition system, characterized in that, Including: A first receiving module, configured to: receive, from at least one terminal, electroencephalogram data of a subject in a resting state and electroencephalogram data of the subject in a game cue response task, wherein the electroencephalogram data of the subject in the resting state includes at least any one of the absolute power spectral density values of the Delta, Theta, Alpha, and Beta frequency bands at 31 electrode points of the whole brain of the subject in the resting state, and the electroencephalogram data of the subject in the game cue response task includes at least any one of the power spectral density values of the Delta, Theta, Alpha, and Beta frequency bands in the time window of the P300 brain electrical component at 8 electrode points of the parieto-occipital lobe of the subject in the game cue response task; An identification module, configured to: obtain an online game disorder identification result of the subject through an online game disorder identification model obtained by the method for constructing an online game disorder identification model according to claim 1 or 2, based on the electroencephalogram data of the subject in the resting state and the electroencephalogram data of the subject in the game cue response task; A first output module, configured to: output the online game disorder identification result of the subject to at least one terminal.

8. A prediction system for the degree of craving for an online game, characterized in that, Comprising: A second receiving module, configured to: receive, from at least one terminal, electroencephalogram data of a subject in a game cue response task, wherein the subject is an online game addict, and the electroencephalogram data of the subject in the game cue response task includes the power spectral density value of the Alpha frequency band at the CP2 electrode point of the brain of the subject in the game cue response task; A prediction module, configured to: obtain a prediction result of the online game craving degree of the subject after a preset time period after the game cue response task through an online game craving degree prediction model obtained by the method for constructing an online game craving degree prediction model according to claim 3 or 4, based on the electroencephalogram data of the subject in the game cue response task; A second output module, configured to: output the prediction result of the online game craving degree of the subject after a preset time period after the game cue response task to at least one terminal.

9. An evaluation system for the intervention effect of online game addiction, characterized in that, Comprising: A third receiving module, configured to: receive, from at least one terminal, electroencephalogram data of a subject in a game cue response task before and after transcranial direct current stimulation intervention, wherein the subject is an online game addict, and the electroencephalogram data of the subject in the game cue response task before and after transcranial direct current stimulation intervention includes the average amplitude difference of the P300 brain electrical component at the Pz electrode point of the brain of the subject in the game cue response task before and after transcranial direct current stimulation intervention; An evaluation module, configured to: obtain an evaluation result of the online game addiction intervention effect of the subject after a preset time period after the transcranial direct current stimulation intervention through an online game addiction intervention effect evaluation model obtained by the method for constructing an online game addiction intervention effect evaluation model according to claim 5 or 6, based on the electroencephalogram data of the subject in the game cue response task before and after the transcranial direct current stimulation intervention; A third output module, configured to: output the evaluation result of the online game addiction intervention effect of the subject after a preset time period after the transcranial direct current stimulation intervention to at least one terminal.

10. An integrated evaluation system for online game addiction, characterized in that, It is obtained by integrating at least two of the online game disorder recognition system described in claim 7, the online game craving degree prediction system described in claim 8, and the online game addiction intervention effect evaluation system described in claim 9.

Images