Mental health risk identification and screening method, system and program product

Abstract

The invention belongs to the technical field of psychological assessment, and particularly discloses a psychological health risk identification and screening method and system and a program product, and the method comprises the steps: collecting a psychological health scale evaluation result of a target object and a drawing test result for comprehensive statistics, and determining a corresponding psychological dimension set; according to the method, emotion recognition is carried out by collecting interview video data and EEG signals of a target object, two emotion recognition results are obtained, a risk psychological dimension set is screened out by using the two emotion recognition results, and a psychological health risk level is judged by correcting an evaluation total score. Through statistics and analysis of multi-source data, comprehensive and diversified assessment and risk screening can be carried out on the mental health condition of an individual, effectiveness and accuracy of an assessment result are ensured, so that psychological workers can comprehensively master the mental health condition of the individual, the psychological workers are assisted to identify potential mental health risks of the individual, and the mental health risk assessment efficiency is improved. And an important basis is provided for individual psychological health education and psychological intervention of psychological workers.

Description

Technical Field

[0001] The present invention belongs to the technical field of psychological assessment, and specifically relates to a method, system and program product for identifying and screening mental health risks. Background Art

[0002] With the development of society, while people are pursuing material wealth, mental health problems are becoming increasingly prominent. The main causes of mental health problems include social pressure, family environment, academic pressure, social isolation, genetic factors, lifestyle, etc. Therefore, it is very important to conduct mental health screening for individuals. Through mental health screening, we can enhance the individual's mental health awareness, discover potential problems in time, and prevent the occurrence of psychological crises.

[0003] At present, when psychological workers conduct mental health screening for individuals, they mainly use common psychological assessment methods, that is, using conversations, observations and test questions to analyze the individual's psychological phenomena. However, this psychological assessment method is relatively simple, and it does not fully grasp the individual's mental health symptoms. The effectiveness and accuracy of the assessment also need to be improved. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system and program product for identifying and screening mental health risks to solve the above-mentioned problems existing in the prior art.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, a method for identifying and screening mental health risks is provided, comprising:

[0007] Obtain the target object's mental health scale assessment results, drawing test results, interview video data, and brain wave measurement signals;

[0008] Determine the scores of each psychological dimension and the total score of the psychological assessment according to the results of the mental health scale assessment, and determine the corresponding first psychological dimension set when the score of any psychological dimension exceeds the set dimension score threshold and the total score of the psychological assessment exceeds the set total assessment score threshold;

[0009] Extracting a second psychological dimension set from the drawing test results, and forming a third psychological dimension set using the first psychological dimension set and the second psychological dimension set;

[0010] Extracting target audio signals and a number of sequentially arranged frame images based on interview video data, extracting image features of each frame image to form an image feature set, and extracting a frequency spectrum feature set of the target audio signal;

[0011] Inputting the image feature set and the spectrum feature set into a preset first emotion recognition network model to perform emotion recognition, and obtaining a corresponding first emotion label and a first normalization parameter;

[0012] Extracting a target frequency signal from the brain wave measurement signal, and determining the frequency domain power characteristics of the target frequency signal;

[0013] Inputting the frequency domain power feature into a preset second emotion recognition network model for emotion recognition, and obtaining a corresponding second emotion label and a second normalization parameter;

[0014] A risk psychological dimension set is screened out from the third psychological dimension set according to the first emotion label and the second emotion label, and a coincidence factor is determined according to the first emotion label, the first normalization parameter, the second emotion label and the second normalization;

[0015] The total score of the psychological assessment and the overlap factor are used to determine the mental health risk level, and the risk psychological dimension set and mental health risk level are output.

[0016] In a possible design, the psychological dimension scores and the total psychological assessment score are determined based on the results of the mental health scale assessment, including:

[0017] Determine the scores of all test items in the mental health scale assessment results and the scores of the test items corresponding to each psychological dimension;

[0018] The scores of each psychological dimension are counted according to the scores of the test items corresponding to each psychological dimension, and the total score of the psychological assessment is calculated using the scores of all test items.

[0019] In a possible design, the target audio signal and a plurality of sequentially arranged frame images are extracted based on the interview video data, and the image features of each frame image are extracted to form an image feature set, and the spectrum feature set of the target audio signal is extracted, including:

[0020] Extracting a target audio signal of a target object from the interview video data, and extracting a number of sequentially arranged frame images from the interview video data at a set time interval;

[0021] Extract image features from each frame of image, and combine the extracted image features of each frame of image in a corresponding order to obtain an image feature set;

[0022] Perform short-time Fourier transform processing on the target audio signal to obtain the corresponding spectrum feature set.

[0023] In a possible design, extracting the target frequency signal from the brain wave measurement signal and determining the frequency domain power feature of the target frequency signal includes:

[0024] Extracting a target frequency signal from the brain wave measurement signal, wherein the target frequency signal is one or more of an alpha wave signal, a beta wave signal, a gamma wave signal, a delta wave signal, and a theta wave signal;

[0025] Perform frequency band decomposition on the target frequency signal, and extract the power spectrum density and differential entropy of each frame signal in each frequency band after the frequency band decomposition;

[0026] The frequency domain power characteristics corresponding to the target frequency signal are obtained by combining power spectral density and differential entropy.

[0027] In one possible design, the first emotion recognition network model adopts a trained and tested DialogueRNN model, the first emotion label includes several first emotion categories, the first normalization parameter includes a first probability parameter corresponding to each first emotion category, and the second emotion recognition network model adopts a trained and tested AMDET model, the second emotion label includes several second emotion categories, and the second normalization parameter includes a second probability parameter corresponding to each second emotion category.

[0028] In a possible design, the step of selecting a risk psychological dimension set from a third psychological dimension set according to the first emotion label and the second emotion label, and determining a coincidence factor according to the first emotion label, the first normalization parameter, the second emotion label and the second normalization includes:

[0029] De-recombining a first emotion classification in the first emotion tag whose corresponding first probability parameter exceeds a set probability value and a second emotion classification in the second emotion tag whose corresponding second probability parameter exceeds a set probability value, to obtain an emotion classification set including several emotion classifications;

[0030] Based on the set emotion classification and psychological dimension association rules, the psychological dimensions associated with the corresponding emotion classification in the emotion classification set in the third psychological dimension set are extracted as risk psychological dimensions, and the risk psychological dimension set is obtained by combining the risk psychological dimensions.

[0031] Using the first emotion classification and the second emotion classification that overlap with each other in the first emotion label and the second emotion label to form an overlapping emotion classification pair;

[0032] Determine the overlapping emotion classification of each overlapping emotion classification pair, and add the first probability parameter corresponding to the first emotion classification and the second probability parameter corresponding to the second emotion classification in each overlapping emotion classification pair to obtain the classification overlap factor of the corresponding overlapping emotion classification pair;

[0033] The classification overlap factors of each overlapping emotion classification pair are added together to obtain the overlap factor.

[0034] In a possible design, the use of the total score of the psychological assessment and the coincidence factor to determine the mental health risk level includes:

[0035] Multiply the total score of the psychological assessment by the coincidence factor to obtain the total score of the assessment correction;

[0036] The assessment corrected total score is substituted into a preset risk level table for matching to determine the corresponding mental health risk level. The risk level table contains several mental health risk levels and the assessment corrected total score ranges corresponding to each mental health risk level.

[0037] In a second aspect, a mental health risk identification and screening system is provided, including an information acquisition unit, an evaluation and statistical unit, a dimension combination unit, a first extraction unit, a first identification unit, a second extraction unit, a second identification unit, a risk screening unit, and a determination output unit, wherein:

[0038] An information acquisition unit, used to obtain the target object's mental health scale evaluation results, drawing test results, interview video data, and brain wave measurement signals;

[0039] An evaluation statistics unit is used to determine the scores of each psychological dimension and the total score of the psychological evaluation according to the evaluation results of the mental health scale, and when the score of any psychological dimension exceeds the set dimension score threshold and the total score of the psychological evaluation exceeds the set total evaluation score threshold, determine the corresponding first psychological dimension set;

[0040] A dimension combination unit, used for extracting a second psychological dimension set from the drawing test result, and forming a third psychological dimension set by using the first psychological dimension set and the second psychological dimension set;

[0041] A first extraction unit is used to extract a target audio signal and a plurality of sequentially arranged frame images based on the interview video data, extract image features of each frame image to form an image feature set, and extract a frequency spectrum feature set of the target audio signal;

[0042] A first recognition unit, used for inputting the image feature set and the spectrum feature set into a preset first emotion recognition network model to perform emotion recognition, and obtain a corresponding first emotion label and a first normalization parameter;

[0043] A second extraction unit is used to extract a target frequency signal from the brain wave measurement signal and determine a frequency domain power feature of the target frequency signal;

[0044] A second recognition unit is used to input the frequency domain power feature into a preset second emotion recognition network model to perform emotion recognition, and obtain a corresponding second emotion label and a second normalization parameter;

[0045] a risk screening unit, configured to screen out a risk psychological dimension set from a third psychological dimension set according to the first emotion label and the second emotion label, and determine a coincidence factor according to the first emotion label, the first normalization parameter, the second emotion label, and the second normalization;

[0046] The determination output unit is used to determine the mental health risk level using the total score of the psychological assessment and the overlap factor, and output a set of risk psychological dimensions and the mental health risk level.

[0047] In a third aspect, a mental health risk identification and screening system is provided, comprising:

[0048] A memory for storing instructions;

[0049] A processor is used to read the instructions stored in the memory and execute any one of the methods described in the first aspect according to the instructions.

[0050] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored on the computer-readable storage medium, and when the instructions are executed on a computer, the computer executes any one of the methods described in the first aspect. In addition, a computer program product is provided, and when the computer program product is executed on a computer, the computer executes any one of the methods described in the first aspect.

[0051] Beneficial effects: The present invention collects the target object's mental health scale assessment results and drawing test results for comprehensive statistics, determines the corresponding psychological dimension set, collects the target object's interview video data and EEG signals for emotion recognition, obtains two emotion recognition results, and uses the two emotion recognition results to screen out the risk psychological dimension set, and corrects the total assessment score to determine the mental health risk level, which can achieve comprehensive and accurate mental health risk identification screening. Through statistics and analysis of multi-source data, the present invention can conduct a comprehensive and diverse assessment and risk screening of an individual's mental health status, ensure the effectiveness and accuracy of the assessment results, so that psychological workers can fully grasp the individual's mental health status, assist psychological workers in identifying the individual's potential mental health risks, and provide an important basis for psychological workers to provide mental health education and psychological intervention for individuals. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1This is a schematic diagram of the steps of the method in Example 1 of the present invention;

[0054] Figure 2 This is a schematic diagram of the structure of the system in Example 2 of the present invention;

[0055] Figure 3 Schematic diagram of the system structure in Example 3 of the present invention. DETAILED DESCRIPTION

[0056] It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. The specific structures and functional details disclosed herein are only used to describe the exemplary embodiments of the present invention. However, the present invention can be embodied in many alternative forms, and it should not be understood that the present invention is limited to the embodiments set forth herein.

[0057] It should be understood that, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments can be understood according to specific circumstances.

[0058] In the following description, certain details are provided to facilitate a complete understanding of the example embodiments. However, it will be appreciated by those of ordinary skill in the art that the example embodiments may be implemented without these certain details. For example, devices may be shown in block diagrams to avoid obscuring the examples with unnecessary details. In other embodiments, well-known processes, structures, and techniques may not be shown in unnecessary detail to avoid obscuring the embodiments.

[0059] Embodiment 1:

[0060] This embodiment provides a method for identifying and screening mental health risks, which can be applied to corresponding processing terminals, such as Figure 1 As shown, the method comprises the following steps:

[0061] S1. Obtain the target subject’s mental health scale assessment results, drawing test results, interview video data, and EEG measurement signals.

[0062] In specific implementation, the psychological worker may ask the target object to fill in the mental health scale in advance, summarize the results of the filling, and score each test item to obtain the target object's mental health scale evaluation results. The mental health scale can use an international general scale, such as SCL-90. The psychological worker may ask the target object to take a painting test in advance, and then summarize the target object's painting test results. The painting test is a method of analyzing the object's psychology through paintings. The psychological worker can analyze the content, color, lines, composition, etc. in the painting to evaluate the object's psychology. The psychological worker may ask the target object to conduct a psychological interview in advance, and record the target object's interview video during the interview. The video content includes the target object's facial expression and movement changes, and the video data contains the recorded audio signal. The psychological worker can perform synchronous brain wave measurement on the target object during other psychological tests to obtain the target object's brain wave measurement signal. Finally, the psychological worker summarizes and uploads the target object's mental health scale evaluation results, painting test results, interview video data, and brain wave measurement signals to the processing terminal so that the processing terminal can perform subsequent analysis and processing.

[0063] S2. Determine the scores of each psychological dimension and the total score of the psychological assessment based on the results of the mental health scale assessment, and determine the corresponding first psychological dimension set when the score of any psychological dimension exceeds the set dimension score threshold and the total score of the psychological assessment exceeds the set total assessment score threshold.

[0064] In specific implementation, the processing terminal first determines the scores of all test items in the mental health scale evaluation results and the scores of the test items corresponding to each psychological dimension. Different psychological dimensions, such as anxiety, depression, hostility, etc., correspond to different test items, and multiple test items are set for each psychological dimension. Then, the scores of each psychological dimension are counted according to the scores of the test items corresponding to each psychological dimension, and the total score of the psychological assessment is calculated using the scores of all test items. When the scores of one or more psychological dimensions exceed the set dimension score threshold and the total score of the psychological assessment exceeds the set assessment total score threshold, the psychological dimensions whose corresponding psychological dimension scores exceed the set dimension score threshold are extracted to form the first psychological dimension set.

[0065] S3. Extract the second psychological dimension set from the drawing test results, and use the first psychological dimension set and the second psychological dimension set to form a third psychological dimension set.

[0066] In specific implementation, the drawing test result includes one or more psychological dimensions of the target object that may have corresponding conditions as assessed by the psychologist based on the drawing test. The processing terminal can extract the corresponding psychological dimensions from the drawing test result to form a second psychological dimension set. Then, the first psychological dimension set and the second psychological dimension set are used to form a third psychological dimension set.

[0067] S4. Extract the target audio signal and a number of sequentially arranged frame images based on the interview video data, extract the image features of each frame image to form an image feature set, and extract the spectrum feature set of the target audio signal.

[0068] In specific implementation, the processing terminal can extract the target audio signal of the target object from the interview video data, and extract a number of sequentially arranged frame images from the interview video data according to a set time interval; then perform image feature extraction on each frame image, and combine the extracted image features of each frame image in a corresponding order to obtain an image feature set; perform short-time Fourier transform processing on the target audio signal to obtain a corresponding spectrum feature set. The short-time Fourier transform divides the audio signal into multiple short time periods (frames) and performs Fourier transform on each frame to obtain a corresponding spectrum diagram, which can finally be combined to obtain a spectrum feature set.

[0069] S5. Input the image feature set and the spectrum feature set into a preset first emotion recognition network model to perform emotion recognition, and obtain a corresponding first emotion label and a first normalization parameter.

[0070] During specific implementation, the processing terminal may input the image feature set and the spectrum feature set into a preset first emotion recognition network model for emotion recognition to obtain the corresponding first emotion label and the first normalization parameter. The first emotion recognition network model may adopt a trained and tested bimodal emotion detection model, such as the DialogueRNN model. The first emotion label includes several first emotion classifications, and the first normalization parameter includes a first probability parameter corresponding to each first emotion classification. The model may finally output the first probability parameter corresponding to the corresponding first emotion classification through a softmax function, and the first probability parameter is a normalized value processed by the softmax function.

[0071] S6. Extract the target frequency signal from the brain wave measurement signal, and determine the frequency domain power characteristics of the target frequency signal.

[0072] In specific implementation, the processing terminal can extract the corresponding target frequency signal from the brain wave measurement signal, including one or more of α wave signal, β wave signal, γ wave signal, δ wave signal and θ wave signal. The electroencephalogram (EEG) signal contains a large number of features related to emotional state. The frequency components in the EEG signal can reflect different states of brain activity. For example, the α wave signal is enhanced when a person is in a relaxed state, the β wave signal is enhanced when the brain is active and highly concentrated, the γ wave signal is associated with the hyperactive state of the brain, and the intensity of the θ wave signal can reflect whether a person feels happy. Therefore, the power characteristics of each wave signal in the frequency domain, such as power spectral density (PSD) and differential entropy (DE), can be used to identify the emotion of the object.

[0073] After extracting the target frequency signal, the processing terminal can perform time window segmentation and frequency band decomposition of the signal, and extract the power spectrum density and differential entropy of each frame signal in each frequency band after the frequency band decomposition, and then use the power spectrum density and differential entropy combination to obtain the frequency domain power characteristics corresponding to the target frequency signal.

[0074] S7. Input the frequency domain power feature into a preset second emotion recognition network model to perform emotion recognition, and obtain a corresponding second emotion label and a second normalization parameter.

[0075] During specific implementation, the processing terminal may input the frequency domain power features into a preset second emotion recognition network model for emotion recognition to obtain the corresponding second emotion label and second normalization parameter. The second emotion recognition network model may adopt a trained and tested AMDET model. The second emotion label includes several second emotion classifications. The second normalization parameter includes a second probability parameter corresponding to each second emotion classification. The model may finally output the second probability parameter corresponding to the corresponding second emotion classification through a softmax function. The second probability parameter is a normalized value processed by the softmax function.

[0076] S8. Filter out a risk psychological dimension set from the third psychological dimension set according to the first emotion label and the second emotion label, and determine the overlap factor according to the first emotion label, the first normalization parameter, the second emotion label and the second normalization.

[0077] In specific implementation, the processing terminal removes and combines the first emotion classification corresponding to the first probability parameter exceeding the set probability value in the first emotion label and the second emotion classification corresponding to the second probability parameter exceeding the set probability value in the second emotion label to obtain an emotion classification set containing several emotion classifications. Then, based on the set emotion classification and psychological dimension association rule, the psychological dimension associated with the corresponding emotion classification in the emotion classification set in the third psychological dimension set is extracted as the risk psychological dimension, and the risk psychological dimension set is obtained by combining each risk psychological dimension. At the same time, the first emotion classification and the second emotion classification that overlap with each other in the first emotion label and the second emotion label can be used to form an overlapping emotion classification pair. Then determine the overlapping emotion classification of each overlapping emotion classification pair, and add the first probability parameter corresponding to the first emotion classification in each overlapping emotion classification pair and the second probability parameter corresponding to the second emotion classification to obtain the classification overlap factor of the corresponding overlapping emotion classification pair. Finally, add the classification overlap factors of each overlapping emotion classification pair to obtain the overlap factor.

[0078] S9. Use the total score of psychological assessment and the overlap factor to determine the mental health risk level, and output the risk psychological dimension set and mental health risk level.

[0079] In specific implementation, the processing terminal can multiply the total score of the psychological assessment by the overlap factor to obtain the total score of the assessment correction. Then the total score of the assessment correction is substituted into the preset risk level table for matching to determine the corresponding mental health risk level. The risk level table contains several mental health risk levels and the total score interval of the assessment correction corresponding to each mental health risk level. Finally, the risk psychological dimension set and the mental health risk level are output to assist psychological workers in making a comprehensive and accurate assessment of the mental health status of the target object based on the risk psychological dimension set and the mental health risk level.

[0080] Through statistics and analysis of multi-source data, this method can conduct a comprehensive and diverse assessment and risk screening of an individual's mental health status, ensuring the effectiveness and accuracy of the assessment results, so that psychological workers can fully grasp the individual's mental health status, assist psychological workers in identifying the individual's potential mental health risks, and provide an important basis for psychological workers to provide mental health education and psychological intervention for the individual.

[0081] Embodiment 2:

[0082] This embodiment provides a mental health risk identification and screening system. Figure 2 As shown, it includes an information acquisition unit, an evaluation and statistics unit, a dimension combination unit, a first extraction unit, a first identification unit, a second extraction unit, a second identification unit, a risk screening unit and a determination output unit, wherein:

[0083] An information acquisition unit, used to obtain the target object's mental health scale evaluation results, drawing test results, interview video data, and brain wave measurement signals;

[0084] An evaluation statistics unit is used to determine the scores of each psychological dimension and the total score of the psychological evaluation according to the evaluation results of the mental health scale, and when the score of any psychological dimension exceeds the set dimension score threshold and the total score of the psychological evaluation exceeds the set total evaluation score threshold, determine the corresponding first psychological dimension set;

[0085] A dimension combination unit, used for extracting a second psychological dimension set from the drawing test result, and forming a third psychological dimension set by using the first psychological dimension set and the second psychological dimension set;

[0086] A first extraction unit is used to extract a target audio signal and a plurality of sequentially arranged frame images based on the interview video data, extract image features of each frame image to form an image feature set, and extract a frequency spectrum feature set of the target audio signal;

[0087] A first recognition unit, used for inputting the image feature set and the spectrum feature set into a preset first emotion recognition network model to perform emotion recognition, and obtain a corresponding first emotion label and a first normalization parameter;

[0088] A second extraction unit is used to extract a target frequency signal from the brain wave measurement signal and determine a frequency domain power feature of the target frequency signal;

[0089] A second recognition unit is used to input the frequency domain power feature into a preset second emotion recognition network model to perform emotion recognition, and obtain a corresponding second emotion label and a second normalization parameter;

[0090] a risk screening unit, configured to screen out a risk psychological dimension set from a third psychological dimension set according to the first emotion label and the second emotion label, and determine a coincidence factor according to the first emotion label, the first normalization parameter, the second emotion label, and the second normalization;

[0091] The determination output unit is used to determine the mental health risk level using the total score of the psychological assessment and the overlap factor, and output a set of risk psychological dimensions and the mental health risk level.

[0092] Embodiment 3:

[0093] This embodiment provides a mental health risk identification and screening system. Figure 3 As shown, at the hardware level, it includes:

[0094] Data interface, used to establish data connection between the processor and the external data terminal;

[0095] A memory for storing instructions;

[0096] A processor is used to read the instructions stored in the memory and execute the mental health risk identification and screening method in Example 1 according to the instructions.

[0097] Optionally, the system further includes an internal bus, through which the processor, the memory and the data interface can be interconnected, and the internal bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc.

[0098] The memory may include, but is not limited to, random access memory (RAM), read only memory (ROM), flash memory, first input first output (FIFO) and / or first in last out (FILO) memory, etc. The processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0099] Embodiment 4:

[0100] This embodiment provides a computer-readable storage medium, on which instructions are stored, and when the instructions are executed on a computer, the computer executes the mental health risk identification and screening method in Embodiment 1. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive, and / or a memory stick, etc. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0101] This embodiment also provides a computer program product, which, when executed on a computer, executes the mental health risk identification and screening method in Embodiment 1. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0102] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying and screening mental health risks, characterized in that: include: Obtain the target object's mental health scale assessment results, drawing test results, interview video data, and brain wave measurement signals; Determine the scores of each psychological dimension and the total score of the psychological assessment according to the results of the mental health scale assessment, and determine the corresponding first psychological dimension set when the score of any psychological dimension exceeds the set dimension score threshold and the total score of the psychological assessment exceeds the set total assessment score threshold; Extracting a second psychological dimension set from the drawing test results, and forming a third psychological dimension set using the first psychological dimension set and the second psychological dimension set; Extracting target audio signals and a number of sequentially arranged frame images based on interview video data, extracting image features of each frame image to form an image feature set, and extracting a frequency spectrum feature set of the target audio signal; Inputting the image feature set and the spectrum feature set into a preset first emotion recognition network model to perform emotion recognition, and obtaining a corresponding first emotion label and a first normalization parameter; Extracting a target frequency signal from the brain wave measurement signal, and determining the frequency domain power characteristics of the target frequency signal; Inputting the frequency domain power feature into a preset second emotion recognition network model for emotion recognition, and obtaining a corresponding second emotion label and a second normalization parameter; A risk psychological dimension set is screened out from the third psychological dimension set according to the first emotion label and the second emotion label, and a coincidence factor is determined according to the first emotion label, the first normalization parameter, the second emotion label and the second normalization; The total score of the psychological assessment and the overlap factor are used to determine the mental health risk level, and the risk psychological dimension set and mental health risk level are output.

2. A method for identifying and screening mental health risks according to claim 1, characterized in that: The psychological dimension scores and the total psychological assessment score are determined based on the results of the mental health scale, including: Determine the scores of all test items in the mental health scale assessment results and the scores of the test items corresponding to each psychological dimension; The scores of each psychological dimension are counted according to the scores of the test items corresponding to each psychological dimension, and the total score of the psychological assessment is calculated using the scores of all test items.

3. A method for identifying and screening mental health risks according to claim 1, characterized in that: The method of extracting a target audio signal and a plurality of sequentially arranged frame images based on the interview video data, extracting image features of each frame image to form an image feature set, and extracting a frequency spectrum feature set of the target audio signal includes: Extracting a target audio signal of a target object from the interview video data, and extracting a number of sequentially arranged frame images from the interview video data at a set time interval; Extract image features from each frame of image, and combine the extracted image features of each frame of image in a corresponding order to obtain an image feature set; Perform short-time Fourier transform on the target audio signal to obtain the corresponding spectrum feature set.

4. A method for identifying and screening mental health risks according to claim 1, characterized in that: The step of extracting the target frequency signal from the brain wave measurement signal and determining the frequency domain power characteristics of the target frequency signal includes: Extracting a target frequency signal from the brain wave measurement signal, wherein the target frequency signal is one or more of an alpha wave signal, a beta wave signal, a gamma wave signal, a delta wave signal, and a theta wave signal; Perform frequency band decomposition on the target frequency signal, and extract the power spectrum density and differential entropy of each frame signal in each frequency band after the frequency band decomposition; The frequency domain power characteristics corresponding to the target frequency signal are obtained by combining power spectral density and differential entropy.

5. A method for identifying and screening mental health risks according to claim 1, characterized in that: The first emotion recognition network model adopts the trained and tested DialogueRNN model, the first emotion label includes several first emotion categories, and the first normalization parameter includes a first probability parameter corresponding to each first emotion category. The second emotion recognition network model adopts the trained and tested AMDET model, the second emotion label includes several second emotion categories, and the second normalization parameter includes a second probability parameter corresponding to each second emotion category.

6. A method for identifying and screening mental health risks according to claim 5, characterized in that: The step of selecting a risk psychological dimension set from a third psychological dimension set according to the first emotion label and the second emotion label, and determining a coincidence factor according to the first emotion label, the first normalization parameter, the second emotion label and the second normalization includes: De-recombining a first emotion classification in the first emotion tag whose corresponding first probability parameter exceeds a set probability value and a second emotion classification in the second emotion tag whose corresponding second probability parameter exceeds a set probability value, to obtain an emotion classification set including several emotion classifications; Based on the set emotion classification and psychological dimension association rules, the psychological dimensions associated with the corresponding emotion classification in the emotion classification set in the third psychological dimension set are extracted as risk psychological dimensions, and the risk psychological dimension set is obtained by combining the risk psychological dimensions. Using the first emotion classification and the second emotion classification that overlap with each other in the first emotion label and the second emotion label to form an overlapping emotion classification pair; Determine the overlapping emotion classification of each overlapping emotion classification pair, and add the first probability parameter corresponding to the first emotion classification and the second probability parameter corresponding to the second emotion classification in each overlapping emotion classification pair to obtain the classification overlap factor of the corresponding overlapping emotion classification pair; The classification overlap factors of each overlapping emotion classification pair are added together to obtain the overlap factor.

7. A method for identifying and screening mental health risks according to claim 1, characterized in that: The use of the total score of the psychological assessment and the coincidence factor to determine the mental health risk level includes: Multiply the total score of the psychological assessment by the coincidence factor to obtain the total score of the assessment correction; The assessment corrected total score is substituted into a preset risk level table for matching to determine the corresponding mental health risk level. The risk level table contains several mental health risk levels and the assessment corrected total score ranges corresponding to each mental health risk level.

8. A mental health risk identification and screening system, characterized in that: It includes an information acquisition unit, an evaluation and statistics unit, a dimension combination unit, a first extraction unit, a first identification unit, a second extraction unit, a second identification unit, a risk screening unit and a determination output unit, wherein: An information acquisition unit, used to obtain the target object's mental health scale evaluation results, drawing test results, interview video data, and brain wave measurement signals; An evaluation statistics unit is used to determine the scores of each psychological dimension and the total score of the psychological evaluation according to the evaluation results of the mental health scale, and when the score of any psychological dimension exceeds the set dimension score threshold and the total score of the psychological evaluation exceeds the set total evaluation score threshold, determine the corresponding first psychological dimension set; A dimension combination unit, used for extracting a second psychological dimension set from the drawing test result, and forming a third psychological dimension set by using the first psychological dimension set and the second psychological dimension set; A first extraction unit is used to extract a target audio signal and a plurality of sequentially arranged frame images based on the interview video data, extract image features of each frame image to form an image feature set, and extract a frequency spectrum feature set of the target audio signal; A first recognition unit, used for inputting the image feature set and the spectrum feature set into a preset first emotion recognition network model to perform emotion recognition, and obtain a corresponding first emotion label and a first normalization parameter; A second extraction unit is used to extract a target frequency signal from the brain wave measurement signal and determine a frequency domain power feature of the target frequency signal; A second recognition unit is used to input the frequency domain power feature into a preset second emotion recognition network model to perform emotion recognition, and obtain a corresponding second emotion label and a second normalization parameter; a risk screening unit, configured to screen out a risk psychological dimension set from a third psychological dimension set according to the first emotion label and the second emotion label, and determine a coincidence factor according to the first emotion label, the first normalization parameter, the second emotion label, and the second normalization; The determination output unit is used to determine the mental health risk level using the total score of the psychological assessment and the overlap factor, and output a set of risk psychological dimensions and the mental health risk level.

9. A mental health risk identification and screening system, characterized in that: include: A memory for storing instructions; A processor is used to read the instructions stored in the memory and execute the mental health risk identification and screening method described in any one of claims 1-7 according to the instructions.

10. A computer program product, characterized in that When the computer program product is run on a computer, the mental health risk identification and screening method described in any one of claims 1-7 is executed.

Images