Reading disorder identification method and system based on electroencephalogram monitoring
By combining EEG technology and writing tasks, the EEG data of subjects during writing tasks was collected, and the problem of insufficient recognition accuracy was solved in the prior art that ignores writing indicators, and more efficient and accurate recognition of dyslexia was achieved.
Patent Information
- Application Number
- CN202411851671.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-16
- Publication Date
- 2025-03-28
AI Technical Summary
The existing EEG interface technology ignores writing indicators when identifying dyslexia, resulting in insufficient recognition accuracy.
By presenting the language materials to the subjects, requiring them to perform writing tasks, and collecting task-state EEG data for classification recognition, combining EEG technology and writing tasks to improve recognition efficiency and accuracy.
This method provides a direct and objective means to evaluate dyslexia by directly collecting EEG data of subjects when performing writing tasks, improving the accuracy and personalization of recognition.
Smart Images
Figure CN119700143A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the technical field of electroencephalogram (EEG) interfaces, and specifically relates to a method and system for identifying dyslexia based on EEG monitoring. Background Art
[0002] At present, traditional clinical methods for the assessment and identification of dyslexia are mainly based on scales and behavioral observations, which require a comprehensive assessment of the individual's cognitive ability, language comprehension, visual processing, and educational background. In addition, the symptoms are diverse and easily confused with other learning disabilities. In the field of physiological monitoring, eye movement monitoring is a common method for identifying dyslexia, but it is greatly affected by the test environment and individual status, and it mostly uses reading tasks for assessment, ignoring the fact that writing is also an important assessment indicator for dyslexia.
[0003] A brain-computer interface is a communication and control pathway between the human brain and the external environment that is established through computers or other electronic devices and is independent of peripheral nerves and muscle tissue. In disease screening, brain-computer interfaces can provide early signs of lesions by monitoring EEG signals and other data, providing key information for the early diagnosis and treatment of neurological diseases. In contrast, electroencephalogram (EEG) monitoring can more directly reflect brain activity, provide more accurate information on cognitive processes, and help bypass the subjectivity and environmental interference of eye movement monitoring, thereby more accurately identifying dyslexia. However, existing EEG interface technologies mostly evaluate and identify dyslexia in a resting state, and also ignore writing indicators, which affects the accuracy of EEG identification of dyslexia. Summary of the invention
[0004] The technical purpose of this application is to address the technical problem that the current dyslexia identification method using brain-computer interface technology ignores writing indicators and the recognition accuracy needs to be improved. It provides a dyslexia identification method and system based on EEG monitoring, combining EEG technology with writing tasks to improve the efficiency and accuracy of dyslexia identification.
[0005] In order to achieve the above technical objectives, this application adopts the following technical solutions.
[0006] In a first aspect, an embodiment of the present application provides a method for identifying dyslexia based on EEG monitoring, comprising: presenting a language material to a subject, and causing the subject to perform a writing task according to the language material, wherein the writing task is to write a target content of a preset category, and there is at least one writing task. If multiple writing tasks are performed, each writing task writes a target content of a different category; Collecting task-state EEG data of the subject when the subject performs the writing task; Classification and recognition are performed according to the task-state EEG data to obtain a reading disability recognition result.
[0007] In some possible implementations, presenting language materials to subjects and causing the subjects to perform a writing task according to the language materials includes: Presetting the task time corresponding to each of the writing tasks according to the difficulty level of each of the writing tasks; A plurality of different language materials are displayed in sequence, each of which has a set display duration. After each display of the language material, the subject is asked to perform a corresponding writing task according to the language material within the corresponding task time. The number of the language materials is the same as the number of the writing tasks.
[0008] In some possible implementations, the preset category of target content includes first category vocabulary, second category vocabulary, first category non-vocabulary and / or second category non-vocabulary, the second category vocabulary is more difficult than the first category vocabulary, and the second category non-vocabulary is more difficult than the first category non-vocabulary.
[0009] In some possible implementations, the method further includes: pre-setting the total task time corresponding to all the writing tasks according to the difficulty level of each of the writing tasks; and presenting language materials to the subjects so that the subjects perform each of the writing tasks according to the language materials within the total task time.
[0010] In some possible implementations, the method further includes: when the subject performs a writing task according to the language material, obtaining the corresponding target content according to the language material; Alternatively, while presenting the language material to the subject, reference materials are provided to the subject, so that when the subject performs the writing task according to the language material, the subject can obtain the corresponding target content from the reference material according to the language material.
[0011] In some possible implementations, classification and identification are performed based on the task-state EEG data to obtain a reading disability identification result, including: Using a task discriminator to classify the task-state EEG data when the subject performs the writing task, to obtain a prediction probability corresponding to the writing task; and determining an absolute value of a difference between the prediction probability and a label value corresponding to the writing task; An abnormality detection is performed on the absolute value of each difference. When an abnormal value exists in the absolute values of all the differences, it is determined that the subject has a risk of dyslexia; otherwise, when no abnormal value exists in all the absolute difference values, the subject does not have a risk of dyslexia.
[0012] In some possible implementations, the method further includes collecting resting-state EEG data of the subject; Performing a prediction based on the resting-state EEG data and the task-state EEG data to predict whether the subject has entered the task state; If it is predicted that the subject has entered the task state, classification and identification are performed based on the task state EEG data.
[0013] In a second aspect, the present application provides a reading disability identification system based on EEG monitoring, including: The material provision module is used to present language materials to the subjects; A task execution module, used to enable the subject to perform a writing task according to the language material, wherein the writing task is to write a preset target content, and there is at least one writing task. If multiple writing tasks are performed, each writing task writes a different target content; A data collection module, used for collecting task-state EEG data of the subject when performing the writing task; The result recognition module is used to perform classification and recognition based on the task-state EEG data to obtain a reading disability recognition result.
[0014] In some possible implementations, the data acquisition module is also used to collect resting-state EEG data of the subject; The system also includes a domain identification module, The domain identification module is used to predict whether the subject enters the task state based on the resting state EEG data and the task state EEG data; If it is predicted that the subject has entered the task state, classification and identification are performed based on the task state EEG data.
[0015] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory and a communication bus; the communication bus is used to realize connection and communication between the processor and the memory; the processor is used to execute one or more programs stored in the memory to implement the steps of a method for identifying dyslexia based on EEG monitoring as provided in any possible implementation manner of the first aspect.
[0016] Compared with the prior art, the dyslexia identification method and system based on EEG monitoring provided in the embodiment of the present application directly collects the EEG data of the subjects when performing writing tasks, combines EEG technology with writing tasks, and provides a direct and objective means to evaluate dyslexia. Compared with traditional methods that rely on subjective reports or behavioral observations, EEG data can better reflect the true activity state of the brain when processing language information. The method allows the execution of multiple writing tasks, each task targeting different types of target content. This multi-task design can more comprehensively evaluate the performance of the subjects when processing different types of language materials, thereby more accurately identifying dyslexia. Since each person's brain activity and the manifestation of dyslexia may be different, the method can achieve personalized dyslexia identification by collecting and analyzing individual EEG data. This helps to provide more targeted intervention and treatment plans for different individuals. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The drawings described herein are for explanation purposes only and are not intended to limit the scope of the present application in any way. In addition, the shapes and proportional dimensions of the components in the drawings are only for illustration purposes and are used to help understand the present application. They do not specifically limit the shapes and proportional dimensions of the components of the present application. Under the guidance of the present application, those skilled in the art can select various possible shapes and proportional dimensions to implement the present application according to specific circumstances. In the drawings: Figure 1 A flowchart of a method for identifying dyslexia based on EEG monitoring provided in an embodiment of the present application; Figure 2 This is a structural block diagram of a dyslexia identification system based on EEG monitoring provided in an embodiment of the present application. DETAILED DESCRIPTION
[0018] In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.
[0019] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more features.
[0020] The existing EEG technology for the assessment and identification of dyslexia is mostly in a resting state. After research, the applicant believes that if it can be combined with the design of writing tasks in traditional assessment methods, it will help to further improve the accuracy of EEG identification of dyslexia. During the writing process, the brain activates two main language areas, namely Broca's area and Wernicke's area. Broca's area is mainly responsible for expressive writing and speaking, while Wernicke's area is responsible for understanding the language of writing and speaking.
[0021] In order to combine EEG technology with writing tasks and improve the efficiency and accuracy of dyslexia identification, the embodiments of the present application provide a dyslexia identification method and system based on EEG monitoring.
[0022] The present application is further described below in conjunction with the accompanying drawings and specific embodiments.
[0023] See also Figure 1 As shown, the method for identifying dyslexia based on EEG monitoring provided in the embodiment of the present application includes: Step S1: presenting language materials to the subjects, and asking them to perform a writing task based on the language materials. The writing task is to write target content of a preset category. There is at least one writing task. If multiple writing tasks are performed, each writing task writes target content of a different category.
[0024] In a specific embodiment, in step S1, the language material can be displayed to the subject using an electronic display device or a traditional physical medium. The display device can be a liquid crystal display device (LCD), an organic light emitting diode display device (OLED), a plasma display panel device (PDP), etc. The traditional physical medium is a physical, touchable medium, such as paper, etc.
[0025] The preset target content may include vocabulary target content and non-vocabulary target content. In some embodiments, in order to further refine the category of the target content, the preset target content may include first-category vocabulary, second-category vocabulary, first-category non-vocabulary and / or second-category non-vocabulary. In a specific embodiment, the writing tasks corresponding to all or part of these target contents may be randomly executed in a loop, or the writing tasks may be executed in a loop in sequence until the termination condition is met (such as the number of writing tasks executed reaches a preset number of times).
[0026] Optionally, the difficulty of the second category of vocabulary is greater than that of the first category of vocabulary, and the difficulty of the second category of non-vocabulary is greater than that of the first category of non-vocabulary.
[0027] In the context of dyslexia recognition, "vocabulary" generally refers to the words or phrases that actually exist in a language, have specific meanings and pronunciations. These vocabulary are the basic units that make up a language and are also the foundation for reading and understanding texts. In contrast to "vocabulary", "non-vocabulary" refers to those character combinations that do not form actual words or phrases in a language. These character combinations may consist of Chinese characters, letters, numbers or other symbols, but they do not have specific meanings and pronunciations in the language.
[0028] In specific embodiments, the vocabulary and non-vocabulary can be classified into grades in advance. The basic vocabulary can be classified as the first type of vocabulary. The basic vocabulary has a relatively high usage frequency and relatively low difficulty in daily communication, learning and work. They form the basic framework of the language and are the vocabulary that beginners need to master first. Those vocabulary that may involve more complex grammatical structures, more professional domain knowledge or more abstract concepts can be classified as the second type of vocabulary, which are usually more common in advanced reading, writing or professional communication.
[0029] It can also be classified according to the usage situation and semantic depth of the vocabulary in a specific context. For example, some vocabulary that frequently appears in daily life or basic learning may be classified as the first type of vocabulary, while some vocabulary used in specific fields or advanced contexts may be classified as the second type of vocabulary.
[0030] In specific embodiments, those character sequences that are composed of common letters or character combinations but do not form actual words or phrases in the language can be classified as the first type of non-vocabulary. The first type of non-vocabulary may have a certain pronunciation rule or pattern, but lacks specific semantic content.
[0031] Compared with the first type of non-vocabulary, the non-vocabulary composed of more complex character combinations or containing more uncommon letters or symbols is classified as the second type of non-vocabulary. The second type of non-vocabulary may be more difficult to pronounce or understand and is less likely to appear in actual language.
[0032] In an embodiment, writing normal words will activate the regions in the brain related to memory, recall and matching the written words, and these regions are located in the Visual Word Form Area (VWFA) of the occipitotemporal region. This will arouse the Broca's area and Wernicke's area in the user's brain and try to decode non-words through these regions instead of recalling by memory. The language materials used in the above tasks can refer to traditional clinical speech disorder recognition scales, and the difficulty levels are distinguished by combining the word length and the complexity of Chinese character strokes. The target content of each writing task contains Chinese characters (such as: "午" and "牛") or Chinese character combinations (such as: "中午" and "水牛") that are difficult for dyslexic patients to distinguish.
[0033] In an embodiment, language materials are displayed to subjects, and the subjects are asked to perform a writing task according to the language materials (which may be on a writing board), including: pre-setting a task time corresponding to each writing task according to the difficulty level of each writing task; displaying a plurality of different language materials in sequence, with each language material displaying a set duration, and after each language material is displayed, allowing the subjects to perform a corresponding writing task according to the language material within the corresponding task time, and the number of language materials is the same as the number of writing tasks.
[0034] Taking the first task, the second task, the third task and the fourth task as an example, the task times corresponding to the first type of vocabulary, the second type of vocabulary, the first type of non-vocabulary and the second type of non-vocabulary, respectively, can be equal or unequal, such as 20 seconds. Set the display time of each language material to 5 seconds. After displaying the language material corresponding to the first task for 5 seconds, the subject is required to perform the first task within 20 seconds (the 25 seconds are the first task cycle); after displaying the language material corresponding to the second task for 5 seconds, the subject is required to perform the second task within 20 seconds (the 25 seconds are the second task cycle); after displaying the language material corresponding to the third task for 5 seconds, the subject is required to perform the third task within 20 seconds (the 25 seconds are the third task cycle); after displaying the language material corresponding to the fourth task for 5 seconds, the subject is required to perform the fourth task within 20 seconds (the 25 seconds are the fourth task cycle). A 5-second time can be set between each task cycle for the subject to rest.
[0035] By setting a fixed task time for each writing task, the embodiment can ensure that all subjects complete the task under the same time conditions, thereby improving the consistency and comparability of the experiment and helping to reduce experimental errors caused by time differences. By setting the task time according to the difficulty level of the task, it can ensure that the subjects have enough time to deal with tasks of different difficulties. For more difficult tasks, providing longer time can reduce frustration or inaccurate results caused by insufficient time; for simpler tasks, they can be completed in a shorter time to improve experimental efficiency.
[0036] In some embodiments, the language materials are presented to the subjects, and the subjects are asked to perform writing tasks according to the language materials, including: presetting the total task time (e.g., 20 minutes) corresponding to all writing tasks according to the difficulty level of each writing task; and presenting the language materials to the subjects (which can be presented for a set time or can be presented continuously), so that the subjects perform each writing task according to the language materials within the total task time. By setting a total task time, the subjects are allowed to complete the tasks within a more flexible time frame, which helps the subjects to allocate time according to their own rhythm and moderation, and improves the efficiency of the experiment.
[0037] In an embodiment, when the subject performs a writing task according to the language material, the subject obtains the corresponding target content according to the language material, such as the content of the language material is the target content, or the language material includes the corresponding target content, and the target content can be found in the language material according to the prompt of the language material; or, while showing the subject the language material, the subject is provided with reference materials, so that when the subject performs a writing task according to the language material, the subject obtains the corresponding target content in the reference material according to the language material, such as the target content can be found in the reference material according to the prompt of the language material.
[0038] The flexibility of the target content acquisition method of the embodiment enables the test to adapt to the needs of dyslexia identification of different difficulty levels and types. When the language material contains clues pointing to the target content in the reference material, the subject needs to integrate information from different sources, which further exercises his information integration and memory retrieval ability.
[0039] Step S2: Collecting the task-state EEG data of the subjects when they perform the writing task.
[0040] The EEG data can be collected using an EEG signal collection device.
[0041] Step S3: Classify and identify the task-state EEG data to obtain dyslexia identification results.
[0042] In some embodiments, step S3 includes: classifying the task-state EEG data of the subject when performing the writing task to obtain the predicted probability of the corresponding writing task; determining the absolute value of the difference between the predicted probability and the label value of the corresponding writing task; performing anomaly detection on the absolute value of each difference, and when there are abnormal values among the absolute values of all differences, it is determined that the subject is at risk of dyslexia; otherwise, when there are no abnormal values among all absolute differences, the subject is not at risk of dyslexia.
[0043] Still take the first task, the second task, the third task and the fourth task as an example, which are predetermined to correspond to the four target contents of the first category of vocabulary, the second category of vocabulary, the first category of non-vocabulary and the second category of non-vocabulary. As an example, the first category corresponds to "simple tasks" and is coded as 0; the second category corresponds to "complex tasks" and is coded as 1. The code corresponding to vocabulary tasks is 0, and the code corresponding to non-vocabulary tasks is 1. Therefore, the writing task label corresponding to the first category of vocabulary is 00, the writing task label corresponding to the second category of vocabulary is 10, the writing task label corresponding to the first category of non-vocabulary is 01, and the writing task label corresponding to the second category of non-vocabulary is 11. The labels can be converted into two-dimensional vectors.
[0044] The task discriminator is used to classify the task-state EEG data. The task discriminator obtains the predicted probability of the corresponding writing task, including [P1, P2], where P1 represents the probability that the sample belongs to a "simple task" or a "complex task"; P2 represents the probability that the sample belongs to a "lexical task" or a "non-lexical task".
[0045] For example, if a writing task is the second task, and its target content is the second category of vocabulary, the corresponding real writing task label is [1, 0], and the probability vector output by the task discriminator is [0.8, 0.3], then subtracting the probability vector from the label vector and taking the absolute value should be: |0.8 - 1| = 0.2, |0.3 - 0| = 0.3; a new two-dimensional vector [0.2, 0.3] is obtained, which indicates the degree of difference between the predicted result and the real label. The smaller the difference, the closer it is to 0, the more accurate the prediction, and the larger the difference, the closer it is to 1, the greater the prediction deviation.
[0046] In some embodiments, the dyslexia identification method based on EEG monitoring also includes collecting resting-state EEG data of the subject, allowing the subject to close his eyes and relax to obtain an EEG baseline record and eliminate physiological conditions that may affect the signal; a domain discriminator can be used to make predictions based on resting-state EEG data and task-state EEG data to predict whether the subject has entered the task state; if it is predicted that the subject has entered the task state, classification and identification are then performed based on the task-state EEG data.
[0047] Take a subject as an example. The total number of writing tasks performed is N, that is, N task-state EEG data of N writing tasks are collected. The subject has n task-state EEG data that are judged to be in the task state. The n task-state EEG data are classified and identified, and n probability vectors are output. After the above calculations are performed on these n probability vectors and the corresponding n writing task labels, the obtained feature vectors are combined into a tensor of the shape (2, n), and the mask vector is used to fill it into a tensor of the shape (2, N) after the sequence. Finally, the probability feature sequence Fi of the subject i is obtained. Fi is used for anomaly detection with the X normal data that have been collected. A statistical method is used to model the feature distribution and check whether Fi is an outlier. For those determined as outliers by anomaly detection, a high-risk warning is given.
[0048] In an embodiment, the dyslexia identification results can be visualized and displayed on a terminal carried by the system. In particular, in some embodiments, the analysis results of high-risk warnings can be submitted to clinical researchers for manual review to further review the risk of dyslexia. If the clinical review is a normal case, the sample will be incorporated into the normal sample library to improve the detection accuracy of the abnormal detection model. If the review is a dyslexia patient, the corresponding medical treatment and health advice will be given. An evaluation report is automatically generated for those that do not need to be reviewed, and an evaluation report is generated after the review is completed for those that need manual review, combined with clinical opinions.
[0049] Specifically, the assessment report includes: whether there is dyslexia; if there is dyslexia, the degree of dyslexia (mild / severe); and which parts of the target content in the task the user has difficulty reading.
[0050] In the embodiment, the task discriminator and the domain discriminator can be implemented using a machine learning model, such as a convolutional neural network (CNN), which can achieve better prediction performance with fewer parameters and relatively lower requirements for training data. Those skilled in the art can clearly understand that in practical applications, historical data needs to be collected in advance to train the task discriminator and the domain discriminator, and no further explanation is given for this.
[0051] The dyslexia identification method based on EEG monitoring provided in the embodiment of the present application is a system for identifying and screening dyslexia and its severity based on the EEG signal characteristics of a specific writing task. The aim is to overcome the current problems of complex dyslexia screening process and low accuracy through the combination of EEG technology and language writing tasks, and to improve the efficiency and accuracy of dyslexia identification.
[0052] Based on the same inventive concept as the above-mentioned method for identifying dyslexia based on EEG monitoring, the present application embodiment also provides a system for identifying dyslexia based on EEG monitoring, such as Figure 2 As shown, it includes a material providing module, a task executing module, a data collecting module and a result identifying module.
[0053] The material provision module is used to present language materials to subjects.
[0054] The task execution module is used to enable the subjects to perform a writing task based on the language material. The writing task is to write a preset target content. There is at least one writing task. If multiple writing tasks are performed, each writing task writes a different type of target content.
[0055] The data acquisition module is used to collect task-state EEG data of subjects when they perform writing tasks.
[0056] The result recognition module is used to classify and recognize task-state EEG data to obtain dyslexia recognition results.
[0057] In some embodiments, the data acquisition module is also used to collect resting-state EEG data of the subject; the dyslexia identification system based on EEG monitoring also includes a domain identification module (not shown in the figure), which is used to make predictions based on resting-state EEG data and task-state EEG data to predict whether the subject has entered the task state; if it is predicted that the subject has entered the task state, classification and identification are performed based on the task-state EEG data.
[0058] Optionally, the dyslexia recognition system based on EEG monitoring also includes an evaluation result feedback module, which is used to visualize the dyslexia recognition results and display them on the terminal carried by the system, or generate an evaluation report, which includes the existence of dyslexia; if dyslexia exists, the degree of dyslexia (mild / severe); for the target content in the writing task, which parts the user has difficulty reading. This application can realize the identification of healthy people, patients with mild dyslexia and patients with severe dyslexia.
[0059] It should be noted that the dyslexia identification system based on EEG monitoring provided in the embodiment of the present application can implement the dyslexia identification method based on EEG monitoring provided in the embodiment of the present application, but the implementation device of the dyslexia identification method based on EEG monitoring provided in the present application includes but is not limited to the dyslexia identification system based on EEG monitoring listed in the present embodiment. All structural deformations and replacements of the prior art made according to the principles of the present application are included in the protection scope of the present application.
[0060] An embodiment of the present application also provides an electronic device, including a processor, a memory and a communication bus; the communication bus is used to realize the connection and communication between the processor and the memory; the processor is used to execute one or more programs stored in the memory to implement the steps of the above method for identifying dyslexia based on EEG monitoring.
[0061] The above is a detailed introduction to the dyslexia identification method and system based on EEG monitoring provided by the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the concept of the present application and should not be construed as limiting the scope of protection of the present application.
Claims
1. A method for identifying dyslexia based on EEG monitoring, characterized in that: include: Presenting language materials to the subjects, and allowing the subjects to perform a writing task according to the language materials, wherein the writing task is to write a target content of a preset category, and there is at least one writing task. If multiple writing tasks are performed, each writing task writes a target content of a different category; collecting task-state EEG data of the subject when the subject performs the writing task; Classification and recognition are performed according to the task-state EEG data to obtain a reading disability recognition result.
2. The method for identifying dyslexia based on EEG monitoring according to claim 1, characterized in that: Presenting language materials to the subjects and causing the subjects to perform a writing task based on the language materials, including: Presetting the task time corresponding to each of the writing tasks according to the difficulty level of each of the writing tasks; A plurality of different language materials are displayed in sequence, each of which has a set display duration. After each display of the language material, the subject is required to perform a corresponding writing task according to the language material within the corresponding task time. The number of the language materials is the same as the number of the writing tasks.
3. The method for identifying dyslexia based on EEG monitoring according to claim 1, characterized in that: The target content of the preset category includes first category vocabulary, second category vocabulary, first category non-vocabulary and / or second category non-vocabulary, the difficulty of the second category vocabulary is greater than that of the first category vocabulary, and the difficulty of the second category non-vocabulary is greater than that of the first category non-vocabulary.
4. The method for identifying dyslexia based on EEG monitoring according to claim 1, characterized in that: The method further comprises: Presetting the total task time corresponding to all the writing tasks according to the difficulty level of each of the writing tasks; The language materials are presented to the subjects, so that the subjects perform the writing tasks according to the language materials within the total task time.
5. The method for identifying dyslexia based on EEG monitoring according to claim 1, characterized in that: The method further comprises: When the subject performs a writing task according to the language material, the corresponding target content is obtained according to the language material; Alternatively, while presenting the language material to the subject, reference materials are provided to the subject, so that when the subject performs the writing task according to the language material, the subject can obtain the corresponding target content from the reference material according to the language material.
6. The method for identifying dyslexia based on EEG monitoring according to claim 1, characterized in that: Classification and recognition are performed according to the task-state EEG data to obtain dyslexia recognition results, including: Using a task discriminator to classify the task-state EEG data when the subject performs the writing task, to obtain a prediction probability corresponding to the writing task; and determining an absolute value of a difference between the prediction probability and a label value corresponding to the writing task; An abnormality detection is performed on the absolute value of each difference. When an abnormal value exists in the absolute values of all the differences, it is determined that the subject has a risk of dyslexia; otherwise, when no abnormal value exists in all the absolute difference values, the subject does not have a risk of dyslexia.
7. The method for identifying dyslexia based on EEG monitoring according to claim 1, characterized in that: The method further includes collecting resting-state EEG data of the subject; Performing a prediction based on the resting-state EEG data and the task-state EEG data to predict whether the subject has entered the task state; If it is predicted that the subject has entered the task state, classification and identification are performed based on the task state EEG data.
8. A reading disability recognition system based on EEG monitoring, characterized in that: include: The material provision module is used to present language materials to the subjects; A task execution module, used to enable the subject to perform a writing task according to the language material, wherein the writing task is to write a preset target content, and there is at least one writing task. If multiple writing tasks are performed, each writing task writes a different target content; A data collection module, used for collecting task-state EEG data of the subject when performing the writing task; The result recognition module is used to perform classification and recognition based on the task-state EEG data to obtain a reading disability recognition result.
9. The reading disability identification system based on EEG monitoring according to claim 8, characterized in that: The data acquisition module is also used to collect the resting-state EEG data of the subject; The system also includes a domain identification module, The domain identification module is used to predict whether the subject enters the task state based on the resting state EEG data and the task state EEG data; If it is predicted that the task state has been entered, classification and identification are performed based on the task state EEG data.
10. An electronic device, characterized in that: It comprises a processor, a memory and a communication bus; the communication bus is used to realize the connection and communication between the processor and the memory; the processor is used to execute one or more programs stored in the memory to realize the steps of the dyslexia identification method based on EEG monitoring as described in any one of claims 1 to 7.