Multi-mode psychological disease risk screening robot, equipment and storage medium

Abstract

The invention relates to the technical field of multi-modal psychological disease risk screening robots, and particularly discloses a multi-modal psychological disease risk screening robot, equipment and a storage medium, and the multi-modal psychological disease risk screening robot comprises a screening robot distribution module, a psychological disease risk analysis module, a database, an initial diagnosis report generation module and a screening accuracy evaluation module; the psychological disease risk corresponding to the psychological counseling person is analyzed in a multi-level mode from the physiological level, the voice level and the face level, and meanwhile the accuracy of report generation of the psychological disease risk screening robot is analyzed according to the change situation of a doctor on an initial diagnosis report generated by the psychological disease risk screening robot. The psychological state and the psychological disease risk of the psychological counseling person can be comprehensively and accurately assessed, various data resources are fully utilized, deviation or omission of assessment of the psychological state of the psychological counseling person is avoided, and then the difficulty and uncertainty of psychological disease risk identification are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of multimodal mental illness risk screening robots, and in particular to a multimodal mental illness risk screening robot, device and storage medium. Background Art

[0002] Traditional mental health screening methods mainly rely on self-assessment questionnaires and structured interviews, which have problems such as low efficiency and inaccurate results. The person being tested may provide untrue answers due to concealment, resulting in inaccurate assessment results. Therefore, a more accurate and realistic psychological screening method is needed.

[0003] The existing technology has also proposed some solutions. For example, the Chinese patent with publication number CN113239050B discloses a medical and psychological data management system, method, equipment and storage medium. It combines the requirements of the construction of a social psychological service system, integrates the patient's basic information, health information, document information, report management and other functions, and integrates mental health projects and related information into the system to meet the development and demand of one-stop service requirements for mental health archive management. The depression risk screening system and method based on EEG signals in virtual reality scenes with publication number CN106407733A collects, processes and analyzes EEG signals in real time under the stimulation of virtual reality scenes through an EEG acquisition system, and finally screens for depression risks through data mining methods.

[0004] The following problems still exist in the existing technology: 1. The psychological counselor's physiological data, voice data and facial data during the consultation process are not combined to analyze the psychological counselor's corresponding mental illness risk from multiple angles and levels at the physiological level, voice level and facial level. It is impossible to comprehensively and accurately assess the psychological state and mental illness risk of the psychological counselor, and it is impossible to make full use of various data resources, resulting in deviations or omissions in the assessment of the psychological state of the psychological counselor, which in turn increases the difficulty and uncertainty of identifying the risk of mental illness.

[0005] 2. The accuracy of the report generated by the mental illness risk screening robot was not analyzed in combination with the doctor's changes to the initial diagnostic report generated by the mental illness risk screening robot. If the doctor's changes were ignored and the robot's report was directly adopted, it might lead to deviations in the diagnostic results, which would in turn affect subsequent treatment plans and psychological intervention measures. If only the robot's report was relied upon, the patient's personalized needs might be ignored, resulting in poor treatment results. Summary of the invention

[0006] In view of this, in order to solve the problems raised in the above background technology, a multimodal mental illness risk screening robot, device and storage medium are proposed.

[0007] The objectives of the present invention can be achieved through the following technical solutions: In the first aspect, the present invention provides a multimodal mental illness risk screening robot, including: a screening robot allocation module, used to record the mental illness risk screening robots placed in front of the mental health counseling department in the target hospital as screening robots, extract the current working status of each screening robot, and allocate the corresponding screening robot to the current psychological counselor, and record it as the target screening robot.

[0008] The mental illness risk analysis module is used to extract the mental health information corresponding to the current psychological counselor during the screening process of the target screening robot, and analyze the mental illness risk index corresponding to the current psychological counselor.

[0009] The database is used to store positive and negative vocabulary in mental health counseling, store the normal speaking speed range in mental health counseling, and store the normal volume range in mental health counseling.

[0010] The initial diagnosis report generation module is used to generate an initial diagnosis report corresponding to the current psychological consultant and send the initial diagnosis report to the target doctor system.

[0011] The screening accuracy assessment module is used to extract the target doctor's change information on the initial diagnosis report corresponding to the current psychological counselor, and to extract the doctor's change information on the initial diagnosis report generated by each screening robot during each historical screening process, to evaluate whether the screening report generated by each screening robot is accurate, and to provide feedback if it is inaccurate.

[0012] The second aspect of the present invention provides a device, the device comprising a processor, a memory and a communication bus. The memory stores a computer-readable program that can be executed by the processor. The communication bus realizes the connection and communication between the processor and the memory. When the processor executes the computer-readable program, the steps in the multimodal mental illness risk screening robot are implemented.

[0013] The third aspect of the present invention proposes a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the multimodal mental illness risk screening robot described in the present invention.

[0014] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention can comprehensively and accurately assess the psychological state and psychological disease risk of the psychological counselor by combining the psychological counselor's physiological data, voice data and facial data during the consultation process to analyze the psychological counselor's corresponding psychological disease risk from multiple angles and levels at the physiological level, voice level and facial level, and make full use of various data resources to avoid bias or omissions in the assessment of the psychological state of the psychological counselor, thereby reducing the difficulty and uncertainty of identifying the risk of psychological disease.

[0015] (2) The present invention analyzes the accuracy of the report generated by the mental illness risk screening robot by combining the doctor's changes to the initial diagnostic report generated by the mental illness risk screening robot, thereby avoiding ignoring the doctor's changes and directly adopting the report of the mental illness risk screening robot, which may lead to deviations in the diagnosis results and affect subsequent treatment plans and psychological intervention measures. By combining the diagnosis results of the doctor and the mental illness risk screening robot, it helps to meet the personalized needs of patients and improve the later treatment effects of psychological counselors. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of the system module structure connection of the present invention.

[0018] Figure 2 The figure is a schematic diagram of the structure of a device of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Example 1

[0021] See also Figure 1 As shown, the present invention provides a multimodal mental illness risk screening robot, including: a screening robot allocation module, a mental illness risk analysis module, a database, an initial diagnosis report generation module and a screening accuracy assessment module.

[0022] The screening robot allocation module is connected to the mental illness risk analysis module, the mental illness risk analysis module is connected to the database, the mental illness risk analysis module is connected to the initial diagnosis report generation module, and the initial diagnosis report generation module is connected to the screening accuracy assessment module.

[0023] The screening robot allocation module is used to record the mental illness risk screening robots placed in front of the mental health counseling department in the target hospital as screening robots, extract the current working status of each screening robot, and allocate the corresponding screening robot to the current psychological counselor, and record it as the target screening robot.

[0024] It should be noted that the current working state includes an idle state, a busy state and a maintenance state.

[0025] In a specific embodiment of the present invention, the specific method of assigning a corresponding screening robot to the current psychological counselor is: extracting each screening robot in an idle state from the current working state of each screening robot, and recording them as idle robots, and randomly assigning an idle robot to the current psychological counselor.

[0026] The mental illness risk analysis module is used to extract the mental health information corresponding to the current psychological consultant during the screening process of the target screening robot, and analyze the mental illness risk index corresponding to the current psychological consultant.

[0027] In a specific embodiment of the present invention, the mental health information includes physiological data, voice data and facial data, wherein the physiological data includes heart rate, blood pressure and skin conductance at each monitoring time point, the voice data includes text, speaking speed and volume during the question-and-answer process, and the facial data includes the number of blinks in each monitoring time period and the interval duration corresponding to each blink.

[0028] It should be noted that the heart rate, blood pressure and skin conductance at each monitoring time point are monitored by a wearable multi-functional health monitoring bracelet, and the text, speaking speed and volume during the question-and-answer process are collected by the comprehensive voice monitoring equipment in the mental illness risk screening robot.

[0029] It should also be noted that the method for collecting the number of blinks in each monitoring time period and the interval duration corresponding to each blink is: the target screening robot performs video capture on the face of the current psychological counselor during the screening process, divides the captured video into pictures in frames, and obtains the time point corresponding to each picture, and scans the pupil area of ​​the current psychological counselor in each picture through a pupil scanner. When the pupil area is 0, it indicates that the current psychological counselor is in a completely closed-eye state. When the pupil area is consistent with the pupil area in a resting state, it indicates that the operator is in a completely open-eye state. The pictures of the current psychological counselor in each completely closed-eye state and each completely open-eye state are extracted, and these pictures are sorted in chronological order. The time points corresponding to the adjacent pictures in the completely closed-eye state and the completely open-eye state are subtracted to obtain the duration of the blink. Similarly, the blink duration corresponding to each blink is analyzed.

[0030] In a specific embodiment of the present invention, the specific process of analyzing the mental illness risk index corresponding to the current psychological consultant is as follows: A1, extracting the heart rate, blood pressure and skin conductance at each monitoring time point, the text, speech speed and volume during the question-and-answer process, the number of blinks in each monitoring time period and the interval duration corresponding to each blink from the mental health information corresponding to the current psychological consultant during the screening process of the target screening robot, and calculating the risk index of the current psychological consultant at the physiological level accordingly. , Voice-level risk index and risk index at the facial level .

[0031] In a specific embodiment of the present invention, the specific process of calculating the risk index of the current psychological consultant corresponding to the physiological level is as follows: B1, recording the heart rate of the current psychological consultant at each monitoring time point as ,in, Indicates the number of the monitoring time point, .

[0032] B2. Extract the maximum and minimum values ​​of the heart rate at each monitoring time point corresponding to the current counselor, and record them as and .

[0033] B3. Calculate the heart rate fluctuation index corresponding to the current psychological consultant , ,in, Indicates the current psychological counselor's corresponding Heart rate at each monitoring time point, and Indicates the stable deviation of heart rate and the extreme deviation of heart rate set as reference. Indicates the number of monitoring time points, Represents a natural constant.

[0034] B4. Based on the blood pressure of the current psychological consultant at each monitoring time point, the blood pressure fluctuation index corresponding to the current psychological consultant is calculated in the same way as the heart rate fluctuation index corresponding to the current psychological consultant. .

[0035] B5. Calculate the normality of the skin conductance of the current psychological consultant based on the skin conductance of each monitoring time point corresponding to the current psychological consultant .

[0036] It should be noted that the specific process of calculating the normality of the skin conductance corresponding to the current psychological consultant is as follows: the skin conductance corresponding to each monitoring time point of the current psychological consultant is recorded as .

[0037] Calculate the normal skin conductance of the current counselor , ,in, and Represents the normal skin conductance and skin conductance deviation of the set reference, respectively.

[0038] B6. Calculate the current psychological counselor's corresponding physiological risk index , ,in, , and They represent the risk assessment weights of the set heart rate fluctuation index, blood pressure fluctuation index and skin conductance normality corresponding to the physiological level. .

[0039] In a specific embodiment of the present invention, The setting value of is 0.34. The setting value of is 0.33. The setting value of is 0.33. In practical applications, in order to comprehensively assess the physiological risks of the consultant, it is usually necessary to comprehensively consider multiple indicators such as heart rate fluctuation index, blood pressure fluctuation index and skin conductance normality. These indicators can complement each other and jointly reflect the consultant's autonomic nervous activation response and emotional state.

[0040] In a specific embodiment of the present invention, the specific process of calculating the risk index of the current psychological consultant at the voice level is as follows: C1, extracting the text in the question-and-answer process corresponding to the current psychological consultant, and performing word segmentation processing on the text to obtain the keywords in the question-and-answer process corresponding to the current psychological consultant, and matching and comparing each keyword with the positive word library and the negative word library in the psychological health consultation stored in the database, and counting the number of positive words and the number of negative words in the question-and-answer process corresponding to the current psychological consultant, and recording them as and .

[0041] C2. Calculate the standardization of words used by the current psychological counselor , ,in, and Respectively represent the number of positive and negative words used as reference.

[0042] C3. Compare the speaking speed of the current counselor during the question-and-answer process with the normal speaking speed range in mental health counseling stored in the database. If the speaking speed of the current counselor during the question-and-answer process is within the normal speaking speed range in mental health counseling, the normality of the speaking speed of the current counselor is recorded as , otherwise, the normal speaking speed of the current psychological counselor is recorded as , thus obtaining the normal speaking speed corresponding to the current psychological counselor , The value of or ,in, .

[0043] In a specific embodiment of the present invention, The value of is 1. The value of is 0.

[0044] C4. Based on the volume of the current counselor during the question-and-answer process, analyze the normality of the current counselor's volume in the same way as the normality of the current counselor's speech speed. .

[0045] It should be noted that the specific method of analyzing the normal volume of the current psychological counselor by analogy is: comparing the volume of the current psychological counselor during the question-and-answer process with the normal volume range in the mental health counseling stored in the database; if the volume of the current psychological counselor during the question-and-answer process is within the normal volume range in the mental health counseling, then the normal volume of the current psychological counselor is recorded as , otherwise, the normal volume corresponding to the current psychological counselor is recorded as , thus obtaining the normal speaking speed corresponding to the current psychological counselor , The value of or ,in,

[0046] In a specific embodiment of the present invention, The value of is 1. The value of is 0.

[0047] C5. Calculate the risk index of the current psychological consultant at the corresponding voice level , ,in, , and They represent the risk assessment weights of the speech level corresponding to the set word standardization, speech speed normality and volume normality. .

[0048] In a specific embodiment of the present invention, The setting value of is 0.33. The setting value of is 0.33. The setting value of is 0.34. In practical applications, in order to comprehensively assess the risk of the consultant's voice, it is usually necessary to comprehensively consider multiple indicators such as word standardization, normality of speech speed and normality of volume. These indicators can complement each other and jointly reflect the consultant's language expression ability and emotional state. At the same time, it is also necessary to conduct a comprehensive analysis and judgment based on the consultant's specific situation and the purpose of the assessment.

[0049] It should be noted that the specific process of calculating the risk index of the current psychological consultant's corresponding facial level is as follows: the number of blinks of the target screening robot corresponding to each monitoring time period of the current psychological consultant during the screening process is recorded as ,in, Indicates the number of the monitoring time period. .

[0050] The minimum value is extracted from the interval duration corresponding to each blink of the current counselor in each monitoring time period during the screening process of the target screening robot, and recorded as .

[0051] Calculate the risk index of the current psychological consultant corresponding to the face level , ,in, and Respectively represent the reference number of blinks and the duration between blinks. Indicates the number of monitoring time periods.

[0052] A2. Calculate the mental illness risk index corresponding to the current psychological consultant , ,in, , and They represent the weights of the risk indexes at the physiological, voice and facial levels corresponding to the risk assessment of mental illness. .

[0053] In a specific embodiment of the present invention, The setting value of is 0.34. The setting value of is 0.33. The value of is set to 0.33. The risk indexes at the physiological level, voice level, and facial level are all important in calculating the mental illness risk index corresponding to the current psychological counselor. However, to determine which level is more important, it is necessary to conduct a comprehensive analysis and judgment based on factors such as the purpose of the assessment, the specific situation of the counselor, and the professional background. In actual operation, the risk indexes at these levels can be used in combination to comprehensively assess the counselor's mental illness risk.

[0054] The embodiments of the present invention analyze the psychological counselor's corresponding mental illness risks from multiple angles and levels from the physiological level, voice level and facial level by combining the psychological counselor's physiological data, voice data and facial data during the consultation process. This can comprehensively and accurately assess the psychological state and mental illness risks of the psychological counselor, make full use of various data resources, avoid bias or omissions in the assessment of the psychological state of the psychological counselor, and thereby reduce the difficulty and uncertainty of identifying mental illness risks.

[0055] The database is used to store the positive word library and the negative word library in mental health consultation, store the normal speech speed range in mental health consultation, and store the normal volume range in mental health consultation. The data sources in the database of this embodiment are shown in Table 1 below.

[0056] Table 1 Data sources in the database

[0057]

[0058] The initial diagnosis report generating module is used to generate an initial diagnosis report corresponding to the current psychological consultant and send the initial diagnosis report to the target doctor system.

[0059] In a specific embodiment of the present invention, the specific process of generating the initial diagnosis report corresponding to the current psychological counselor is: D1. Compare the physiological risk index, voice risk index, facial risk index and mental illness risk index corresponding to the current psychological counselor with the risk index ranges corresponding to each physiological risk level, each voice risk level, each facial risk level and each mental illness risk level stored in the database, and obtain the physiological risk level, voice risk level, facial risk level and mental illness risk level corresponding to the current psychological counselor.

[0060] D2. Generate a corresponding initial diagnosis report based on the current psychological counselor’s corresponding physiological risk level, voice risk level, facial risk level and mental illness risk level.

[0061] The screening accuracy evaluation module is used to extract the target doctor's change information on the initial diagnosis report corresponding to the current psychological counselor, and to extract the doctor's change information on the initial diagnosis report generated by each screening robot during each historical screening process, to evaluate whether the screening report generated by each screening robot is accurate, and to provide feedback if it is inaccurate.

[0062] It should be noted that the target doctor's modification information for the initial diagnosis report corresponding to the current psychological counselor and the doctor's modification information for the initial diagnosis reports produced by each screening robot in each historical screening process are all extracted from the doctor's diagnosis system.

[0063] In a specific embodiment of the present invention, the change information includes a final physiological level risk level, a final voice level risk level, a final facial level risk level and a final mental illness risk level.

[0064] In a specific embodiment of the present invention, the specific process of evaluating whether the screening reports generated by each screening robot are accurate is as follows: E1. Extract the final physiological risk level, final voice risk level, final facial risk level and final mental illness risk level from the change information of the target doctor's initial diagnosis report corresponding to the current psychological consultant, and compare them with the physiological risk level, voice risk level, facial risk level and mental illness risk level corresponding to the current psychological consultant, respectively. The level with the same risk level is recorded as the accurate diagnosis level, and the number of accurate diagnosis levels corresponding to the current psychological consultant is counted, recorded as .

[0065] E2. Calculate the accuracy of the initial diagnosis report corresponding to the current psychological consultant , ,in, It indicates the percentage of accurate diagnosis levels of the set reference.

[0066] E3. Based on the doctor's change information on the initial diagnosis reports produced by each screening robot in each historical screening process, the accuracy of the initial diagnosis reports produced by each screening robot in each historical screening process is calculated in the same way as the accuracy of the initial diagnosis report corresponding to the current psychological counselor, and the average is calculated to obtain the accuracy of the initial diagnosis reports produced by each screening robot in the historical screening process.

[0067] E4. Compare the accuracy of the initial diagnosis report produced by each screening robot in the historical screening process with the accuracy of the initial diagnosis report of the set reference. If the accuracy of the initial diagnosis report produced by a screening robot in the historical screening process is greater than or equal to the accuracy of the initial diagnosis report of the set reference, it indicates that the screening report generated by the screening robot is accurate; otherwise, it indicates that the screening report generated by the screening robot is inaccurate.

[0068] The embodiment of the present invention analyzes the accuracy of the report generated by the mental illness risk screening robot in combination with the doctor's changes to the initial diagnostic report generated by the mental illness risk screening robot, avoids ignoring the doctor's changes and directly adopting the report of the mental illness risk screening robot, which may lead to deviations in the diagnosis results and affect subsequent treatment plans and psychological intervention measures. The combination of the diagnosis results of the doctor and the mental illness risk screening robot helps to meet the patient's personalized needs and improve the later treatment effect of the psychological counselor.

[0069] Example 2

[0070] See also Figure 2 As shown, the present invention proposes a device, which includes a processor, a memory and a communication bus. The memory stores a computer-readable program that can be executed by the processor. The communication bus realizes the connection and communication between the processor and the memory. When the processor executes the computer-readable program, the steps in the multimodal mental illness risk screening robot are implemented.

[0071] Example 3

[0072] The present invention proposes a storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the steps in the multimodal mental illness risk screening robot described in the present invention.

[0073] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A multimodal mental illness risk screening robot, characterized in that: include: The screening robot allocation module is used to record the mental illness risk screening robots placed in front of the mental health consultation department in the target hospital as screening robots, extract the current working status of each screening robot, and allocate the corresponding screening robot to the current psychological consultant, and record it as the target screening robot; The mental illness risk analysis module is used to extract the mental health information corresponding to the current psychological consultant during the screening process of the target screening robot, and analyze the mental illness risk index corresponding to the current psychological consultant; A database for storing positive and negative word libraries used in mental health counseling, storing a normal speech speed range in mental health counseling, and storing a normal volume range in mental health counseling; An initial diagnosis report generation module is used to generate an initial diagnosis report corresponding to the current psychological consultant and send the initial diagnosis report to the target doctor system; The screening accuracy assessment module is used to extract the target doctor's change information on the initial diagnosis report corresponding to the current psychological counselor, and to extract the doctor's change information on the initial diagnosis report generated by each screening robot during each historical screening process, to evaluate whether the screening report generated by each screening robot is accurate, and to provide feedback if it is inaccurate.

2. A multimodal mental illness risk screening robot according to claim 1, characterized in that: The mental health information includes physiological data, voice data and facial data, wherein the physiological data includes heart rate, blood pressure and skin conductance at each monitoring time point, the voice data includes the text, speaking speed and volume during the question-and-answer process, and the facial data includes the number of blinks in each monitoring time period and the interval duration corresponding to each blink.

3. A multimodal mental illness risk screening robot according to claim 2, characterized in that: The specific process of analyzing the mental illness risk index corresponding to the current psychological consultant is as follows: A1. Extract the heart rate, blood pressure and skin conductance at each monitoring time point, the text, speech speed and volume during the question-and-answer process, the number of blinks in each monitoring time period and the interval length of each blink from the psychological health information of the current psychological counselor during the screening process, and calculate the risk index of the current psychological counselor at the physiological level based on this. , Voice-level risk index and risk index at the facial level ; A2. Calculate the mental illness risk index corresponding to the current psychological consultant , ,in, , and They represent the weights of the risk indexes at the physiological, voice and facial levels corresponding to the risk assessment of mental illness. .

4. A multimodal mental illness risk screening robot according to claim 3, characterized in that: The specific process of calculating the risk index of the current psychological consultant corresponding to the physiological level is: B1. Record the heart rate of the current counselor at each monitoring time point as ,in, Indicates the number of the monitoring time point, ; B2. Extract the maximum and minimum values ​​of the heart rate at each monitoring time point corresponding to the current counselor, and record them as and ; B3. Calculate the heart rate fluctuation index corresponding to the current psychological consultant , ,in, Indicates the current psychological counselor's corresponding Heart rate at each monitoring time point, and Indicates the stable deviation of heart rate and the extreme deviation of heart rate set as reference. Indicates the number of monitoring time points, represents a natural constant; B4. Based on the blood pressure of the current psychological consultant at each monitoring time point, the blood pressure fluctuation index corresponding to the current psychological consultant is calculated in the same way as the heart rate fluctuation index corresponding to the current psychological consultant. ; B5. Calculate the normality of the skin conductance of the current psychological consultant based on the skin conductance of each monitoring time point corresponding to the current psychological consultant ; B6. Calculate the current psychological counselor's corresponding physiological risk index , ,in, , and They represent the risk assessment weights of the set heart rate fluctuation index, blood pressure fluctuation index and skin conductance normality corresponding to the physiological level. .

5. The multimodal mental illness risk screening robot according to claim 3, characterized in that: The specific process of calculating the risk index of the current psychological consultant at the voice level is as follows: C1. Extract the text in the question-and-answer process corresponding to the current psychological consultant, and perform word segmentation on the text to obtain the keywords in the question-and-answer process corresponding to the current psychological consultant, and match and compare each keyword with the positive word library and negative word library in the mental health consultation stored in the database, and count the number of positive words and negative words in the question-and-answer process corresponding to the current psychological consultant, and record them as and ; C2. Calculate the standardization of words used by the current psychological counselor , ,in, and Respectively represent the number of positive and negative words used in the setting reference; C3. Compare the speaking speed of the current counselor during the question-and-answer process with the normal speaking speed range in mental health counseling stored in the database. If the speaking speed of the current counselor during the question-and-answer process is within the normal speaking speed range in mental health counseling, the normality of the speaking speed of the current counselor is recorded as , otherwise, the normal speaking speed of the current psychological counselor is recorded as , thus obtaining the normal speaking speed corresponding to the current psychological counselor , The value of or ,in, ; C4. Based on the volume of the current counselor during the question-and-answer process, analyze the normality of the current counselor's volume in the same way as the normality of the current counselor's speech speed. ; C5. Calculate the risk index of the current psychological consultant at the corresponding voice level , ,in, , and They represent the risk assessment weights of the set word standardization, speech speed normality, and volume normality corresponding to the speech level. .

6. The multimodal mental illness risk screening robot according to claim 4, characterized in that: The specific process of generating the initial diagnosis report corresponding to the current psychological consultant is as follows: D1. Compare the physiological risk index, voice risk index, facial risk index and mental illness risk index of the current psychological consultant with the risk index ranges corresponding to the physiological risk level, voice risk level, facial risk level and mental illness risk level stored in the database, and obtain the physiological risk level, voice risk level, facial risk level and mental illness risk level corresponding to the current psychological consultant; D2. Generate a corresponding initial diagnosis report based on the current psychological counselor’s corresponding physiological risk level, voice risk level, facial risk level and mental illness risk level.

7. A multimodal mental illness risk screening robot according to claim 6, characterized in that: The change information includes a final physiological level risk level, a final voice level risk level, a final facial level risk level and a final mental illness risk level.

8. The multimodal mental illness risk screening robot according to claim 7, characterized in that: The specific process of evaluating whether the screening reports generated by each screening robot are accurate is as follows: E1. Extract the final physiological risk level, final voice risk level, final facial risk level, and final mental illness risk level from the target doctor's change information on the initial diagnosis report corresponding to the current psychological consultant, and compare them with the physiological risk level, voice risk level, facial risk level, and mental illness risk level corresponding to the current psychological consultant, respectively. Record the level with the same risk level as the accurate diagnosis level, and count the number of accurate diagnosis levels corresponding to the current psychological consultant, recorded as ; E2. Calculate the accuracy of the initial diagnosis report corresponding to the current psychological consultant , ,in, It indicates the percentage of accurate diagnosis levels of the set reference; E3. Based on the doctor's modification information of the initial diagnosis reports produced by each screening robot in each historical screening process, the accuracy of the initial diagnosis reports produced by each screening robot in each historical screening process is calculated in the same way as the accuracy of the initial diagnosis reports corresponding to the current psychological counselor, and the average is calculated to obtain the accuracy of the initial diagnosis reports produced by each screening robot in the historical screening process; E4. Compare the accuracy of the initial diagnosis report produced by each screening robot in the historical screening process with the accuracy of the initial diagnosis report of the set reference. If the accuracy of the initial diagnosis report produced by a screening robot in the historical screening process is greater than or equal to the accuracy of the initial diagnosis report of the set reference, it indicates that the screening report generated by the screening robot is accurate; otherwise, it indicates that the screening report generated by the screening robot is inaccurate.

9. A device, characterized in that: include: Processor, memory and communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, the steps in the multimodal mental illness risk screening robot as described in any one of claims 1-8 are implemented.

10. A storage medium, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the multimodal mental illness risk screening robot as described in any one of claims 1-8.

Images