Student mental health monitoring system based on multi-modal evaluation and multi-role report

Abstract

The invention relates to the technical field of psychological health detection, in particular to a student psychological health monitoring system based on multi-modal evaluation and a multi-role report, which comprises a data acquisition module, a data processing module, a report generation module, a resource allocation module, a data processing module and a monitoring module, the configuration adjustment module is in communication connection with the report generation module and is used for dynamically adjusting the configuration of the psychological teacher based on the analysis result of the report generation module; the user interface module is in communication connection with the report generation module and the resource allocation module and is used for displaying corresponding analysis reports to users of different roles; and visual display of a resource allocation result is provided. Through the characteristics of comprehensiveness, intelligence, safety and individuation, many problems existing in a traditional method are effectively solved. The system not only can provide more accurate and comprehensive mental health assessment, but also can provide powerful decision support for educational practice.

Description

Technical Field

[0001] The present invention relates to the technical field of mental health detection, and particularly to a student mental health monitoring system based on multi-modal evaluation and multi-role reports. Background Art

[0002] With the increasing social attention to mental health issues, the mental health status of students has become a key factor affecting the quality of education and the all-round development of students. Traditional methods for monitoring students' mental health mainly rely on regular questionnaires and face-to-face interviews. Although these methods can reflect students' mental conditions to a certain extent, they have many limitations.

[0003] Firstly, traditional methods often can only provide static and one-sided evaluation results. A single questionnaire is difficult to capture the dynamic changes of students' mental states, nor can it comprehensively reflect students' mental performances in different situations. Secondly, these methods usually only focus on individual students and ignore the influence of external factors such as the personality traits of parents, the mental health status of teachers, the living environment, and risk life events on students' mental health. Moreover, traditional evaluation results often lack pertinence and are difficult to provide effective guiding suggestions for educational participants in different roles (such as teachers, parents, school administrators).

[0004] In addition, existing mental health monitoring systems also have deficiencies in data security and privacy protection. Students' mental health data belongs to highly sensitive personal information. How to protect students' privacy while making full use of the data is a major challenge faced by existing systems. At the same time, traditional methods also seem powerless in resource allocation and are difficult to dynamically adjust the allocation of mental health resources according to evaluation results, resulting in low resource utilization efficiency.

[0005] Finally, existing mental health monitoring methods generally lack the ability of self-optimization and continuous learning. As time goes by and society changes, the factors affecting students' mental health are also constantly changing, while static evaluation systems are difficult to adapt to this change, which easily leads to a gradual decrease in the accuracy and effectiveness of evaluation results.

[0006] In view of the above problems, there is an urgent need for a system that can comprehensively, dynamically, and securely monitor students' mental health status and provide personalized guidance for different roles. The present invention is an innovative solution proposed in response to this need. Summary of the Invention

[0007] The technical problem to be solved by the present invention is how to construct a student mental health monitoring system based on multi-modal evaluation and multi-role reports to achieve a comprehensive evaluation, dynamic monitoring, and personalized intervention of students' mental health status.

[0008] The present invention proposes a student mental health monitoring system based on multimodal assessment and multi-role reports, including:

[0009] A data acquisition module for:

[0010] Collecting standardized psychological test questionnaire data of students;

[0011] Collecting behavioral experiment data of students, including implicit association experiment data;

[0012] Collecting relevant data of students' parents, teachers, and living environments;

[0013] A data processing module, communicatively connected to the data acquisition module, for:

[0014] Receiving multimodal assessment data sent by the data acquisition module;

[0015] Based on the multimodal assessment data, using machine learning algorithms for comprehensive analysis and prediction;

[0016] A report generation module, communicatively connected to the data processing module, for:

[0017] Generating personalized analysis reports for different roles based on the analysis results of the data processing module;

[0018] Generating comparative analysis reports between schools, regions, and provinces and cities;

[0019] A resource allocation module, communicatively connected to the report generation module, for:

[0020] Dynamically adjusting the configuration of psychological teachers based on the analysis results of the report generation module;

[0021] A user interface module, communicatively connected to the report generation module and the resource allocation module, for:

[0022] Displaying corresponding analysis reports to users of different roles;

[0023] Providing a visual display of the resource allocation results.

[0024] Preferably, the data acquisition module includes:

[0025] A questionnaire data acquisition unit for collecting standardized psychological test questionnaire data;

[0026] A behavioral experiment data acquisition unit for collecting behavioral experiment data such as implicit association experiments;

[0027] An environment data acquisition unit for collecting relevant data of students' parents, teachers, and living environments;

[0028] Among them, the behavioral experiment data acquisition unit obtains the implicit attitude data of students by recording the reaction time and error rate of students in the implicit association experiment.

[0029] Preferably, the data processing module includes:

[0030] A data preprocessing unit for cleaning, standardizing, and feature extracting the multimodal evaluation data;

[0031] A machine learning analysis unit for analyzing and predicting the preprocessed data using machine learning algorithms;

[0032] Among them, the machine learning analysis unit uses a deep learning model to perform a fusion analysis on the questionnaire data, behavioral experiment data, and environmental data, and generates a comprehensive evaluation result of the mental health status of students.

[0033] Preferably, the report generation module includes:

[0034] A personalized report generation unit for generating targeted analysis reports according to different user roles;

[0035] A comparative analysis report generation unit for generating comparative analysis reports between schools, regions, and provinces and cities;

[0036] Among them, the personalized report generation unit selectively presents relevant mental health indicators and suggestions according to different user roles to improve the practicality and operability of the report.

[0037] Preferably, the resource allocation module includes:

[0038] A demand analysis unit for evaluating the demand for mental health resources based on the analysis results of the report generation module;

[0039] A resource allocation unit for dynamically adjusting the configuration of psychological teachers according to the demand analysis results;

[0040] Among them, the resource allocation unit uses an intelligent optimization algorithm to comprehensively consider the number of students, the severity of psychological problems, and the existing resource status, and generates an optimal resource allocation plan.

[0041] Preferably, it further includes:

[0042] An early warning module, communicatively connected to the data processing module and the report generation module, for:

[0043] Setting a mental health risk threshold based on the analysis results of the data processing module;

[0044] When the mental health indicators of students exceed the preset threshold, generating a warning message;

[0045] Send the warning information to the report generation module and include it in the analysis reports of relevant roles.

[0046] Preferably, the warning module further includes:

[0047] A risk level assessment unit for grading the warning information and classifying mental health risks into three levels: low, medium, and high;

[0048] An intervention suggestion generation unit for generating corresponding intervention suggestions according to the risk level and including them in the analysis reports.

[0049] Preferably, it further includes:

[0050] A data security module, communicatively connected to the data collection module and the data processing module, for:

[0051] Encrypting and storing the collected data;

[0052] Controlling access permissions to the data;

[0053] Performing desensitization processing on the data processing process to protect the privacy of students.

[0054] Preferably, the user interface module includes:

[0055] A role identification unit for identifying user roles, including students, parents, teachers, and school administrators;

[0056] A report display unit for displaying corresponding analysis reports according to user roles;

[0057] An interaction function unit for providing functions such as report interpretation, problem feedback, and suggestion submission.

[0058] Preferably, it further includes:

[0059] A system evaluation module, communicatively connected to the data processing module and the report generation module, for:

[0060] Collecting feedback information from users on the reports generated by the system;

[0061] Evaluating the accuracy and effectiveness of the system based on the feedback information;

[0062] Automatically adjusting the analysis algorithm of the data processing module and the report template of the report generation module according to the evaluation results.

[0063] Advantages of the present invention:

[0064] From a macro perspective, the system proposed in this paper builds a closed-loop student mental health management ecosystem by integrating multimodal data collection, intelligent data analysis, personalized report generation, and dynamic resource allocation. This system not only comprehensively and accurately assesses students' mental health but also provides macro-level data support for education decision-makers, contributing to the development of more scientific and effective education policies.

[0065] In terms of system architecture, the present invention adopts a modular design, with each functional module being both relatively independent and closely coordinated. For example, the collaborative work of the data acquisition and data processing modules enables the system to collect students' mental health data from multiple dimensions and conduct in-depth analysis using advanced machine learning algorithms. This synergistic effect significantly improves the accuracy and comprehensiveness of mental health assessments.

[0066] In terms of security and privacy protection, the data security module of this invention forms an organic unity with other functional modules. Through technologies such as encrypted storage, permission control, and data desensitization, the system maximizes the value of data while protecting student privacy. This balance resolves the contradiction between data utilization and privacy protection in traditional systems.

[0067] Another key feature of this system is its personalized report generation for different roles. The report generation module works in conjunction with the user interface module, enabling students, parents, teachers, and school administrators to receive tailored analysis reports and recommendations. This personalized information presentation significantly enhances the system's practicality and operability.

[0068] In terms of resource allocation, the resource allocation module of the present invention achieves intelligent and dynamic allocation of mental health resources through collaboration with the data processing module and the report generation module. This not only improves resource utilization efficiency but also better meets the actual needs of students, achieving twice the result with half the effort.

[0069] Particularly noteworthy is the introduction of an early warning module and a system evaluation module, which empower the system with proactive intervention and self-optimization capabilities. The early warning module monitors students' mental health indicators in real time, enabling early detection of potential psychological issues and enabling early intervention. The system evaluation module continuously collects user feedback and optimizes various system functions, ensuring that the system remains up-to-date and maintains its effectiveness and advancement.

[0070] From a microscopic perspective, the present invention has innovations and breakthroughs in many technical details. For example, in the data collection stage, the system not only collects traditional questionnaire data, but also introduces behavioral experiment data and environmental data. This multi-modal data collection method greatly enriches the evaluation dimensions. In the data analysis stage, the system adopts advanced deep learning algorithms, which can extract valuable features and patterns from complex multi-modal data.

[0071] In summary, the student mental health monitoring system based on multi-modal evaluation and multi-role reporting of the present invention effectively solves many problems existing in traditional methods through its comprehensive, intelligent, secure, and personalized characteristics. The system can not only provide a more accurate and comprehensive mental health assessment, but also provide strong decision-making support for educational practice, which is of great significance for improving students' mental health level, optimizing the allocation of educational resources, and promoting educational fairness. The popularization and application of this innovative solution is expected to bring a qualitative leap to the field of student mental health management and make important contributions to building a more healthy and harmonious educational ecosystem. Brief Description of the Drawings

[0072] Figure 1 is the main workflow diagram of the present invention;

[0073] Figure 2 is the logic block diagram of the data collection module of the present invention;

[0074] Figure 3 is the logic block diagram of the data processing module of the present invention;

[0075] Figure 4 is the logic block diagram of the report generation module of the present invention;

[0076] Figure 5 is the logic block diagram of the resource allocation module of the present invention;

[0077] Figure 6 is the logic block diagram of the user interface module of the present invention;

[0078] Figure 7 is the logic block diagram of the early warning module of the present invention;

[0079] Figure 8 is the logic block diagram of the data security module of the present invention;

[0080] Figure 9 is the logic block diagram of the system evaluation module of the present invention;

[0081] Figure 10 is the classification diagram of the current risks and potential risks of the present invention;

[0082] Figure 11 is the schematic diagram of the risk grading of the present invention. Detailed implementation manners

[0083] Please refer to the appendix Figures 1-9 , the present invention provides a student mental health monitoring system based on multimodal evaluation and multi-role reports, which can comprehensively and accurately evaluate the mental health status of students and provide personalized analysis reports for users of different roles. The technical solutions of the present invention will be described in detail below.

[0084] As described in claim 1, the system of the present invention includes a data acquisition module 1, a data processing module 2, a report generation module 3, a resource allocation module 4, and a user interface module 5. These modules work together to form a complete closed-loop system for student mental health monitoring.

[0085] Specifically, the data acquisition module 1 is used to collect various types of data, including standardized psychological test questionnaire data of students, behavioral experiment data (such as implicit association experiment data), and relevant data of students' parents, teachers, and living environments. This multimodal data acquisition method enables the system to comprehensively evaluate the mental health status of students from multiple dimensions. For example, when collecting standardized psychological test questionnaire data, the system can use widely used tools such as the SCL-90 Symptom Checklist or the Self-Rating Depression Scale (SDS). When collecting behavioral experiment data, the system can use the Implicit Association Test (IAT) to evaluate the potential psychological tendencies of students.

[0086] The data processing module 2 is communicatively connected to the data acquisition module 1 and is used to receive multimodal evaluation data and perform comprehensive analysis and prediction. The present invention preferably uses machine learning algorithms to process the data to improve the accuracy of analysis and the reliability of prediction. For example, algorithms such as Support Vector Machine (SVM) or Random Forest can be used to perform fusion analysis on multimodal data. Specifically, the following data fusion algorithm can be adopted:

[0087]

[0088] Among them, F is the final fusion result, n is the number of data modalities, w i is the weight of the i-th data modality, and f i (x i ) is the feature function of the i-th data modality. By adjusting the weights of different data modalities, the fusion effect can be optimized and the accuracy of mental health assessment can be improved.

[0089] The report generation module 3 is communicatively connected to the data processing module 2, and generates personalized analysis reports for different roles based on the analysis results, as well as comparative analysis reports between schools, regions, provinces and cities. This multi-role and multi-level report generation mechanism can meet the needs of different users and improve the practicality and operability of the reports. For example, for the report of an individual student, it may include specific mental health index scores and improvement suggestions; while for school administrators, the report may focus on overall statistical data and resource allocation suggestions.

[0090] The resource allocation module 4 is communicatively connected to the report generation module 3 and is used to dynamically adjust the allocation of psychological teachers according to the analysis results. This function helps to optimize the allocation of mental health resources and improve the resource utilization efficiency. For example, the system can automatically calculate the number of psychological teachers required based on the mental health risk level of students and give specific allocation suggestions.

[0091] The user interface module 5 is communicatively connected to the report generation module 3 and the resource allocation module 4, and is responsible for presenting the corresponding analysis reports to users of different roles and providing a visual display of the resource allocation results. The design of this module focuses on the user experience to ensure the clear presentation and convenient access of information.

[0092] The data collection module 1 further includes a questionnaire data collection unit 11, a behavioral experiment data collection unit 12, and an environmental data collection unit 13. This sub-divided structure makes the data collection process more systematic and professional.

[0093] Among them, the questionnaire data collection unit 11 is responsible for collecting data of standardized psychological test questionnaires. The present invention can adopt a variety of mature psychological assessment scales, such as the SCL-90 and SDS mentioned above, as well as the Adolescent Self-Rating Life Events Checklist (ASLEC), etc. These scales have been verified through long-term practice and have good reliability and validity, and can provide reliable basic data for mental health assessment.

[0094] The behavioral experiment data collection unit 12 is specifically used for collecting behavioral experiment data such as implicit association experiments. The implicit association experiment is an effective method for measuring an individual's latent attitude. By recording the reaction time and error rate of students in the experiment, implicit attitude data of students can be obtained. For example, when evaluating a student's attitude towards learning, word pairs such as "learning - pleasant" and "learning - painful" can be designed, and by measuring the reaction speed of students to these word pairs, their implicit attitude towards learning can be inferred.

[0095] The environmental data collection unit 13 is responsible for collecting relevant data of students' parents, teachers, and living environment. These data can be obtained through methods such as questionnaires, interviews, or environmental observations. For example, a Vulnerable Life Events Questionnaire can be designed to assess the impact of students' life risk events on students, a Vulnerable Traits Questionnaire to assess the impact of students' cognitive tendencies and emotional traits on students, a Vulnerable Environment to assess the impact of family and school environments on students, a Family Environment Scale to assess the family atmosphere of students, or a Teacher Rating Scale to collect teachers' observations of students. The collection of these environmental data provides important background information for comprehensively understanding the mental health status of students.

[0096] The data processing module 2 includes a data preprocessing unit 21 and a machine learning analysis unit 22. This structural design ensures the scientificity and reliability of data analysis.

[0097] The data preprocessing unit 21 is responsible for cleaning, standardizing, and feature extraction of multi-modal assessment data. This step is crucial for improving the accuracy of subsequent analysis. For example, in the data cleaning stage, the system will delete outliers and missing values to ensure the quality of the data. In the standardization process, the Z-score standardization method can be adopted to convert data with different dimensions to the same scale:

[0098]

[0099] where z is the standardized score, x is the original score, μ is the mean, and σ is the standard deviation.

[0100] The machine learning analysis unit 22 then uses a deep learning model to perform fusion analysis on the preprocessed questionnaire data, behavioral experiment data, and environmental data to generate a comprehensive assessment result of the students' mental health status. The present invention preferably adopts a multi-modal deep learning model, such as a Multimodal Attention Network, to make full use of the complementary information of different modal data. The core idea of this model is to perform weighted fusion of features of different modalities through an attention mechanism:

[0101]

[0102] where H is the fused feature representation, n is the number of data modalities, α i is the attention weight of the i-th data modality, h i is the feature representation of the i-th data modality. The attention weight α i can be calculated through the softmax function:

[0103]

[0104] where ei is the importance score of the i-th data modality and can be obtained through neural network learning.

[0105] In this way, the system of the present invention can adaptively adjust the importance of different data modalities, thereby obtaining more accurate and comprehensive mental health assessment results. This multi-modal fusion analysis method not only improves the accuracy of the assessment but also can capture potential mental health problems that are difficult to discover with single-modal data.

[0106] Generally speaking, the student mental health monitoring system based on multi-modal assessment and multi-role reports of the present invention provides a comprehensive, accurate, and efficient solution for student mental health monitoring through multi-dimensional data collection, advanced data processing algorithms, and personalized report generation mechanisms. The system can not only timely detect students' mental health problems but also provide valuable decision-making support for educators and administrators, which is of great significance for improving the overall education quality and promoting the all-round development of students.

[0107] The report generation module 3 of the present invention includes a personalized report generation unit 31 and a comparative analysis report generation unit 32. This module design aims to meet the diverse needs of different users and provide highly targeted and practical analysis reports.

[0108] The core function of the personalized report generation unit 31 is to generate targeted analysis reports according to different user roles. The system of the present invention can identify user roles such as students, parents, teachers, and school administrators, etc., and customize corresponding report content for them. For example, for student users, the report may focus on the assessment results and improvement suggestions of personal mental health status; while for teacher users, the report may include statistical data on the overall mental health status of the class and targeted educational guidance suggestions.

[0109] Preferably, the system of the present invention uses natural language generation technology to improve the readability and personalization of the report. Specifically, a template-based method combined with dynamic content generation technology can be used. The system first selects an appropriate report template according to the user role, and then dynamically fills in the key information points in the template based on the data analysis results. For example, when describing the mental health status of a student, the system may use the following template:

[0110] "Student [Name]'s [Mental Health Dimension] index is [Score], at the [Level] level. It is recommended that [Suggestion Content]."

[0111] Among them, the content in the square brackets will be dynamically generated according to the actual situation. This method not only ensures the professionalism and consistency of the report but also provides personalized content for each user.

[0112] The comparative analysis report generation unit 32 is responsible for generating comparative analysis reports between schools, regions, and provinces or municipalities. Such reports are mainly targeted at education administrators and decision-makers, aiming to provide an analysis of the mental health status at the macro level. The system of the present invention adopts a multi-dimensional comparative analysis method, which not only compares the overall mental health level but also conducts in-depth analysis on specific mental health dimensions.

[0113] In an embodiment of the present invention, the system uses a radar chart to visualize the performance of different regions or schools in each mental health dimension. Each axis of the radar chart represents a mental health dimension, such as emotional stability, learning pressure, interpersonal relationships, etc. By comparing the scores of different entities in each dimension, administrators can intuitively understand the advantages and disadvantages of each region or school, so as to formulate targeted improvement strategies.

[0114] In addition, the system of the present invention also introduces a time series analysis method to track the dynamic changes in the mental health status. For example, the system can generate monthly or quarterly trend reports to show the change trends of the mental health levels of each region or school. This kind of dynamic analysis helps to timely discover potential problems and evaluate the effectiveness of intervention measures.

[0115] The resource allocation module 4 of the present invention includes a demand analysis unit 41 and a resource allocation unit 42. The design of this module aims to optimize the allocation of mental health resources and improve the resource utilization efficiency.

[0116] The demand analysis unit 41 evaluates the demand for mental health resources based on the analysis results of the report generation module 3. The system of the present invention adopts a multi-factor comprehensive evaluation model to quantify the resource demand. This model takes into account multiple key factors, such as the total number of students, the incidence rate of psychological problems, the severity of the problems, etc. For example, the following formula can be used to calculate the resource demand index:

[0117]

[0118] where RDI is the resource demand index, N is the total number of students, P is the incidence rate of psychological problems, S is the average severity of the problems, α, β, γ are the weight coefficients of each factor, and the subscript max represents the maximum value of each index. By adjusting the weight coefficients, the system can flexibly adapt to the resource demand evaluation in different situations.

[0119] The resource allocation unit 42 is responsible for dynamically adjusting the allocation of psychology teachers according to the demand analysis results. The system of the present invention adopts an intelligent optimization algorithm to generate an optimal resource allocation plan. Preferably, the system uses a genetic algorithm to solve this multi-objective optimization problem. The objective function of the algorithm can be defined as:

[0120]

[0121] Among them, n is the number of schools or regions, and R i is the amount of resources allocated to the i-th school or region, and D i is its resource demand, C is the total cost, and C max is the maximum acceptable cost, and w1 and w2 are weight coefficients. This objective function takes into account both the balance of resource allocation and cost control.

[0122] In this way, the system of the present invention can dynamically adjust the allocation of mental health resources according to actual needs to ensure the most effective use of resources.

[0123] The system of the present invention further includes an early warning module 6, which is communicatively connected to the data processing module 2 and the report generation module 3. The introduction of the early warning module 6 greatly enhances the active intervention ability of the system and helps to detect and handle potential mental health problems in a timely manner.

[0124] Please refer to Figures 10-11 The early warning module 6 first sets mental health risk thresholds based on the analysis results of the data processing module 2. These thresholds are set based on a large amount of historical data and expert knowledge and are differentially set according to different mental health dimensions. For example, for the tendency of depression, the system may set multiple early warning levels:

[0125] Mild early warning: The score of the depression scale exceeds 50 points;

[0126] Moderate early warning: The score of the depression scale exceeds 65 points;

[0127] Severe early warning: The score of the depression scale exceeds 80 points;

[0128] Here, our early warning method comprehensively considers six aspects: suicidal ideation, psychological distress, event impact, vulnerable situation, vulnerable traits, and protective resources;

[0129] Among them, suicidal ideation, psychological distress, and event impact comprehensively represent the current risk;

[0130] Vulnerable situation, vulnerable traits, and positive psychological qualities comprehensively represent the potential risk;

[0131] Serious current risk is a third-level early warning;

[0132] Medium current risk and serious potential risk are second-level early warnings;

[0133] Moderate current risk or mild current risk and serious potential risk are first-level early warnings;

[0134] When the mental health indicators of students exceed the preset thresholds, the early warning module 6 will generate early warning information. These early warning information will be graded according to the severity and include detailed problem descriptions and preliminary intervention suggestions.

[0135] Subsequently, the warning information will be sent to the report generation module 3 and incorporated into the analysis reports of relevant roles. For example, for severe warnings, the system will immediately send emergency notifications to school psychologists and relevant management personnel, and at the same time add a special warning note and help-seeking guide to the student's personal report.

[0136] The warning module 6 also includes a risk level assessment unit 61 and an intervention suggestion generation unit 62, further enhancing the system's warning and intervention capabilities.

[0137] The risk level assessment unit 61 is responsible for grading the warning information and classifying mental health risks into three levels: low, medium, and high. This grading is not only based on the scores of a single indicator but also takes into account the comprehensive influence of multiple relevant factors. For example, when assessing the risk of suicide, the system will comprehensively consider factors such as the degree of depression, social support level, and recent major life events. The calculation of the risk level can use the weighted scoring method:

[0138] Risk Score=∑ i =1 n w i ·s i

[0139] where n is the number of factors considered, w i is the weight of the i-th factor, and s i is the standardized score of the i-th factor. According to the final Risk Score, the system classifies the risk level as:

[0140] Low risk: Risk Score < 0.3;

[0141] Medium risk: 0.3 ≤ Risk Score < 0.7;

[0142] High risk: Risk Score ≥ 0.7;

[0143] The intervention suggestion generation unit 62 then generates corresponding intervention suggestions according to the risk level and incorporates them into the analysis report. The system of the present invention uses a rule-based expert system to generate intervention suggestions. This expert system contains a large number of if-then rules, which are formulated by psychology experts based on the latest research results and clinical experience. For example:

[0144] IF the risk level = low risk AND the main problem = learning pressure

[0145] THEN the suggestion = "appropriately adjust the learning plan and increase relaxation and rest time"

[0146] IF the risk level = high risk AND the main problem = depressive symptoms

[0147] THEN Suggestion = "Arrange face-to-face psychological counseling immediately and consider referring to a professional psychiatrist."

[0148] In this way, the system of the present invention can provide targeted intervention suggestions for mental health problems in different situations, effectively supporting schools and families to take timely and appropriate intervention measures.

[0149] In summary, the student mental health monitoring system based on multi-modal assessment and multi-role reporting of the present invention provides an intelligent, efficient, and personalized solution for student mental health management through its comprehensive functional design and advanced technology implementation. This system can not only accurately evaluate the mental health status of students, but also timely warn of potential risks, and provide targeted analysis reports and intervention suggestions for users of different roles, which is of great significance for improving the mental health level of students and optimizing the allocation of educational resources.

[0150] The system of the present invention further includes a data security module 7, which is communicatively connected to the data collection module 1 and the data processing module 2. In the context of increasing emphasis on privacy protection today, the introduction of the data security module 7 is crucial for ensuring the compliance and credibility of the system.

[0151] The data security module 7 is mainly responsible for three aspects of work: encrypting and storing the collected data, controlling access permissions to the data, and desensitizing the data processing process. These three aspects constitute a multi-level data security protection system, effectively protecting the privacy information of students.

[0152] In terms of encrypting and storing data, the system of the present invention uses the Advanced Encryption Standard (AES) algorithm for data encryption. AES is a symmetric encryption algorithm, which has the characteristics of high security and high efficiency. Preferably, the system uses the AES-256 mode with a 256-bit key, and this mode is widely considered to be able to resist attacks by quantum computers. The encryption process can be expressed as:

[0153] C = E K (P),

[0154] where C is the ciphertext, E is the encryption function, K is the key, and P is the plaintext. In this way, even if the data storage medium is illegally accessed, the attacker cannot directly read the original data.

[0155] In terms of controlling data access permissions, the system of the present invention implements a Role-Based Access Control (RBAC) policy. The RBAC model associates users, roles, and permissions, achieving flexible and precise permission management. For example, the system may define the following roles and permissions:

[0156] Students: Can only access their own personal data and reports;

[0157] Teachers: Can access the aggregated data of the students in the classes they teach, but cannot view specific personal information;

[0158] Counselors: Can access the detailed information of the referred students;

[0159] System administrators: Can perform system maintenance, but have no direct access to student data;

[0160] Through this fine-grained permission control, the system ensures that each user can only access the minimum data set necessary for their role.

[0161] In terms of data processing desensitization, the system of the present invention employs a variety of technical means. For direct identifiers (such as names, student IDs, etc.), the system uses a hash function for irreversible conversion. For quasi-identifiers (such as age, grade, etc.), the system adopts the K-anonymization technology for processing. K-anonymization ensures that under any combination of quasi-identifiers, at least K records have the same value, thus preventing individual identification through these attributes.

[0162] For example, for the quasi-identifier of age, the system may divide it into coarser-grained intervals:

[0163] Original data: 14 years old, 15 years old, 16 years old;

[0164] After desensitization: 14 - 16 years old;

[0165] In this way, while protecting personal privacy, the system still retains the analytical value of the data.

[0166] The user interface module 5 of the present invention includes a role recognition unit 51, a report display unit 52, and an interaction function unit 53. This module design aims to provide an intuitive, friendly, and feature-rich user interface, enhancing the usability and user experience of the system.

[0167] The role recognition unit 51 is responsible for recognizing user roles, including students, parents, teachers, and school administrators, etc. The system of the present invention adopts a multi-factor authentication mechanism to ensure the accuracy and security of role recognition. For example, the system may combine methods such as username and password, SMS verification code, and biometric recognition (such as fingerprint or facial recognition) for identity verification. Preferably, the system also implements context-aware abnormal login detection, which will trigger additional verification steps if unusual login locations or devices are detected.

[0168] The report display unit 52 displays corresponding analysis reports according to the user role. The system of the present invention adopts a responsive design to ensure that the reports can be well presented on different devices (such as desktop computers, tablets, and mobile phones). The content and form of the reports will be dynamically adjusted according to the user role. For example:

[0169] For student users, the report may mainly consist of concise and easy-to-understand charts and text descriptions, focusing on presenting personal mental health status and improvement suggestions.

[0170] For teacher users, the report may contain more statistical data and trend analysis to help them understand the overall situation of the class.

[0171] For school administrators, the report may focus on presenting the mental health overview of the whole school, including comparative analysis of each grade and each class, etc.

[0172] The interaction function unit 53 provides functions such as report interpretation, problem feedback, and suggestion submission. The system of the present invention integrates natural language processing technology in this unit, enabling users to interact with the system in a more natural way. For example, the system may provide a chatbot interface, and users can directly ask: "Why is my depression index so high?" The system will analyze this question and give a targeted explanation based on the user's personal data.

[0173] In addition, the interaction function unit 53 also supports users to submit feedback and suggestions. The system will automatically classify and perform sentiment analysis on these feedbacks to help administrators timely understand user needs and system usage situations.

[0174] For example, the system may use the following sentiment classification criteria:

[0175] Positive: Users express satisfaction or praise;

[0176] Neutral: Users put forward objective suggestions or ask questions;

[0177] Negative: Users express dissatisfaction or complaints;

[0178] In this way, the system can continuously optimize the user experience and improve user satisfaction.

[0179] The system of the present invention further includes a system evaluation module 8, which is communicatively connected to the data processing module 2 and the report generation module 3. The introduction of the system evaluation module 8 reflects the self-optimization ability of the system of the present invention, ensuring that the system can continuously improve its performance and accuracy.

[0180] The system evaluation module 8 is first responsible for collecting feedback information from users on the reports generated by the system. This feedback information may come from multiple channels, such as the rating function in the user interface, questionnaires, and the actual usage behavior data of users, etc. For example, the system may track the following metrics:

[0181] The reported click-through rate and reading duration;

[0182] The score given by users to the report content (such as accuracy, practicality, comprehensibility, etc.);

[0183] The proportion of users taking actions based on the report suggestions;

[0184] Based on this feedback information, the system evaluation module 8 will evaluate the accuracy and effectiveness of the system. The system of the present invention adopts a multi-dimensional evaluation model, comprehensively considering factors such as the accuracy of the report, user satisfaction, and the effectiveness of intervention suggestions. The evaluation result can be represented by a comprehensive score:

[0185] Score = w1·Accuracy + w2·Satisfaction + w3·Effectiveness,

[0186] where w1, w2, and w3 are the weights of each factor, Accuracy represents the accuracy score of the report, Satisfaction represents the user satisfaction score, and Effectiveness represents the effectiveness score of the intervention suggestions.

[0187] According to the evaluation result, the system evaluation module 8 will automatically adjust the analysis algorithm of the data processing module 2 and the report template of the report generation module 3. This automatic adjustment mechanism adopts machine learning technology. Specifically, the system uses a reinforcement learning algorithm to optimize its decision-making process.

[0188] For example, when optimizing the report template, the system may try different content organization methods or expression methods and adjust these choices according to user feedback. This process can be modeled by a Markov decision process (MDP):

[0189] State S: The current report template configuration;

[0190] Action A: The adjustments made to the template (such as changing the content order, modifying the wording, etc.);

[0191] Reward R: The change in the user feedback score;

[0192] Transition probability P: The probability of changing from one template configuration to another;

[0193] The goal of the system is to find an optimal policy π * such that the long-term cumulative reward is maximized:

[0194]

[0195] where γ is the discount factor, R tis the reward obtained at time step t.

[0196] Through this continuous learning and optimization mechanism, the system of the present invention can continuously improve its performance, adapt to changes in user needs, and maintain its advancement and effectiveness in the field of student mental health monitoring.

[0197] In summary, the student mental health monitoring system based on multimodal assessment and multi-role reporting of the present invention provides an intelligent, safe, efficient and evolving solution for student mental health management through its comprehensive functional design, advanced technology implementation and self-optimization capabilities. The system can not only accurately assess students' mental health status, but also protect user privacy, provide a personalized user experience, and continuously improve its performance through self-assessment and optimization. These characteristics give the system of the present invention significant advantages in improving students' mental health, optimizing the allocation of educational resources, and supporting educational decision-making, and are of great significance to promoting the informatization and modernization of education.

[0198] In addition, the system of the present invention can realize the following contents: 1. Consultation appointment system, including students making appointments for consultation, teachers completing case consultation record viewing, editing and management, teachers calling questionnaires online, etc.; 2. Guidance education system, including teachers uploading audio and video materials independently and distributing them to students for learning, online courses across time periods and continuous use, customized classification of online courses, customized configuration of visibility and availability of different courses to different types and levels of users, and pushing courses to designated users; 3. High school subject selection system, for student subjects, a variety of rich and complete subject assessment professional reports, support for high school students' enrollment planning assessment reports and including major and school recommendations; 4. Municipal-level mental education system, including checking the early warning status of schools in each region, uploading course content in the background, and distributing it to different stages of study, etc.

[0199] It should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A student mental health monitoring system based on multimodal assessment and multi-role reports, characterized in that , including: A data acquisition module, which is used for: Collecting the standardized psychological test questionnaire data of students; Collecting the behavioral experiment data of students, including the implicit association experiment data; Collecting the relevant data of students' parents, teachers and living environment; A data processing module, which is communicatively connected to the data acquisition module and is used for: Receiving the multi-modal assessment data sent by the data acquisition module; Based on the multi-modal assessment data, using machine learning algorithms for comprehensive analysis and prediction; A report generation module, which is communicatively connected to the data processing module and is used for: Based on the analysis results of the data processing module, generating personalized analysis reports for different roles; Generating comparative analysis reports among schools, regions and provinces / municipalities; A resource allocation module, which is communicatively connected to the report generation module and is used for: Based on the analysis results of the report generation module, dynamically adjusting the configuration of psychological teachers; A user interface module, which is communicatively connected to the report generation module and the resource allocation module and is used for: Displaying the corresponding analysis reports to users of different roles; Providing a visual display of the resource allocation results.

2. The system according to claim 1, wherein , the data acquisition module includes: A questionnaire data acquisition unit, which is used for collecting the standardized psychological test questionnaire data; A behavioral experiment data acquisition unit, which is used for collecting behavioral experiment data such as implicit association experiments; An environment data acquisition unit, which is used for collecting the relevant data of students' parents, teachers and living environment; Among them, the behavioral experiment data acquisition unit obtains the implicit attitude data of students by recording the reaction time and error rate of students in the implicit association experiment.

3. The system according to claim 1, wherein , the data processing module includes: A data preprocessing unit, which is used for cleaning, standardizing and feature extraction of the multi-modal assessment data; A machine learning analysis unit, which is used for analyzing and predicting the preprocessed data using machine learning algorithms; Among them, the machine learning analysis unit adopts a deep learning model to conduct a fusion analysis of the questionnaire data, behavioral experiment data and environment data, and generates a comprehensive evaluation result of the mental health status of students.

4. The system according to claim 1, characterized in that , the report generation module includes: A personalized report generation unit, which is used for generating targeted analysis reports according to different user roles; A comparative analysis report generation unit, which is used for generating comparative analysis reports among schools, regions and provinces / municipalities; Among them, the personalized report generation unit selectively presents relevant mental health indicators and suggestions according to different user roles to improve the practicality and operability of the report.

5. The system according to claim 1, wherein , the resource allocation module includes: A demand analysis unit, which is used for evaluating the demand for mental health resources based on the analysis results of the report generation module; A resource allocation unit, which is used for dynamically adjusting psychological teachers according to the demand analysis results. Specifically, students can make appointments for psychological counseling through the system, and teachers can also record psychological counseling and configure case reports through the system; Among them, the resource allocation unit adopts an intelligent optimization algorithm, comprehensively considers the number of students, the severity of psychological problems and the existing resource status, and generates an optimal resource allocation plan.

6. The system according to claim 1, wherein , further including: An early warning module, which is communicatively connected to the data processing module and the report generation module and is used for: Set a mental health risk threshold based on the analysis results of the data processing module; Generate a warning message when a student's mental health indicator exceeds the preset threshold; Send the warning message to the report generation module and include it in the analysis reports of relevant roles.

7. The system according to claim 6, wherein , The warning module further includes: A risk level assessment unit for grading warning messages and classifying mental health risks into three levels: low, medium, and high; An intervention suggestion generation unit for generating corresponding intervention suggestions according to the risk level and including them in the analysis report.

8. The system according to claim 1, wherein , Further includes: A data security module communicatively connected to the data collection module and the data processing module for: Encrypting and storing the collected data; Controlling access rights to the data; Performing desensitization processing on the data processing process to protect students' privacy.

9. The system according to claim 1, characterized in that , The user interface module includes: A role recognition unit for recognizing user roles, including students, parents, teachers, and school administrators; A report display unit for displaying corresponding analysis reports according to user roles; An interaction function unit for providing functions such as report interpretation, problem feedback, and suggestion submission.

10. The system according to claim 1, wherein , Further includes: A system evaluation module communicatively connected to the data processing module and the report generation module for: Collecting feedback information from users on the reports generated by the system; Evaluating the accuracy and effectiveness of the system based on the feedback information; Automatically adjusting the analysis algorithm of the data processing module and the report template of the report generation module according to the evaluation results.

Images