Mental health assessment method, system, device, medium and program product
By using a mental health assessment method based on group characteristics and multi-source data, this approach addresses the issues of strong subjectivity and lack of objective data in existing technologies, enabling more accurate and convenient mental health assessments and providing personalized intervention recommendations.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUA DATA TECH (SHANGHAI) CO LTD
- Filing Date
- 2026-04-13
- Publication Date
- 2026-05-15
AI Technical Summary
Existing mental health assessment methods rely on subjective responses, are greatly influenced by emotional and cultural factors, lack objective behavioral data support, have long assessment cycles, and are difficult to achieve accurate and convenient assessments.
Psychological testing is conducted based on the group characteristics of the test subjects to obtain multi-source data. Psychological characteristic models and clustering models are used for analysis to generate mental health assessment results and provide personalized intervention suggestions.
It improves the objectivity and accuracy of mental health assessments, enhances the differentiation of testing among different groups, and provides more accurate assessment results and personalized intervention measures.
Smart Images
Figure CN122050844A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of information processing technology, specifically to a mental health assessment method, system, device, medium, and program product. Background Technology
[0002] Mental health assessment is a crucial component of public health management and clinical psychological intervention. Currently, the most widely used methods for mental health assessment are still self-report scales (common types include SCL-90, PHQ-9, GAD-7, etc.). These methods have the following limitations: users rely on subjective completion, making them highly susceptible to the influence of emotions, culture, and other factors; the assessment results lack objective behavioral data support, and traditional questionnaires struggle to incorporate objective indicators such as voice, facial expressions, and gestures; and the assessment cycle is long. Therefore, existing mental health assessment methods are insufficient for objectively, accurately, and conveniently assessing users' mental health. Summary of the Invention
[0003] The technical problem to be solved by this disclosure is to overcome the shortcomings of existing psychological assessment methods in that they are difficult to objectively, accurately and conveniently assess users' mental health, and to provide a mental health assessment method, system, device, medium and program product.
[0004] This disclosure solves the above-mentioned technical problems through the following technical solution:
[0005] This disclosure provides a method for assessing mental health, including:
[0006] The test subjects are subjected to psychological tests based on the group characteristics of the test subjects to obtain test data, which includes multi-source data generated by the test subjects during the psychological test process;
[0007] The psychological characteristics of the test subject are obtained based on the test data;
[0008] The psychological health assessment results of the test subjects are obtained based on the aforementioned psychological characteristics.
[0009] Optionally, the step of conducting psychological tests on the test subjects based on the group characteristics of the test subjects to obtain test data includes:
[0010] The type of test task and / or the presentation format of the test task are selected based on the characteristics of the group.
[0011] The test object is tested based on the test task to obtain test data.
[0012] Optionally, obtaining the psychological characteristics of the test subject based on the test data includes:
[0013] The test data is input into a psychological feature model to obtain the psychological features of the test subject. The psychological feature model includes a multi-attention fusion layer.
[0014] Optionally, the step of obtaining the mental health assessment result of the test subject based on the psychological characteristics includes:
[0015] The mental health assessment results of the test subjects are obtained based on the group characteristics and psychological characteristics.
[0016] Optionally, the method further includes:
[0017] Based on the test data, obtain the first behavioral characteristics of the test object;
[0018] The second behavioral characteristics of the test object are obtained based on the historical test data of the test object;
[0019] The first behavioral feature and the second behavioral feature are input into the clustering model to obtain the clustering category of the test object;
[0020] The psychological and behavioral labels of the test subjects are obtained based on the clustering categories.
[0021] Optionally, the method further includes: generating intervention recommendations corresponding to the test subject based on the mental health assessment results.
[0022] Optionally, the method further includes: generating intervention recommendations corresponding to the test subjects based on the mental health assessment results, group characteristics, and / or psychological and behavioral labels.
[0023] Optionally, the group characteristics include at least one of the following: age group, educational background, and stage of disease development.
[0024] Optionally, the test data includes at least one of the following: voice data, video data, motion data, and interactive event streams.
[0025] Optionally, the psychological characteristics include at least one of the following: emotional stability, psychological stress index, social positivity, and attention concentration.
[0026] Optionally, the psychological and behavioral labels include at least one of the following: high-pressure stable type, low-pressure irritable type, and passive avoidant type.
[0027] Optionally, the mental health assessment results include at least one of the following: a mental health index, scores for various psychological characteristics, and a risk level, wherein the risk level corresponds to the mental health index.
[0028] This disclosure also provides a mental health assessment system, including:
[0029] The testing module is used to acquire the group characteristics of the test subjects, and to test the test subjects based on the group characteristics to obtain test data. The test data includes multi-source data generated by the test subjects during the psychological testing process.
[0030] A psychological feature acquisition module is used to acquire the psychological features of the test subject based on the test data.
[0031] The assessment module is used to obtain the mental health assessment results of the test subject based on the psychological characteristics.
[0032] Optionally, the psychological feature acquisition module is further used for:
[0033] The test data is input into a psychological feature model to obtain the psychological features of the test subject. The psychological feature model includes a multi-attention fusion layer.
[0034] Optionally, the evaluation module is also used for
[0035] The mental health assessment results of the test subjects are obtained based on the group characteristics and psychological characteristics.
[0036] Optionally, the system further includes:
[0037] The psychological behavior label acquisition module is used to acquire the first behavioral characteristics of the test subject based on the test data.
[0038] The second behavioral characteristics of the test object are obtained based on the historical test data of the test object;
[0039] The first behavioral feature and the second behavioral feature are input into the clustering model to obtain the clustering category of the test object;
[0040] The psychological and behavioral labels of the test subjects are obtained based on the clustering categories.
[0041] Optionally, the system further includes:
[0042] The suggestion generation module is used to generate intervention suggestions corresponding to the test subjects based on the mental health assessment results.
[0043] Optionally, the system further includes:
[0044] The suggestion generation module is used to generate intervention suggestions corresponding to the test subjects based on the mental health assessment results, group characteristics, and / or psychological behavior labels.
[0045] Optionally, the group characteristics include at least one of the following: age group, educational background, and stage of disease development.
[0046] Optionally, the test data includes at least one of the following: voice data, video data, motion data, and interactive event streams.
[0047] Optionally, the psychological characteristics include at least one of the following: emotional stability, psychological stress index, social positivity, and attention concentration.
[0048] Optionally, the psychological and behavioral labels include at least one of the following: high-pressure stable type, low-pressure irritable type, and passive avoidant type.
[0049] Optionally, the mental health assessment results include at least one of the following: a mental health index, scores for various psychological characteristics, and a risk level, wherein the risk level corresponds to the mental health index.
[0050] This disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and used to run on the processor, characterized in that the processor executes the computer program to implement the aforementioned mental health assessment method.
[0051] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the aforementioned mental health assessment method.
[0052] This disclosure also provides a computer program product, including a computer program, characterized in that the computer program, when executed by a processor, implements the aforementioned mental health assessment method.
[0053] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this disclosure.
[0054] The positive and progressive effects of this disclosure are as follows: conducting psychological tests on test subjects based on the group characteristics of the test subjects can achieve differentiated testing for different groups, thereby improving the universality of the test; obtaining the psychological characteristics of test subjects based on multi-source test data is more accurate and objective than assessment based solely on scales; obtaining the psychological health assessment results of test subjects based on more precise psychological characteristics can further improve the objectivity and accuracy of psychological health assessment. Attached Figure Description
[0055] Figure 1 A flowchart of a mental health assessment method provided as an exemplary embodiment of this disclosure;
[0056] Figure 2 A schematic diagram of a mental health assessment system provided as an exemplary embodiment of this disclosure;
[0057] Figure 3This is a schematic diagram of an electronic device provided as an exemplary embodiment of the present disclosure. Detailed Implementation
[0058] The present disclosure is further illustrated below by way of embodiments, but the present disclosure is not limited to the scope of the embodiments described herein.
[0059] The prefixes such as "first" and "second" used in this disclosure are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this disclosure does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not be construed as an unnecessary limitation. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.
[0060] In this embodiment of the disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good morals.
[0061] Example 1
[0062] Figure 1 A flowchart of a mental health assessment method provided for an exemplary embodiment of this disclosure includes the following steps:
[0063] S1. Conduct psychological tests on the test subjects based on the group characteristics of the test subjects to obtain test data.
[0064] The test data includes multi-source data generated by the test subject during the psychological test. In an optional implementation, the test data includes at least one of the following: voice data, video data, motion data, interactive event streams, etc.
[0065] Group characteristics refer to the group type to which the test subjects belong. In one optional implementation, group characteristics include at least one of the following: age group, educational background, stage of disease development, etc. This approach makes the expressions of test subjects with the same group characteristics more comparable in psychological testing, thereby avoiding testing bias caused by physiological and other differences.
[0066] S2. Obtain the psychological characteristics of the test subjects based on the test data.
[0067] In one alternative implementation, the psychological characteristics include at least one of the following: emotional stability, psychological stress index, social positivity, attention concentration, etc.
[0068] S3. Obtain the psychological health assessment results of the test subjects based on their psychological characteristics.
[0069] In one optional implementation, the mental health assessment results include at least one of the following: a mental health index, scores for various mental characteristics, risk levels, etc., with the risk levels corresponding to the mental health index.
[0070] In an optional implementation, test data of the test object can be acquired using devices such as microphone arrays, cameras, touch screen acquisition units, and interactive interfaces. The test data can be used as a raw signal stream for subsequent calculations or evaluations. The raw signal stream is shown below:
[0071] ;
[0072] Where A(t) is voice data, V(t) is video data, M(t) is motion data, and I(t) is the interactive event stream.
[0073] In one optional implementation, the sampling frequency of the test data can be: 16kHz for audio, 25fps for video, and ≥60Hz for touch sampling.
[0074] In one optional implementation, after obtaining the test data, it is necessary to preprocess the test data. The preprocessing process may include: noise reduction, time synchronization, feature extraction, etc.
[0075] When performing timing synchronization, a hybrid time stamp mechanism can be used to ensure that the synchronization error of test data for different modes is no more than 25ms.
[0076] During feature extraction, a multimodal feature vector set can be obtained. These include acoustic features (speech rate, tone), visual features (AU coding), and motor features (reaction time, trajectory smoothness).
[0077] In an optional implementation, step S1 specifically includes:
[0078] S11. Select the type of test task and the presentation format of the test task based on group characteristics.
[0079] S12. Test the test object based on the test task to obtain test data.
[0080] Selecting the type and presentation format of test tasks based on group characteristics can achieve "cross-group universal design". This can be achieved by introducing differentiated parameter configurations. For example, group characteristics include age groups, which may include children, teenagers, adults, the elderly, etc. By obtaining the group characteristics of the test subjects, that is, the age group to which the test subjects belong, the corresponding test task type and presentation format can be dynamically matched.
[0081] In one optional implementation, the type of test task may include task duration, interaction complexity, and workload intensity; the presentation format of the test task may include voice-modal tasks, behavioral-modal tasks, and scale-modal tasks. For test subjects whose group characteristics are children or adolescents, test tasks with higher interaction complexity, greater workload, and voice or behavioral modalities can be used; for test subjects whose group characteristics are elderly, test tasks with lower interaction complexity, less workload, and voice or behavioral modalities can be used.
[0082] In an optional implementation, step S2 specifically includes:
[0083] Test data is input into a psychological feature model to obtain the psychological features of the test subjects. The psychological feature model includes a multi-attention fusion layer.
[0084] The psychological feature model can be achieved using a Transformer multi-head attention fusion layer.
[0085] In an optional implementation, psychological characteristics may include at least one of the following: emotional stability, psychological stress index, social positivity, and attention concentration. Psychological characteristics may be represented using a psychological dimension vector, which may be:
[0086] ;
[0087] in, For emotional stability; The psychological stress index; For social positivity; To ensure concentration; This is a set of multimodal feature vectors obtained based on test data.
[0088] Specifically, emotional stability can be used to reflect the degree of fluctuation in a test subject's emotional state during the evaluation process. It can be determined based on the following characteristics in speech data: the amplitude of speech rate variation, the degree of fundamental frequency fluctuation, the pause rate, and the stability of speech energy; it can be determined based on the following characteristics in video data: the frequency of facial expression changes, the ratio of negative to positive micro-expressions; and it can also be determined based on characteristics such as the stability of operational rhythms in motion data. When determining emotional stability, to improve accuracy, the weighting of speech and video data can be set higher than that of motion data.
[0089] The psychological stress index can be used to reflect the tension and stress level of test subjects during tasks or assessments. It can be determined based on the following characteristics in speech data: speech pause rate, voice tremor, or instability of voice energy; and based on the following characteristics in motion data or interactive event streams: reaction delay, operational error rate, and behavioral rhythm. To improve the accuracy of the psychological stress index, the changing trends of the above characteristics over time should be carefully considered.
[0090] Social engagement can be used to reflect the initiative and participation of a test subject during interaction or expression. It can be determined based on the following characteristics in voice data: frequency of active vocalization, response delay, etc.; based on the following characteristics in motion or video data: duration of gaze, intensity of facial expression feedback, etc.; and based on the following characteristics in the interaction event stream: the proportion of proactive actions or responses, etc. To improve accuracy in determining a test subject's social engagement, a comprehensive assessment using data from various modalities can be conducted.
[0091] Attentional concentration can be used to reflect the level of focus of a test subject during the execution of a test task. It can be determined based on the following characteristics in speech data: response delay, interruption of expression; based on the following characteristics in video data: gaze stability, frequency of gaze deviation, etc.; and based on the following characteristics in motion data or interactive event streams: deviation of operation trajectory, proportion of invalid operations, etc. When determining the attentional concentration of a test subject, video data, motion data, and interactive event streams should be given particular attention.
[0092] In an optional implementation, if the mental health assessment results include a mental health index, step S3 specifically includes:
[0093] Using psychological dimension vectors To calculate the Comprehensive Mental Health Index (MHI) using the following input method:
[0094] ;
[0095] in, It can be determined based on large sample statistical learning methods, and the sum of the four can be set to 1.
[0096] In an optional implementation, the mental health assessment results may also include a risk level, and different mental health indices may be classified into different risk levels according to practical needs, which may be divided into four different levels.
[0097] In an optional implementation, the mental health assessment results may also include scores for various psychological characteristics, with the specific scores configurable according to practical needs.
[0098] In an optional implementation, step S3 specifically includes: obtaining the mental health assessment results of the test subjects based on group characteristics and psychological characteristics.
[0099] When calculating the mental health index, different settings can be set for test subjects corresponding to different group characteristics. , , , When determining the risk level, thresholds can be set for test subjects with different group characteristics.
[0100] In an optional implementation, the method further includes the following steps:
[0101] S41. Obtain the first behavioral characteristics of the test object based on the test data.
[0102] S42. Obtain the second behavioral characteristics of the test object based on the test object's historical test data.
[0103] S43. Input the first behavioral feature and the second behavioral feature into the clustering model to obtain the clustering category of the test object.
[0104] S44. Obtain psychological and behavioral labels of test subjects based on clustering categories.
[0105] In an optional implementation, the first and second behavioral characteristics may include: mean and variance of reaction time, consistency of behavioral rhythm, trend of interaction frequency, current mental health index, historical mean of mental health index, trend or fluctuation of mental health index, historical statistical values of each psychological characteristic, long-term average of emotional stability, degree of fluctuation of stress index, stability of attention concentration, etc. Behavioral characteristics are used to reflect the long-term patterns of the test subject at the behavioral level. By inputting multiple behavioral characteristics into a Gaussian Mixture Model (GMM), the problem of insufficient information caused by judging based on a single feature can be avoided.
[0106] In an optional implementation, psychological behavior labels are used to reflect the psychological behavior patterns of the test subjects. These labels may include: high-pressure stable type, low-pressure irritable type, passive avoidance type, etc.
[0107] In an optional implementation, the clustering model may employ a Gaussian Mixture Model (GMM) and a clustering decision model. The Gaussian Mixture Model is as follows:
[0108] ;
[0109] Where x is the behavioral feature vector, namely the first behavioral feature and the second behavioral feature. The behavioral features can be combined according to preset weights, or normalized and uniformly used as the behavioral feature vector input to the GMM model; K is the number of clusters. The first behavioral feature and the second behavioral feature include behavioral features of different categories, such as operational rhythm and interaction activity, and K is the total number of categories. The weights of the behavioral features for the k-th category; Let be the center vector of the behavioral features of the k-th category, which corresponds to the typical performance of the behavioral features of this category; Let be the covariance matrix of the behavioral features of the k-th category, used to characterize the fluctuation range and correlation of the behavioral features of this category in each dimension.
[0110] In an optional implementation, by inputting the first behavioral feature and the second behavioral feature into the GMM model, the GMM model outputs the probability distribution of the test object under each cluster category. Since the first behavioral feature and the second behavioral feature correspond to different time windows, the behavioral features under different time windows should be input into the GMM model separately to obtain the probability distribution of the test object under each cluster category when the test is conducted in different time windows.
[0111] After obtaining the probability distribution, it is necessary to further determine the cluster category of the test objects using the clustering determination formula. The clustering determination formula is as follows:
[0112] ;
[0113] in, : The behavioral feature vector extracted within the i-th time window or evaluation period; : Parameters from the k-th class in the aforementioned GMM model; Under the k-th Gaussian distribution, the behavioral feature vector The conditional probability; The final clustering category of the test objects selected.
[0114] Since different clustering categories correspond to different psychological and behavioral labels, the psychological and behavioral labels of the test subjects can be obtained.
[0115] By combining the first and second behavioral features with a clustering model, we can refer to both the test subject's historical behavioral characteristics and the test subject's behavioral characteristics during this psychological test. This allows us to more stably characterize an individual's long-term psychological and behavioral patterns, more accurately obtain the test subject's psychological and behavioral labels, and support the dynamic updating of the test subject's psychological and behavioral labels over time.
[0116] In an optional implementation, the method further includes:
[0117] S51. Generate intervention recommendations corresponding to the test subjects based on the results of mental health assessments.
[0118] Intervention recommendations can be made based on data such as the mental health index, scores of various psychological characteristics, and risk levels from the mental health assessment results.
[0119] Taking the generation of mental health assessment results based on scores of various psychological characteristics as an example:
[0120] The psychological stress index reflects the level of tension and stress load of test subjects during the assessment process. A higher value indicates a more significant stress state. If the psychological stress index exceeds the preset threshold, it indicates that the individual is under high stress. The intervention goal is to reduce physiological arousal level and subjective tension. Intervention suggestions include breathing relaxation and meditation training. If the emotional stability is below the preset threshold, the intervention goal is to improve emotional awareness, expression, and regulation abilities. Since emotional stability reflects the fluctuation of an individual's emotional state, a lower value indicates weaker emotional regulation ability or greater emotional fluctuations. Intervention suggestions include emotional expression training. Social activity reflects an individual's initiative and participation in interaction or communication. A lower value indicates weaker willingness to participate in social activities or a lower level of interaction. If social activity is below the preset threshold, the intervention goal is to enhance social participation motivation and interactive behavior. Intervention suggestions include social interaction training.
[0121] It can also combine scores of multiple psychological characteristics to generate intervention suggestions. For example, when the psychological stress index is high and the emotional stability is low, the intervention suggestions should prioritize stress relief and emotion regulation.
[0122] In an optional implementation, the method further includes:
[0123] S52. Based on the results of mental health assessments and group characteristics, generate intervention recommendations corresponding to the test subjects.
[0124] For example, intervention recommendations can be based on the results of mental health assessments to determine their basic content, and on the presentation method and explanatory language to determine the group characteristics. Alternatively, the basic content of intervention recommendations can be determined by combining the results of mental health assessments and group characteristics. For example, for test subjects whose group characteristics are children, intervention recommendations can focus on behavioral guidance and family advice; for test subjects whose group characteristics are adults, intervention recommendations can focus on self-regulation; and for test subjects whose group characteristics are elderly, intervention recommendations can focus on follow-up and supportive interventions.
[0125] In an optional implementation, the method further includes:
[0126] S53. Based on the results of mental health assessments and psychological and behavioral labels, generate intervention recommendations corresponding to the test subjects.
[0127] For example, when social positivity is low and the psychological behavior is labeled as passive-avoidant, the recommendation intensity for social training should be increased in the intervention suggestions.
[0128] In an optional implementation, the method further includes:
[0129] S54. Based on mental health assessment results, group characteristics, and psychological and behavioral labels, generate intervention recommendations corresponding to the test subjects.
[0130] You can choose to execute any of the steps S51 to S54 depending on the practice. If you choose to execute step S53 or S54, you need to execute step S4 before executing this step.
[0131] In an optional implementation, the method further includes:
[0132] S6. Generate a psychological and behavioral profile file of the test subject.
[0133] The psychological and behavioral profile file may include: basic identification information, such as the identification of the test subject, the profile generation time, and the time window covered by the profile; psychological health assessment results; psychological and behavioral labels; and statistics on the psychological and behavioral labels of the test subject under different time windows.
[0134] In an optional implementation, the method further includes:
[0135] S71. Track the changing trends of mental health assessment results over time and construct a trend curve.
[0136] S72. Adjust intervention recommendations based on the trend curve.
[0137] In an optional implementation, the trend T(t) can be defined as the rate of change of MHI over time. If the trend T(t) < 0 (decline in mental state), a reassessment or intervention adjustment mechanism is automatically triggered, for example:
[0138]
[0139] in, : Indicates the time point of the i-th evaluation of the test object. The obtained mental health index; : These represent the start and end time points within the trend calculation window, respectively; T(t): represents the overall trend of mental health status within the stated time window. When T(t) < 0, it indicates a downward trend in mental health status; when T(t) > 0, it indicates an improving trend in mental health status. When T(t) < 0 and the decline exceeds a preset threshold, the system automatically triggers a reassessment mechanism; when T(t) < 0 and continues to decline, the system adjusts the intensity or type of intervention recommendations; when T(t) ≥ 0, the system maintains or gradually reduces the intervention frequency.
[0140] In an alternative implementation, to avoid the impact of fluctuations in a single assessment on trend judgment, a robust trend can be calculated based on multiple time points, for example:
[0141] ;
[0142] This method improves the stability of trend estimation by taking the median of the rate of change of multiple time points, and is suitable for evaluating scenarios where the time is uneven or the data contains noise.
[0143] In an optional implementation, during trend tracking, the system can combine the psychological and behavioral tags of the test subjects to determine whether a decline in psychological state is accompanied by a shift in behavioral patterns; and to apply different trend thresholds and optimization strategies to different types of behavioral profiles.
[0144] This implementation plan conducts psychological tests on test subjects based on the group characteristics of the test subjects, which can achieve differentiated testing for different groups, thereby improving the universality of the test; the psychological characteristics of the test subjects are obtained based on multi-source test data, which is more accurate and objective than assessment based on a single scale; the psychological health assessment results of the test subjects are obtained based on more precise psychological characteristics, which can further improve the objectivity and accuracy of psychological health assessment.
[0145] Example 2
[0146] Corresponding to the aforementioned embodiments of mental health assessment methods, this disclosure also provides embodiments of mental health assessment systems.
[0147] Figure 2 A schematic diagram of a mental health assessment system provided as an exemplary embodiment of this disclosure, the system comprising:
[0148] Test Module 1 is used to conduct psychological tests on the test subjects based on the group characteristics of the test subjects in order to obtain test data.
[0149] The test data includes multi-source data generated by the test subject during the psychological test. In an optional implementation, the test data includes at least one of the following: voice data, video data, motion data, interactive event streams, etc.
[0150] Group characteristics refer to the group type to which the test subjects belong. In one optional implementation, group characteristics include at least one of the following: age group, educational background, stage of disease development, etc. This approach makes the expressions of test subjects with the same group characteristics more comparable in psychological testing, thereby avoiding testing bias caused by physiological and other differences.
[0151] Psychological Feature Acquisition Module 2 is used to acquire the psychological features of the test subjects based on the test data.
[0152] In one alternative implementation, the psychological characteristics include at least one of the following: emotional stability, psychological stress index, social positivity, attention concentration, etc.
[0153] Assessment module 3 is used to obtain the mental health assessment results of the test subjects based on their psychological characteristics.
[0154] In one optional implementation, the mental health assessment results include at least one of the following: a mental health index, scores for various mental characteristics, risk levels, etc., with the risk levels corresponding to the mental health index.
[0155] In an optional implementation, test data of the test object can be acquired using devices such as microphone arrays, cameras, touch screen acquisition units, and interactive interfaces. The test data can be used as a raw signal stream for subsequent calculations or evaluations. The raw signal stream is shown below:
[0156] ;
[0157] Where A(t) is voice data, V(t) is video data, M(t) is motion data, and I(t) is the interactive event stream.
[0158] In one optional implementation, the sampling frequency of the test data can be: 16kHz for audio, 25fps for video, and ≥60Hz for touch sampling.
[0159] In one optional implementation, after obtaining the test data, it is necessary to preprocess the test data. The preprocessing process may include: noise reduction, time synchronization, feature extraction, etc.
[0160] When performing timing synchronization, a hybrid time stamp mechanism can be used to ensure that the synchronization error of test data for different modes is no more than 25ms.
[0161] During feature extraction, a multimodal feature vector set can be obtained. These include acoustic features (speech rate, tone), visual features (AU coding), and motor features (reaction time, trajectory smoothness).
[0162] In an optional implementation, the test module is also used for:
[0163] The type and presentation format of the test task are selected based on the characteristics of the group; the test object is tested based on the test task to obtain test data.
[0164] Selecting the type and presentation format of test tasks based on group characteristics can achieve "cross-group universal design". This can be achieved by introducing differentiated parameter configurations. For example, group characteristics include age groups, which may include children, teenagers, adults, the elderly, etc. By obtaining the group characteristics of the test subjects, that is, the age group to which the test subjects belong, the corresponding test task type and presentation format can be dynamically matched.
[0165] In one optional implementation, the type of test task may include task duration, interaction complexity, and workload intensity; the presentation format of the test task may include voice-modal tasks, behavioral-modal tasks, and scale-modal tasks. For test subjects whose group characteristics are children or adolescents, test tasks with higher interaction complexity, greater workload, and voice or behavioral modalities can be used; for test subjects whose group characteristics are elderly, test tasks with lower interaction complexity, less workload, and voice or behavioral modalities can be used.
[0166] In an optional implementation, the psychological feature acquisition module is also used for:
[0167] The test data is input into the psychological feature model to obtain the psychological features of the test subjects. The psychological feature model includes a multi-attention fusion layer; the psychological feature model can use the Transformer multi-head attention fusion layer.
[0168] In an optional implementation, psychological characteristics may include at least one of the following: emotional stability, psychological stress index, social positivity, and attention concentration. Psychological characteristics may be represented using a psychological dimension vector, which may be:
[0169] ;
[0170] in, For emotional stability; The psychological stress index; For social positivity; To ensure concentration; This is a set of multimodal feature vectors obtained based on test data.
[0171] Specifically, emotional stability can be used to reflect the degree of fluctuation in a test subject's emotional state during the evaluation process. It can be determined based on the following characteristics in speech data: the amplitude of speech rate variation, the degree of fundamental frequency fluctuation, the pause rate, and the stability of speech energy; it can be determined based on the following characteristics in video data: the frequency of facial expression changes, the ratio of negative to positive micro-expressions; and it can also be determined based on characteristics such as the stability of operational rhythms in motion data. When determining emotional stability, to improve accuracy, the weighting of speech and video data can be set higher than that of motion data.
[0172] The psychological stress index can be used to reflect the tension and stress level of test subjects during tasks or assessments. It can be determined based on the following characteristics in speech data: speech pause rate, voice tremor, or instability of voice energy; and based on the following characteristics in motion data or interactive event streams: reaction delay, operational error rate, and behavioral rhythm. To improve the accuracy of the psychological stress index, the changing trends of the above characteristics over time should be carefully considered.
[0173] Social engagement can be used to reflect the initiative and participation of a test subject during interaction or expression. It can be determined based on the following characteristics in voice data: frequency of active vocalization, response delay, etc.; based on the following characteristics in motion or video data: duration of gaze, intensity of facial expression feedback, etc.; and based on the following characteristics in the interaction event stream: the proportion of proactive actions or responses, etc. To improve accuracy in determining a test subject's social engagement, a comprehensive assessment using data from various modalities can be conducted.
[0174] Attentional concentration can be used to reflect the level of focus of a test subject during the execution of a test task. It can be determined based on the following characteristics in speech data: response delay, interruption of expression; based on the following characteristics in video data: gaze stability, frequency of gaze deviation, etc.; and based on the following characteristics in motion data or interactive event streams: deviation of operation trajectory, proportion of invalid operations, etc. When determining the attentional concentration of a test subject, video data, motion data, and interactive event streams should be given particular attention.
[0175] In an optional implementation, if the mental health assessment results include a mental health index, the assessment module is also used for:
[0176] Using psychological dimension vectors To calculate the Comprehensive Mental Health Index (MHI) using the following input method:
[0177] ;
[0178] in, It can be determined based on large sample statistical learning methods, and the sum of the four can be set to 1.
[0179] In an optional implementation, the mental health assessment results may also include a risk level, and different mental health indices may be classified into different risk levels according to practical needs, which may be divided into four different levels.
[0180] In an optional implementation, the mental health assessment results may also include scores for various psychological characteristics, with the specific scores configurable according to practical needs.
[0181] In an optional implementation, the assessment module is also used to: obtain mental health assessment results of the test subjects based on group characteristics and psychological characteristics.
[0182] When calculating the mental health index, different settings can be set for test subjects corresponding to different group characteristics. , , , When determining the risk level, thresholds can be set for test subjects with different group characteristics.
[0183] In an optional implementation, the system further includes:
[0184] The psychological behavior label acquisition module is used to: acquire the first behavioral characteristics of the test subject based on the test data, acquire the second behavioral characteristics of the test subject based on the test subject's historical test data; input the first and second behavioral characteristics into the clustering model to obtain the clustering category of the test subject; and acquire the psychological behavior label of the test subject based on the clustering category.
[0185] In an optional implementation, the first and second behavioral characteristics may include: mean and variance of reaction time, consistency of behavioral rhythm, trend of interaction frequency, current mental health index, historical mean of mental health index, trend or fluctuation of mental health index, historical statistical values of each psychological characteristic, long-term average of emotional stability, degree of fluctuation of stress index, stability of attention concentration, etc. Behavioral characteristics are used to reflect the long-term patterns of the test subject at the behavioral level. By inputting multiple behavioral characteristics into a Gaussian Mixture Model (GMM), the problem of insufficient information caused by judging based on a single feature can be avoided.
[0186] In an optional implementation, psychological behavior labels are used to reflect the psychological behavior patterns of the test subjects. These labels may include: high-pressure stable type, low-pressure irritable type, passive avoidance type, etc.
[0187] In an optional implementation, the clustering model may employ a Gaussian Mixture Model (GMM) and a clustering decision model. The Gaussian Mixture Model is as follows:
[0188] ;
[0189] Where x is the behavioral feature vector, namely the first behavioral feature and the second behavioral feature. The behavioral features can be combined according to preset weights, or normalized and uniformly used as the behavioral feature vector input to the GMM model; K is the number of clusters. The first behavioral feature and the second behavioral feature include behavioral features of different categories, such as operational rhythm and interaction activity, and K is the total number of categories. The weights of the behavioral features for the k-th category; Let be the center vector of the behavioral features of the k-th category, which corresponds to the typical performance of the behavioral features of this category; Let be the covariance matrix of the behavioral features of the k-th category, used to characterize the fluctuation range and correlation of the behavioral features of this category in each dimension.
[0190] In an optional implementation, by inputting the first behavioral feature and the second behavioral feature into the GMM model, the GMM model outputs the probability distribution of the test object under each cluster category. Since the first behavioral feature and the second behavioral feature correspond to different time windows, the behavioral features under different time windows should be input into the GMM model separately to obtain the probability distribution of the test object under each cluster category when the test is conducted in different time windows.
[0191] After obtaining the probability distribution, it is necessary to further determine the cluster category of the test objects using the clustering determination formula. The clustering determination formula is as follows:
[0192] ;
[0193] in, : The behavioral feature vector extracted within the i-th time window or evaluation period; : Parameters from the k-th class in the aforementioned GMM model; Under the k-th Gaussian distribution, the behavioral feature vector The conditional probability; The final clustering category of the test objects selected.
[0194] Since different clustering categories correspond to different psychological and behavioral labels, the psychological and behavioral labels of the test subjects can be obtained.
[0195] By combining the first and second behavioral features with a clustering model, we can refer to both the test subject's historical behavioral characteristics and the test subject's behavioral characteristics during this psychological test. This allows us to more stably characterize an individual's long-term psychological and behavioral patterns, more accurately obtain the test subject's psychological and behavioral labels, and support the dynamic updating of the test subject's psychological and behavioral labels over time.
[0196] In an optional implementation, the system further includes:
[0197] The suggestion generation module is used to generate intervention suggestions corresponding to the test subjects based on the results of mental health assessments.
[0198] Intervention recommendations can be made based on data such as the mental health index, scores of various psychological characteristics, and risk levels from the mental health assessment results.
[0199] Taking the generation of mental health assessment results based on scores of various psychological characteristics as an example:
[0200] The psychological stress index reflects the level of tension and stress load of test subjects during the assessment process. A higher value indicates a more significant stress state. If the psychological stress index exceeds the preset threshold, it indicates that the individual is under high stress. The intervention goal is to reduce physiological arousal level and subjective tension. Intervention suggestions include breathing relaxation and meditation training. If the emotional stability is below the preset threshold, the intervention goal is to improve emotional awareness, expression, and regulation abilities. Since emotional stability reflects the fluctuation of an individual's emotional state, a lower value indicates weaker emotional regulation ability or greater emotional fluctuations. Intervention suggestions include emotional expression training. Social activity reflects an individual's initiative and participation in interaction or communication. A lower value indicates weaker willingness to participate in social activities or a lower level of interaction. If social activity is below the preset threshold, the intervention goal is to enhance social participation motivation and interactive behavior. Intervention suggestions include social interaction training.
[0201] It can also combine scores of multiple psychological characteristics to generate intervention suggestions. For example, when the psychological stress index is high and the emotional stability is low, the intervention suggestions should prioritize stress relief and emotion regulation.
[0202] In an optional implementation, the system further includes:
[0203] The suggestion generation module is used to generate intervention suggestions corresponding to the test subjects based on mental health assessment results and group characteristics.
[0204] For example, intervention recommendations can be based on the results of mental health assessments to determine their basic content, and on the presentation method and explanatory language to determine the group characteristics. Alternatively, the basic content of intervention recommendations can be determined by combining the results of mental health assessments and group characteristics. For example, for test subjects whose group characteristics are children, intervention recommendations can focus on behavioral guidance and family advice; for test subjects whose group characteristics are adults, intervention recommendations can focus on self-regulation; and for test subjects whose group characteristics are elderly, intervention recommendations can focus on follow-up and supportive interventions.
[0205] In an optional implementation, the system further includes:
[0206] The suggestion generation module is used to generate intervention suggestions corresponding to the test subjects based on the results of mental health assessments and psychological behavior labels.
[0207] For example, when social positivity is low and the psychological behavior is labeled as passive-avoidant, the recommendation intensity for social training should be increased in the intervention suggestions.
[0208] In an optional implementation, the system further includes:
[0209] The suggestion generation module is used to generate intervention suggestions corresponding to the test subjects based on mental health assessment results, group characteristics, and psychological and behavioral labels.
[0210] In an optional implementation, the system further includes:
[0211] The profile generation module is used to generate psychological and behavioral profile files for the test subjects.
[0212] The psychological and behavioral profile file may include: basic identification information, such as the identification of the test subject, the profile generation time, and the time window covered by the profile; psychological health assessment results; psychological and behavioral labels; and statistics on the psychological and behavioral labels of the test subject under different time windows.
[0213] In an optional implementation, the system further includes:
[0214] The trend analysis module is used to track the changing trends of mental health assessment results over time and construct trend curves; intervention recommendations are then adjusted based on these trend curves.
[0215] In an optional implementation, the trend T(t) can be defined as the rate of change of MHI over time. If the trend T(t) < 0 (decline in mental state), a reassessment or intervention adjustment mechanism is automatically triggered, for example:
[0216]
[0217] in, : Indicates the time point of the i-th evaluation of the test object. The obtained mental health index; : These represent the start and end time points within the trend calculation window, respectively; T(t): represents the overall trend of mental health status within the stated time window. When T(t) < 0, it indicates a downward trend in mental health status; when T(t) > 0, it indicates an improving trend in mental health status. When T(t) < 0 and the decline exceeds a preset threshold, the system automatically triggers a reassessment mechanism; when T(t) < 0 and continues to decline, the system adjusts the intensity or type of intervention recommendations; when T(t) ≥ 0, the system maintains or gradually reduces the intervention frequency.
[0218] In an alternative implementation, to avoid the impact of fluctuations in a single assessment on trend judgment, a robust trend can be calculated based on multiple time points, for example:
[0219] ;
[0220] This method improves the stability of trend estimation by taking the median of the rate of change of multiple time points, and is suitable for evaluating scenarios where the time is uneven or the data contains noise.
[0221] In an optional implementation, during trend tracking, the system can combine the psychological and behavioral tags of the test subjects to determine whether a decline in psychological state is accompanied by a shift in behavioral patterns; and to apply different trend thresholds and optimization strategies to different types of behavioral profiles.
[0222] This implementation plan conducts psychological tests on test subjects based on the group characteristics of the test subjects, which can achieve differentiated testing for different groups, thereby improving the universality of the test; the psychological characteristics of the test subjects are obtained based on multi-source test data, which is more accurate and objective than assessment based on a single scale; the psychological health assessment results of the test subjects are obtained based on more precise psychological characteristics, which can further improve the objectivity and accuracy of psychological health assessment.
[0223] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs.
[0224] Example 3
[0225] Figure 3This is a schematic diagram of the structure of an electronic device according to an example embodiment of the present disclosure. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the mental health assessment method described in any of the above embodiments. Figure 3 The electronic device 30 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.
[0226] like Figure 3 As shown, the electronic device 30 can be manifested as a general-purpose computing device, such as a server device. The components of the electronic device 30 may include, but are not limited to: at least one processor 31, at least one memory 32, and a bus 33 connecting different system components (including memory 32 and processor 31).
[0227] Bus 33 includes a data bus, an address bus, and a control bus.
[0228] The memory 32 may include volatile memory, such as random access memory (RAM) 321 and / or cache memory 322, and may further include read-only memory (ROM) 323.
[0229] The memory 32 may also include a program tool 325 (or utility) having a set (at least one) program module 324, such program module 324 including but not limited to: an operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.
[0230] The processor 31 executes various functional applications and data processing by running computer programs stored in the memory 32, such as the mental health assessment method provided in any of the above embodiments.
[0231] Electronic device 30 can also communicate with one or more external devices 34 (e.g., keyboard, pointing device, etc.). This communication can be performed through input / output (I / O) interface 35. Furthermore, electronic device 30 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public network, such as the Internet) via network adapter 36. As shown, network adapter 36 communicates with other modules of electronic device 30 via bus 33. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with electronic device 30, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.
[0232] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.
[0233] Example 4
[0234] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the mental health assessment method provided in any of the above embodiments.
[0235] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.
[0236] Example 5
[0237] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described mental health assessment methods.
[0238] The program code for executing the computer program product of this disclosure can be written in any combination of one or more programming languages, and the program code can be executed entirely on a user device, partially on a user device, as a stand-alone software package, partially on a user device and partially on a remote device, or entirely on a remote device.
[0239] While specific embodiments of this disclosure have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of this disclosure is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of this disclosure, but all such changes and modifications fall within the scope of protection of this disclosure.
Claims
1. A method for assessing mental health, characterized in that, include: The type of test task and / or the presentation format of the test task are selected based on the group characteristics of the test subjects to conduct psychological tests on the test subjects in order to obtain test data, which includes multi-source data generated by the test subjects during the psychological test. The psychological characteristics of the test subject are obtained based on the test data and the multi-attention fusion layer; The mental health assessment results of the test subjects are obtained based on the group characteristics and psychological characteristics.
2. The mental health assessment method as described in claim 1, characterized in that, The mental health assessment results include a mental health index, and the steps for obtaining the mental health assessment results of the test subjects based on the group characteristics and the psychological characteristics include: The mental health index of the test subject is obtained based on the aforementioned psychological characteristics and the mental health index calculation formula, wherein the mental health index calculation formula is: ； MHI stands for Mental Health Index. , , , For different dimensions of psychological characteristics, The weights of the psychological characteristics corresponding to different dimensions are set based on the group characteristics.
3. The mental health assessment method as described in claim 1, characterized in that, The method further includes: Based on the test data, obtain the first behavioral characteristics of the test object; The second behavioral characteristics of the test object are obtained based on the historical test data of the test object; The first behavioral feature and the second behavioral feature are input into the clustering model to obtain the clustering category of the test object; The psychological and behavioral labels of the test subjects are obtained based on the clustering categories.
4. The mental health assessment method as described in any one of claims 1-3, characterized in that, The method further includes: Based on the results of the mental health assessment, intervention recommendations are generated for the test subjects.
5. The mental health assessment method as described in any one of claims 1-3, characterized in that, The method further includes: Based on the mental health assessment results, group characteristics, and / or psychological and behavioral labels, intervention recommendations are generated that correspond to the test subjects.
6. The mental health assessment method as described in any one of claims 1-3, characterized in that, The group characteristics include at least one of the following: age group, educational background, and stage of disease development; And / or, The test data includes at least one of the following: voice data, video data, motion data, and interactive event streams; And / or, The psychological characteristics include at least one of the following: emotional stability, psychological stress index, social positivity, and attention concentration; And / or, Psychological and behavioral labels include at least one of the following: high-pressure stable type, low-pressure irritable type, and passive avoidant type; And / or, The mental health assessment results include at least one of the following: mental health index, scores of various psychological characteristics, and risk level, wherein the risk level corresponds to the mental health index.
7. A mental health assessment system, characterized in that, include: The testing module is used to select the type of testing task and / or the presentation format of the testing task based on the group characteristics of the test subjects to conduct psychological tests on the test subjects and obtain test data, which includes multi-source data generated by the test subjects during the psychological test. The psychological feature acquisition module is used to acquire the psychological features of the test subject based on the test data and the multi-attention fusion layer. The assessment module is used to obtain the mental health assessment results of the test subjects based on the group characteristics and the psychological characteristics.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the mental health assessment method according to any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the mental health assessment method according to any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the mental health assessment method as described in any one of claims 1 to 6.