Risk screening method for mental health
By building a psychological risk testing environment and deep learning models, and dynamically generating personalized screening processes, we solve the efficiency and accuracy issues of traditional mental health screening and achieve efficient and personalized mental health risk identification.
Patent Information
- Application Number
- CN202510816100.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-18
- Publication Date
- 2025-09-19
AI Technical Summary
Traditional mental health screening methods are unable to capture the dynamic interactions of physiology, behavior, environment, and psychology, are inefficient, cannot be dynamically adjusted according to individual characteristics, lack causal chain explanations, and are difficult to identify advanced psychological risks.
Build a psychological risk testing environment, conduct psychological health risk tests for basic and advanced items, dynamically generate screening processes, delete avoidance items, use deep learning models to analyze psychological health indexes, and generate personalized screening paths.
It improves the efficiency and accuracy of mental health screening, dynamically adapts to individual characteristics, and solves the problems of data comprehensiveness and clinical interpretability of traditional screening.
Smart Images

Figure CN120674079A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of mental health analysis, and more particularly, to a risk screening method for mental health. Background Art
[0002] Mental health risk screening, a key tool for early identification of mental disorders, currently relies primarily on traditional questionnaires, physiological indicator monitoring, and expert interviews. These technologies are widely used in scenarios such as grassroots screening, corporate employee health management, and campus psychological protection. However, as mental health needs become more complex and precise, their inherent flaws are becoming increasingly prominent.
[0003] Traditional methods often rely on fixed-structure questionnaires, which fail to capture the dynamic interactions of "physiology-behavior-environment-psychology." Furthermore, a comprehensive screening process requires all tests to be completed regardless of baseline risk, resulting in low efficiency and a poor user experience. Screening for advanced psychological risks (such as PTSD and ADHD) lacks a scientific sequencing logic, and traditional methods are implemented at a fixed level, unable to dynamically adjust based on individual characteristics. Screening results based on traditional statistical models lack causal chain explanations, making it difficult to answer key questions such as "Why is social mental health risk judged to be high?" This limits their application in clinical scenarios.
[0004] Based on the above problems, the present invention proposes a risk screening method for mental health. Summary of the Invention
[0005] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a risk screening method for mental health.
[0006] To achieve the above object, the present invention provides the following technical solutions: A risk screening method for mental health, comprising the following steps: Step 1: Build a psychological risk testing environment for various mental health risks; Step 2: First, arrange for the test-taker to take a psychological risk test for each basic mental health risk to screen whether the test-taker has the basic mental health risk; Step 3: Determine the advanced mental health risk screening process for the test-taker, and conduct mental health risk screening for each advanced item on the test-taker in accordance with the advanced mental health risk screening process; Step 4: After conducting a mental health risk screening for an advanced item on each test subject, determine whether there are any avoidance advanced items in the advanced mental health risk screening process. If there are any avoidance advanced items, delete the avoidance advanced items in the advanced mental health risk screening process and then conduct mental health risk screening for subsequent advanced items.
[0007] Furthermore, it is screened whether the test subject has mental health risks of basic items: obtain the basic mental health feature set of the test subject corresponding to each basic item, import the basic mental health feature set of each basic item into the corresponding mental health analysis model, and then derive the mental health index of each basic item, set the mental health threshold index, and when the mental health index of a basic item is less than the mental health threshold index, the test subject is screened for mental health risks of the basic item.
[0008] Furthermore, the steps for generating a basic mental health feature set corresponding to a basic item of the tester are as follows: arrange the tester to conduct a psychological risk test in a psychological risk test environment of a mental health risk; during the psychological risk test, collect various collected data corresponding to the mental health risk in real time; after the psychological risk test, perform data preprocessing and feature extraction on various collected data to extract the features of various collected data, and combine the features of various collected data in a set manner to form a basic mental health feature set.
[0009] Furthermore, the steps for determining the advanced mental health risk screening process of the tester are as follows: obtain the advanced screening process index for each advanced item of the tester, and sort all advanced items in descending order according to the value of the advanced screening process index to generate the advanced mental health risk screening process of the tester.
[0010] Furthermore, the steps for obtaining the advanced screening process index of the advanced item are as follows: select an advanced item, and simultaneously determine all risk reference testers for the tester; when a risk reference tester has a mental health index for the advanced item, mark the risk reference tester as an advanced screening basis tester, and calculate the sum and average of the mental health indexes of all advanced screening basis testers to calculate the average advanced screening basis index; when a risk reference tester does not have a mental health index for the advanced item, mark the risk reference tester as an advanced screening avoider, obtain the total number of advanced screening avoiders, and calculate the advanced screening process index of the advanced item based on the average advanced screening basis index and the total number of advanced screening avoiders.
[0011] Furthermore, all risk reference testers for the tester are determined: all testers who have previously completed the mental health risk screening are marked as historical testers, the risk reference index of each historical tester is obtained, and a risk reference threshold index is set. When the risk reference index of a historical tester is greater than or equal to the risk reference threshold index, the historical tester is marked as a risk reference tester.
[0012] Furthermore, the steps for obtaining the risk reference index of historical testers are as follows: select a historical tester, obtain the basic mental health feature set of the historical tester corresponding to each basic item, obtain the basic mental health feature set of the tester corresponding to each basic item, and then determine the basic reference degree of each basic item, set the basic reference dividing degree, when the basic reference degree of a basic item is greater than or equal to the basic reference dividing degree, increase the reference quantity by one, and mark the basic item as a reference basic item, calculate the sum and average of the basic reference degrees of each reference basic item, and calculate the average target basic reference degree. Finally, multiply the reference quantity by the average target basic reference degree to calculate the risk reference index of the historical tester.
[0013] Furthermore, after each mental health risk screening of an advanced item is performed on a test subject, the mental health index of the advanced item is obtained. When the mental health index of the advanced item is less than the mental health threshold index, all advanced items that do not have a relationship with the advanced item in the advanced mental health relationship tree are marked as avoidance advanced items, and all avoidance advanced items in the advanced mental health risk screening process are deleted. After deletion, the mental health risk screening of the remaining advanced items is performed on the test subject in turn according to the advanced mental health risk screening process; When the mental health index of the advanced item is greater than or equal to the mental health threshold index, the test subject will be screened for mental health risks in the next advanced item according to the advanced mental health risk screening process.
[0014] Compared with the prior art, the present invention has the following beneficial effects: The method of the present invention builds a psychological risk testing environment for various mental health risks, which facilitates the subsequent intuitive and accurate analysis of the tester's data performance for different mental health risks. First, through the psychological risk test of the tester's basic mental health risks, it can not only quickly analyze whether the tester has basic mental health risks, but also intelligently generate a "high risk first, avoidance items later" mental health risk testing process for the risk reference tester, and dynamically delete the avoidance advanced items during the execution of the process to eliminate interference, ensuring that the screening path always targets the tester's real risk combination. This hierarchical, progressive and dynamically optimized screening mechanism not only improves the efficiency of basic item screening, but also enhances the accuracy of risk identification through personalized advanced processes, avoids invalid screening, and solves the core pain points of traditional mental health screening in terms of data comprehensiveness, dynamic adaptability, and clinical interpretability. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a principle flow chart of the present invention; Figure 2 Screening flowchart for advanced mental health risk screening. DETAILED DESCRIPTION
[0016] like Figure 1-Figure 2 , a risk screening method for mental health, the steps are as follows: Step 1: Build a psychological risk testing environment for various mental health risks (the mental health risks that require psychological risk testing include but are not limited to the following: social, depression, anxiety, stress, PTSD, ADHD, and bipolar disorder. Among them, social, depression, anxiety, and stress are basic mental health risks, and PTSD, ADHD, bipolar disorder, and other items are advanced mental health risks. Each mental health risk requires a corresponding psychological risk testing environment); The principles of setting up a psychological risk testing environment are basically similar. For different mental health risks, the Unity HDRP rendering pipeline creates typical test scenarios such as offices and campuses, embeds physiological data collection interfaces, and ARFoundation simulates augmented reality stressors (such as sudden pop-up work task notifications) to configure scene parameters. For example, for social interactions, the crowd density is set to 1.5 people / m. 2 , social distance of 1-1.5m, noise level of 50-60dB. For stress testing, set the task density to 2 tasks / 15 minutes, the task interface complexity to a multi-level menu (≥10 buttons), environmental interference to pop-up ads (≥3 times / 5 minutes), and time pressure to a countdown to 30% of the remaining time. Configure interaction trigger events. For example, using social interaction as an example, add a sudden call-out event (triggered without warning). For example, using hostility and aggression as an example, add an NPC intentional collision event (with a 0% probability of apology). Use Unreal Engine's physics engine (such as PhysX) to simulate physical phenomena and motion laws, such as gravity, collision, and friction. Add controllers through Blueprint programming or C++ programming to facilitate human intervention in the simulation process. Design multiple interaction methods and use Blueprint visual scripting or C++ scripting to implement logical control and interactive functions in the environment, thereby building a psychological risk testing environment for various mental health risks.
[0017] Step 2: First, arrange for the test-taker to take a psychological risk test for each basic mental health risk item, generate a basic mental health feature set corresponding to each basic item, and further screen whether the test-taker has a mental health risk in the basic item; The steps for generating a basic mental health feature set corresponding to a basic item of a tester are as follows: arrange the tester to conduct a psychological risk test in a psychological risk test environment of a mental health risk (such as arrange the tester to conduct a psychological risk test in a social psychological risk test environment). During the psychological risk test, collect various collected data corresponding to the mental health risk in real time (each mental health risk corresponds to multiple collected data, and the collected data corresponding to different mental health risks may have repeated collected data or non-repeated collected data. Taking the social psychological risk test environment as an example, it is necessary to collect the tester's eye tracking data, EEG data, electromyographic activity data, skin conductance data and other collected data. Taking the depression psychological risk test environment as an example, it is necessary to collect EEG data, body movement data, skin conductance data and other collected data). After the psychological risk test, perform data preprocessing and feature extraction on each collected data to extract the features of each collected data, and combine the features of each collected data into a basic mental health feature set in a set manner; Further screen whether the test subject has mental health risks for basic items: obtain the basic mental health feature set of the test subject corresponding to each basic item, import the basic mental health feature set of each basic item into the corresponding mental health analysis model, and then derive the mental health index of each basic item, set the mental health threshold index (the setting of the mental health threshold index is the result of the combined effect of statistical standards, clinical guidelines, individual characteristics, and engineering optimization). When the mental health index of a basic item is less than the mental health threshold index, the test subject is screened for mental health risks for the basic item (when the mental health index of a basic item is greater than or equal to the mental health threshold index, it means that the test subject does not have mental health risks for the basic item); Each mental health risk corresponds to a mental health analysis model. Whether basic or advanced, each mental health analysis model is built on a deep learning model. The following uses social mental health risk as an example to explain the process of building a mental health analysis model: A deep learning model is built, and multiple basic mental health feature sets of social networks are collected. Using these basic mental health feature sets as the baseline data, the constructed deep learning model is trained. During this process, each basic mental health feature set is assigned a mental health index, ranging from 1 to 50. The magnitude of the mental health index has a clear meaning: a higher value indicates a lower mental health risk for the testee. The multiple basic mental health feature sets of social networks are then divided into training, validation, and test sets in a specific ratio of 70%:20%:10%. The deep learning model is first repeatedly trained using the training set. During the training process, the validation set is used to verify the model's performance during the training phase. Based on the verification results, the model parameters are adjusted and the model structure is optimized to achieve greater accuracy and stability. Finally, a mental health analysis model for social mental health risks is built.
[0018] Step 3: Determine all risk reference testers for the test subject, and then determine the advanced mental health risk screening process for the test subject (the advanced mental health risk screening process has all advanced mental health risks sorted in order), and conduct mental health risk screening for each advanced item on the test subject in turn according to the advanced mental health risk screening process; Determine all risk reference testers for the tester: Mark all testers who have previously completed the mental health risk screening as historical testers, obtain the risk reference index of each historical tester, set the risk reference threshold index (the risk reference threshold index is set based on statistical standards and historical data), and when the risk reference index of a historical tester is greater than or equal to the risk reference threshold index, mark the historical tester as a risk reference tester (when the risk reference index of a historical tester is less than the risk reference threshold index, do not mark it); The steps for obtaining the risk reference index of a historical tester are as follows: select a historical tester, obtain the basic mental health feature set corresponding to each basic item of the historical tester, obtain the basic mental health feature set corresponding to each basic item of the (current) tester, and then determine the basic reference degree of each basic item, set the basic reference cutoff degree (the basic reference cutoff degree is greater than 0, and the basic reference cutoff degree is set based on statistical standards and historical data), when the basic reference degree of a basic item is greater than or equal to the basic reference cutoff degree, increase the reference quantity by one, and mark the basic item as a reference basic item, calculate the sum and average of the basic reference degrees of each reference basic item, and calculate the average target basic reference degree. Finally, multiply the reference quantity by the average target basic reference degree to calculate the risk reference index of the historical tester; The steps for determining the basic reference degree of a basic item are as follows: select a basic item, vectorize the basic psychological health feature set of the historical test takers corresponding to the basic item, and convert it into a vector M=(m1,m2,...,m n ), vectorize the basic mental health feature set of the (current) tester corresponding to the basic item and convert it into a vector P=(p1,p2,...,p n ), Calculate the basic reference degree of the basic item; The steps for determining the advanced mental health risk screening process of the test subject are as follows: obtaining the advanced screening process index of each advanced item for the test subject, and sorting all the advanced items in descending order according to the values of the advanced screening process index to generate the advanced mental health risk screening process of the test subject; The steps for obtaining the advanced screening process index of the advanced items are as follows: select an advanced item, and simultaneously determine all risk reference testers. When a risk reference tester has a mental health index for the advanced item, the risk reference tester is marked as an advanced screening basis person, and the mental health indexes of all advanced screening basis persons are summed and averaged to calculate the average advanced screening basis index, which is marked as CR (be). When a risk reference tester does not have a mental health index for the advanced item, the risk reference tester is marked as an advanced screening avoider, and the total number of advanced screening avoiders is marked as PE (dc). Calculate the advanced screening process index of the advanced item Among them, gg1 is the advanced screening basis coefficient, gg2 is the avoider number coefficient, the advanced screening basis coefficient is 0.28, and the avoider number coefficient is 1.09.
[0019] Step 4: After performing a mental health risk screening for one advanced item on each test subject, obtain the mental health index of the advanced item. When the mental health index of the advanced item is less than the mental health threshold index, mark all advanced items that do not have a relationship with the advanced item in the advanced mental health relationship tree as avoidance advanced items, delete all avoidance advanced items in the advanced mental health risk screening process, and after deletion, perform mental health risk screening for the remaining advanced items on the test subject in turn according to the advanced mental health risk screening process; When the mental health index of the advanced item is greater than or equal to the mental health threshold index, the test subject will be screened for the mental health risk of the next advanced item according to the advanced mental health risk screening process; The advanced mental health relationship tree contains various advanced items in the form of a relationship tree. If two advanced items cannot exist at the same time, then there is no relationship between the two advanced items in the advanced mental health relationship tree. For example, acute stress disorder (ASD) and post-traumatic stress disorder (PTSD) are mutually exclusive (that is, they cannot exist at the same time), then there is no relationship between ASD and PTSD in the advanced mental health relationship tree. If two advanced items can exist at the same time, then there is a relationship between the two advanced items in the advanced mental health relationship tree. For example, post-traumatic stress disorder (PTSD) and attention deficit hyperactivity disorder (ADHD) are not mutually exclusive, then there is a relationship between PTSD and ADHD in the advanced mental health relationship tree.
[0020] The above method builds a psychological risk testing environment for various mental health risks, which facilitates the subsequent intuitive and accurate analysis of the tester's data performance for different mental health risks. First, through the psychological risk testing of the tester's basic mental health risks, it can not only quickly analyze whether the tester has basic mental health risks, but also intelligently generate a "high risk first, avoidance items later" mental health risk testing process for the risk reference tester. During the execution of the process, the avoidance advanced items are dynamically deleted to eliminate interference, ensuring that the screening path always targets the tester's true risk combination. This hierarchical, progressive and dynamically optimized screening mechanism not only improves the efficiency of basic item screening, but also enhances the accuracy of risk identification through personalized advanced processes, avoids invalid screening, and solves the core pain points of traditional mental health screening in data comprehensiveness, dynamic adaptability, and clinical interpretability.
[0021] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.
[0022] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0023] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0024] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0025] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.
Claims
1. A risk screening method for mental health, characterized in that: Here are the steps: Step 1: Build a psychological risk testing environment for various mental health risks; Step 2: First, arrange for the test-taker to take a psychological risk test for each basic mental health risk to screen whether the test-taker has the basic mental health risk; Step 3: Determine the advanced mental health risk screening process for the test-taker, and conduct mental health risk screening for each advanced item on the test-taker in accordance with the advanced mental health risk screening process; Step 4: After conducting a mental health risk screening for an advanced item on each test subject, determine whether there are any avoidance advanced items in the advanced mental health risk screening process. If there are any avoidance advanced items, delete the avoidance advanced items in the advanced mental health risk screening process and then conduct mental health risk screening for subsequent advanced items.
2. A risk screening method for mental health according to claim 1, characterized in that: Screen out whether the test subject has mental health risks for basic items: obtain the basic mental health feature set of the test subject corresponding to each basic item, import the basic mental health feature set of each basic item into the corresponding mental health analysis model, and then derive the mental health index of each basic item, set the mental health threshold index, and when the mental health index of a basic item is less than the mental health threshold index, screen out the test subject for mental health risks for the basic item.
3. A risk screening method for mental health according to claim 2, characterized in that: The steps for generating a basic mental health feature set corresponding to a basic item of the tester are as follows: arrange the tester to take a psychological risk test in a psychological risk test environment of a mental health risk; during the psychological risk test, collect various collected data corresponding to the mental health risk in real time; after the psychological risk test, perform data preprocessing and feature extraction on various collected data to extract the features of various collected data, and combine the features of various collected data in a set manner to form a basic mental health feature set.
4. A risk screening method for mental health according to claim 1, characterized in that: The steps for determining the advanced mental health risk screening process for the tester are as follows: obtain the advanced screening process index for each advanced item of the tester, and sort all advanced items in descending order according to the value of the advanced screening process index to generate the advanced mental health risk screening process for the tester.
5. A risk screening method for mental health according to claim 4, characterized in that: The steps for obtaining the advanced screening process index of advanced items are as follows: select an advanced item, and simultaneously determine all risk reference testers for the tester; when a risk reference tester has a mental health index for the advanced item, mark the risk reference tester as an advanced screening basis tester, and calculate the sum and average of the mental health indexes of all advanced screening basis testers to calculate the average advanced screening basis index; when a risk reference tester does not have a mental health index for the advanced item, mark the risk reference tester as an advanced screening avoider, obtain the total number of advanced screening avoiders, and calculate the advanced screening process index of the advanced item based on the average advanced screening basis index and the total number of advanced screening avoiders.
6. A risk screening method for mental health according to claim 5, characterized in that: Determine all risk reference testers for the tester: mark all testers who have previously completed the mental health risk screening as historical testers, obtain the risk reference index of each historical tester, set the risk reference threshold index, and when the risk reference index of a historical tester is greater than or equal to the risk reference threshold index, mark the historical tester as a risk reference tester.
7. A risk screening method for mental health according to claim 6, characterized in that: The steps for obtaining the risk reference index of historical testers are as follows: select a historical tester, obtain the basic mental health feature set of the historical tester corresponding to each basic item, obtain the basic mental health feature set of the tester corresponding to each basic item, and then determine the basic reference degree of each basic item, set the basic reference dividing degree, when the basic reference degree of a basic item is greater than or equal to the basic reference dividing degree, increase the reference quantity by one, and mark the basic item as a reference basic item, calculate the sum and average of the basic reference degrees of each reference basic item, and calculate the average target basic reference degree. Finally, multiply the reference quantity by the average target basic reference degree to calculate the risk reference index of the historical tester.
8. A risk screening method for mental health according to claim 1, characterized in that: After performing a mental health risk screening for an advanced item on each test subject, obtain the mental health index of the advanced item. When the mental health index of the advanced item is less than the mental health threshold index, mark all advanced items that do not have a relationship with the advanced item in the advanced mental health relationship tree as avoidance advanced items, delete all avoidance advanced items in the advanced mental health risk screening process, and after deletion, perform mental health risk screening for the remaining advanced items on the test subject in turn according to the advanced mental health risk screening process; When the mental health index of the advanced item is greater than or equal to the mental health threshold index, the test subject will be screened for mental health risks in the next advanced item according to the advanced mental health risk screening process.
Citation Information
Patent Citations
Mental health risk screening system assisted by big data
CN118538403A
Method and system for rapidly screening health status of overseas students
CN118629637A