Intelligent psychological assessment system based on electronic sandbox and its operation method
The intelligent psychological assessment system using an electronic sandbox addresses subjectivity and inconsistency in mental health assessments by employing neural networks and semantic analysis to provide reliable and efficient evaluations of depression, anxiety, and obsession.
Patent Information
- Application Number
- CN202411019798.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-07-29
AI Technical Summary
Existing psychological assessment methods for mental health disorders such as depression, anxiety, and obsession suffer from subjectivity, inconsistency in interpretation, and a 'black box' state, leading to low reliability and inefficiency.
An intelligent psychological assessment system using an electronic sandbox that includes a capability module for emotion evaluation, a task module for content analysis, an evidence module for semantic processing, and a reliability and validity analysis module, employing convolutional neural networks, Transformer networks, and semantic similarity measurement to provide accurate psychological assessments.
The system enhances the reliability and efficiency of psychological assessments by providing objective, comprehensive evaluations of mental health dimensions through semantic analysis and validation techniques.
Smart Images

Figure CN118553381B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of medical care informatics, and in particular to an intelligent psychological assessment system based on an electronic sandbox and an operation method thereof. Background Art
[0002] An important research area in psychology is the effective measurement and assessment of an individual's mental health dimensions such as depression, anxiety, and compulsion. Accurate assessment results can help individuals prevent mental illness and, combined with appropriate intervention techniques, effectively improve their mental health.
[0003] However, the symptoms of mental health disorders such as depression, anxiety, and obsession are usually subjective, and individual self-reports are easily affected by subjectivity and recall bias. In order to obtain accurate psychological assessment results, it is necessary to rely on the professional knowledge of psychologists to conduct rigorous assessments. The symptoms of mental health disorders are complex and diverse, and a variety of assessment tools and methods are needed for comprehensive psychological assessments.
[0004] In related technologies, self-reports, projective tests, game environment assessments, and clinical interviews with psychologists are usually used to assess users' mental health disorders. The standard environment and expected effects of these methods may lead to test anxiety and deception, and the independence of data points limits comprehensiveness. Visual projective tests focus on personality and mental health, but there are problems with subjectivity and consistency of interpretation. Game environment assessments perform well in cognitive abilities, but are complex to develop and the subjects' behavioral decision-making data are often ignored, remaining in a "black box" state, resulting in low reliability of psychological assessments. Summary of the invention
[0005] The present invention provides an intelligent psychological assessment system based on an electronic sandbox and an operating method thereof, so as to solve the defects of the psychological health disorder observation method in the prior art, such as the contradiction between subjectivity and consistency of interpretation and the easy existence of a "black box" state, which leads to low reliability of psychological assessment, thereby improving the reliability and efficiency of psychological assessment.
[0006] The present invention provides an intelligent psychological assessment system based on an electronic sandbox, comprising:
[0007] A capability module, the capability module is used to evaluate the target emotion of the user and obtain an evaluation result; the target emotion includes healthy emotion and negative emotion, and the negative emotion includes at least one of depression, anxiety, compulsion and anger;
[0008] A task module, which is used to preliminarily identify and summarize the content created by the user through the electronic sandbox to obtain the sandbox theme summary result;
[0009] An evidence module is used to determine the sand table theme according to external psychological semantics and the basic semantic perception of the sand tools extracted from the sand table theme induction result, and to process the sand table theme and the evaluation result in turn through semantic similarity measurement and hierarchical clustering to obtain the user psychological evaluation result; wherein the evaluation result is used to represent the symptom degree of the user in multiple psychological dimensions.
[0010] According to an intelligent psychological assessment system based on an electronic sandbox provided by the present invention, the evidence module includes:
[0011] A sand table theme recognition module, which is used to extract the corresponding basic semantic perception of sand tools from the sand table theme induction result through a pre-trained convolutional neural network, integrate the introduced external psychological semantics and the basic semantic perception of sand tools through a self-attention operation and a Transformer network, and classify the sand table theme according to the integrated features to obtain a sand table theme recognition result;
[0012] A psychological stratification assessment module is used to perform word segmentation and extract topic features on the sandbox theme recognition results to obtain feature phrases, extract multiple semantic features from the feature phrases using a text pre-training model, calculate the semantic similarity between the multiple semantic features, and cluster the feature phrases according to the distance information corresponding to the semantic similarity to obtain the user psychological assessment results.
[0013] According to an intelligent psychological assessment system based on an electronic sandbox provided by the present invention, the task module includes:
[0014] A sandbox scene creation module, the sandbox scene creation module is used to obtain the position of the object of interest placed by the user in the sandbox of the electronic sandbox and the target operation on the sandbox terrain; the target operation includes at least one of digging sand, piling sand and smoothing the terrain;
[0015] The sandbox theme theory module is used to perform theme identification and induction on the position of the object of interest and the target operation to obtain the sandbox theme induction result.
[0016] According to an intelligent psychological assessment system based on an electronic sandbox provided by the present invention, the system further comprises:
[0017] The external test implementation module is used to determine the external reference value corresponding to the user's psychological evaluation result according to the SCL-90 scale, and evaluate the symptom degree of the user in different mental health dimensions according to the external reference value.
[0018] According to an intelligent psychological assessment system based on an electronic sandbox provided by the present invention, the system further comprises:
[0019] A reliability and validity analysis module, which performs reliability verification on the user's psychological evaluation results and their external reference values according to the alpha coefficient and the intra-group correlation coefficient, and performs validity verification on the user's psychological evaluation results and their external reference values according to the target validity; wherein the target validity includes at least one of structural validity, convergent validity, discriminant validity and criterion-related validity.
[0020] The present invention also provides an operation method of an intelligent psychological assessment system based on an electronic sandbox, comprising:
[0021] The target emotion of the user is evaluated through the ability module to obtain an evaluation result; the content created by the user through the electronic sand table is preliminarily identified and summarized through the task module to obtain a sand table theme summary result; the target emotion includes healthy emotions and bad emotions, and the bad emotions include at least one of depression, anxiety, compulsion and anger;
[0022] The sand tray theme is determined by the evidence module according to the external psychological semantics and the basic semantic perception of the sand tools extracted from the sand tray theme induction result, and the sand tray theme and the evaluation result are processed by semantic similarity measurement and hierarchical clustering in turn to obtain the user psychological evaluation result; wherein, the evaluation result is used to represent the symptom degree of the user in multiple psychological dimensions.
[0023] According to an operation method of an intelligent psychological evaluation system based on an electronic sandbox provided by the present invention, after obtaining the user psychological evaluation result, the method further includes:
[0024] The external test implementation module determines the external reference value corresponding to the user's psychological evaluation result according to the SCL-90 scale, and evaluates the symptom degree of the user in different mental health dimensions according to the external reference value.
[0025] According to an operation method of an intelligent psychological evaluation system based on an electronic sandbox provided by the present invention, after obtaining the user psychological evaluation result, the method further includes:
[0026] The reliability and validity analysis module performs reliability verification on the user's psychological assessment results and their external reference values according to the α coefficient and the intra-group correlation coefficient, and performs validity verification on the user's psychological assessment results and their external reference values according to the target validity; wherein the target validity includes at least one of structural validity, convergent validity, discriminant validity and criterion-related validity.
[0027] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, an operating method of an intelligent psychological assessment system based on an electronic sandbox as described above is implemented.
[0028] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the operating method of the intelligent psychological assessment system based on the electronic sandbox as described in any one of the above is implemented.
[0029] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the operating method of any of the above-mentioned intelligent psychological assessment systems based on the electronic sandbox.
[0030] The intelligent psychological assessment system based on the electronic sand table provided by the present invention and its operation method evaluate the target emotion of the user through the ability module to obtain the assessment result, preliminarily identify and summarize the content created by the user through the electronic sand table through the task module to obtain the sand table theme summary result, determine the sand table theme according to the external psychological semantics and the basic semantic perception of the sand tool extracted from the sand table theme summary result through the evidence module, and successively process the sand table theme and the assessment result through semantic similarity measurement and hierarchical clustering to obtain the user psychological assessment result, thereby improving the reliability and efficiency of the psychological assessment and providing an effective tool for the assessment of mental health problems. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0032] Figure 1 It is one of the structural schematic diagrams of the intelligent psychological evaluation system based on the electronic sandbox provided by the present invention.
[0033] Figure 2 It is a schematic diagram of the mapping relationship between the psychological dimensions, diagnostic elements and sand tray themes provided by the present invention.
[0034] Figure 3 It is a structural schematic diagram of the sandbox theme identification module provided by the present invention.
[0035] Figure 4 It is a flowchart of the working mechanism of the psychological stratification assessment module provided by the present invention.
[0036] Figure 5This is the second structural schematic diagram of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention.
[0037] Figure 6 It is a flow chart of the operation mechanism of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention.
[0038] Figure 7 It is a flow chart of the operation method of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention.
[0039] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention.
[0040] Reference numerals:
[0041] 110: capability module; 120: task module; 121: sandbox scene creation module;
[0042] 122: Sand tray theme theory module; 130: Evidence module; 131: Sand tray theme identification module;
[0043] 132: Psychological stratification assessment module; 140: External test implementation module;
[0044] 150: Reliability and validity analysis module. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.
[0046] Combine the following Figure 1-Figure 7 The invention describes an intelligent psychological assessment system based on an electronic sandbox and an operation method thereof.
[0047] Figure 1 This is one of the structural diagrams of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention. Figure 1 As shown, the intelligent psychological assessment system based on the electronic sandbox includes: an ability module 110, a task module 120 and an evidence module 130.
[0048] The capability module 110 is used to evaluate the user's target emotion and obtain an evaluation result; the target emotion includes healthy emotion and bad emotion, and the bad emotion includes at least one of depression, anxiety, compulsion and anger.
[0049] In this embodiment, healthy emotions may include joy, cheerfulness, and peace, among others.
[0050] In this embodiment, the ability module 110 can be used to score the user's health in the three psychological dimensions of depression, anxiety, and compulsion. The ability module 110 comprehensively examines the three dimensions of depression, anxiety, and compulsion to conduct a comprehensive assessment of the visitor's mental health status.
[0051] In this embodiment, the capability module 110 can identify the creative content left by the user in the electronic sandbox and determine the target emotion. For example, the sandbox image is input into the emotion recognition model set by the capability module 110 to obtain the corresponding target emotion.
[0052] In this embodiment, the capability module 110 may also directly identify the user portrait to determine the user's current target emotion, where the user portrait includes the user's facial expressions and body movements.
[0053] The task module 120 is used to preliminarily identify and summarize the content created by the user through the electronic sandbox to obtain the sandbox theme summary results.
[0054] In this embodiment, the task module 120 is used to perform the evaluation task of the electronic sandbox, and the evaluation task includes: sandbox scene creation and sandbox theme theory; for example, the sandbox scene creation module 121 allows the subject to create freely in the electronic sandbox, while the sandbox theme theory module 122 realizes the identification and induction of themes through artificial intelligence technology.
[0055] In this embodiment, a sandbox theme theoretical model may be set in the task module 120, and the theoretical model is used to preliminarily identify the content created by the user in the electronic sandbox to obtain classification information related to the sandbox theme.
[0056] The evidence module 130 is used to determine the sand tray theme based on external psychological semantics and the basic semantic perception of the sand tools extracted from the sand tray theme induction results, and to process the sand tray theme and the evaluation results in turn through semantic similarity measurement and hierarchical clustering to obtain the user's psychological evaluation results; wherein the evaluation results are used to indicate the degree of the user's symptoms in multiple psychological dimensions.
[0057] In this embodiment, the evidence module 130 is configured to construct the association between the above-mentioned ability model and the task model through an electronic sand tray, that is, from the sand tray theme score to the psychological diagnostic elements, and then to the association between the three mental health dimensions of depression, anxiety, and compulsion, and finally to further identify and summarize the sand tray theme through artificial intelligence technology to obtain a more accurate sand tray theme.
[0058] Figure 2 is a schematic diagram of the mapping relationship between the psychological dimensions, diagnostic elements and sand table themes provided by the present invention. Figure 2 In the illustrated embodiment, it is assumed that the psychological dimensions include "depression", "anxiety" and "compulsion", and the diagnostic elements corresponding to "depression" include "low mood", "slow thinking" and "lack of energy", and the theme corresponding to "low mood" is "emptiness, injury, other trauma themes", the theme corresponding to "slow thinking" is "split, flow, confusion", and the theme corresponding to "lack of energy" is "energy"; the diagnostic elements corresponding to "anxiety" include "negative expectations", "sensitivity" and "anxiety", and the theme corresponding to "negative expectations" is "threat", the theme corresponding to "sensitivity" is "restriction, hindrance and energy", and the theme corresponding to "anxiety" is "injury, flow"; the diagnostic elements corresponding to "compulsion" include "compulsive behavior", "compulsive thinking" and "countering compulsion", and the theme corresponding to "compulsive behavior" is "stereotypes, other trauma themes", the theme corresponding to "compulsive thinking" is "spirituality", and the theme corresponding to "countering compulsion" is "split threat".
[0059] In this embodiment, the evidence module 130 is designed through the hierarchical analysis method of the psychoanalyst, and the sand tray theme hierarchical identification (HIST) model is constructed with the help of artificial intelligence technology. The sand tray theme is quickly and accurately identified through the steps of basic semantic perception of sand tools, introduction of external psychological semantics and final theme classification task; and then the scores of the three dimensions of depression, anxiety and compulsion are obtained through the sand tray theme. For example, the sand tray theme and target emotion are automatically integrated and summarized as diagnostic elements through artificial intelligence technology, and presented in the form of a database to facilitate the analysis of other psychological dimensions, comprehensively display related symptoms, and provide convenience for in-depth research.
[0060] The intelligent psychological assessment system based on the electronic sand table provided by the embodiment of the present invention evaluates the target emotion of the user through the ability module 110 to obtain the assessment result, preliminarily identifies and summarizes the content created by the user through the electronic sand table through the task module 120 to obtain the sand table theme summary result, determines the sand table theme according to the external psychological semantics and the basic semantic perception of the sand tool extracted from the sand table theme summary result through the evidence module 130, and obtains the user psychological assessment result by processing the sand table theme and the assessment result through semantic similarity measurement and hierarchical clustering in turn, thereby improving the reliability and efficiency of psychological assessment and providing an effective tool for the assessment of mental health problems.
[0061] In some embodiments, the evidence module 130 includes: a sand table theme identification module 131, which is used to extract the corresponding basic semantic perception of sand tools from the sand table theme induction results through a pre-trained convolutional neural network, and integrate the introduced external psychological semantics and basic semantic perception of sand tools through self-attention operation and Transformer network, and classify the sand table theme according to the integrated features to obtain the sand table theme identification results; a psychological stratification evaluation module 132, which is used to perform word segmentation and extract theme features on the sand table theme identification results to obtain feature phrases, use a text pre-training model to extract multiple semantic features from the feature phrases, calculate the semantic similarity between multiple semantic features, and cluster the feature phrases according to the distance information corresponding to the semantic similarity to obtain the user psychological evaluation results.
[0062] In this embodiment, the sandbox theme identification module 131 may include: a basic semantic perception module, an external knowledge introduction module, and a theme comprehensive classification module.
[0063] Figure 3 is a schematic diagram of the structure of the sandbox theme identification module 131 provided by the present invention. Figure 3 In the illustrated embodiment, the basic semantic perception module is used to extract basic semantic information (target category information includes category 1, category 2...), especially the category information of sand tool targets, from the sand table image, and use a general residual backbone network as an image encoder. For a given sand table image, it is input into the image encoder to obtain a corresponding feature vector; wherein the sand table image can be extracted from the sand table theme induction result, or it can be directly photographed.
[0064] Specifically, the basic semantic perception module extracts basic semantic information, especially the category information of the sand tool target, from the sand table image I through the following formula:
[0065] ;
[0066] in, It is the basic semantic information vector extracted by the pre-trained convolutional neural network ResNet.
[0067] In this embodiment, the external knowledge introduction module is used to introduce external psychological attribute knowledge into the model, that is, the psychological attributes of each sand tool target, and then integrate the external knowledge into the model by encoding the psychological attribute information into a high-level semantic vector, using the object category index, and weighting it in combination with the existence probability of the sand tool object; the self-attention operation and the Transformer network are used to integrate the external knowledge feature information; the specific calculation process is expressed as follows:
[0068] ;
[0069] ;
[0070] ;
[0071] in, Represents the psychological attribute knowledge of the sand tool target, Indicates that Transformer The encoded high-level semantic vector, It is a high-level semantic vector after weighted integration by the self-attention mechanism. It is a comprehensive feature vector that combines the basic semantic vector and the weighted feature vector.
[0072] In this embodiment, the theme comprehensive classification module is used to integrate basic semantic information and external knowledge information (external knowledge includes existing related vocabulary) to finally classify the sandbox theme.
[0073] Specifically, the topic comprehensive classification module can be set as a topic comprehensive classification model, which is trained through the following steps:
[0074] (1) Expand the size of the sample data set through image enhancement technology. The specific calculation process is as follows:
[0075] ;
[0076] (2) Use the pre-trained model to extract the initial image features, which can be expressed as follows:
[0077] ;
[0078] in, For the j Samples The corresponding semantic feature vector; by Input to the fully connected layer to obtain a multi-label classification vector, which is expressed as follows:
[0079] ;
[0080] in, for The corresponding label classification vector.
[0081] (3) Finally, the sandbox theme is identified based on the multi-label classification vector, and the classification error is constructed using the cross entropy loss function. The basic semantic classification error and the theme classification error are integrated through the relative weight coefficient, which is used for back propagation to train the model. When the model converges, the corresponding theme comprehensive classification model is obtained. The cross entropy loss function is expressed by the following formula:
[0082] ;
[0083] in, For the i The labels corresponding to the sample data are used; the most useful one is to use the above-trained comprehensive theme classification model to realize the accurate classification of the sand table theme according to the sand table theme induction results.
[0084] In this embodiment, the psychological stratification assessment module 132 obtains the user's psychological assessment results through the following steps:
[0085] (1) Construct a set of concept-related descriptions. Specifically, by investigating psychology literature and collecting descriptive language about the sand tray theme, including experimental results, observations, and analytical conclusions, a set of descriptions was formed, which provided a basis for subsequent inductive analysis.
[0086] (2) Construct a set of concept-related phrases; specifically, after obtaining the descriptive language set ( ,in, Indicates After extracting the description texts), we use NLTK word segmenter to segment the relevant sentences, extract the characteristic phrases about the sandbox theme, and construct a set of concept-related phrases ( ;in Indicates from text The extracted feature texts); the above phrase set can more accurately represent the sandbox theme and provide material for subsequent semantic similarity measurement.
[0087] (3) Semantic similarity measurement; based on the obtained phrase set ( ), using the BERT text pre-training model in natural language processing technology to extract features and obtain vector representation , the process can be expressed as: ; Then use cosine similarity to calculate the semantic similarity between each feature:
[0088] ;
[0089] in, Represents two features and The cosine similarity between them is used to quantify the relationship between different topic features through semantic similarity measurement, providing quantitative support for establishing conceptual relationships.
[0090] (4) Hierarchical clustering processing: by measuring semantic similarity, the distance relationship between feature phrases is obtained :
[0091] ;
[0092] Specifically, based on the obtained distance information, a hierarchical clustering method is used to cluster the characteristic phrases. The hierarchical clustering steps are as follows:
[0093] (1) Initialize clustering using the following formula:
[0094] ;
[0095] in, C is the set of cluster centers.
[0096] (2) Calculate the inter-cluster distance using the single chain method:
[0097] ;
[0098] in, For the k Cluster centers, For the Cluster centers.
[0099] (3) Merge the nearest clusters by the following formula:
[0100] ;
[0101] ;
[0102] (4) Repeat the merging process until the termination condition is reached; each cluster corresponds to a target emotion score related to the sandbox theme and contains semantically similar feature phrases.
[0103] In this embodiment, the clustering results are subjected to a hierarchical analysis, and the obtained clusters are regarded as summarized characteristic elements. The characteristic phrases within each cluster are semantically similar, reflecting a class of similar symptom manifestations, while different clusters represent different semantic differences. These clusters can be regarded as diagnostic elements, and their number reflects the diversity and quantity of diagnostic elements.
[0104] Figure 4 is a flowchart of the working mechanism of the psychological stratification assessment module 132 provided by the present invention. Figure 4 In the illustrated embodiment, a set of descriptions related to concept A is collected and organized, and the description set is segmented to obtain a set of phrases related to concept A, the semantic similarity between the characteristic phrases in the phrase set is calculated, a semantic similarity heat map is generated, and the phrase set is hierarchically clustered according to the semantic similarity heat map to obtain a corresponding hierarchical clustering result graph, and the characteristic elements of concept A are obtained from the graph; wherein the hierarchical relationship between different concepts is obtained by calculating the semantic similarity based on sample concept data in a preliminary experiment.
[0105] The intelligent psychological assessment system based on the electronic sand table provided by the embodiment of the present invention uses a pre-trained convolutional neural network to extract the corresponding basic semantic perception of the sand tool from the sand table theme induction result through the sand table theme recognition module 131, and integrates the introduced external psychological semantics and the basic semantic perception of the sand tool through the self-attention operation and the Transformer network to realize the sand table theme classification; the sand table theme recognition result is segmented and the theme feature phrases are extracted through the psychological layered evaluation module 132, and multiple semantic features are extracted from the feature phrases, and then the feature phrases are clustered according to the distance information corresponding to the semantic similarity to obtain the user psychological assessment result, thereby improving the accuracy of the sand table theme classification and the accuracy of the user psychological assessment.
[0106] In some embodiments, the task module 120 includes: a sandbox scene creation module 121, which is used to obtain the position of objects of interest placed by the user in the sandbox of the electronic sandbox and the target operation on the sandbox terrain; the target operation includes at least one of digging sand, piling sand and smoothing the terrain; a sandbox theme theory module 122, which is used to perform theme identification and induction on the positions of objects of interest and target operations to obtain sandbox theme summarization results.
[0107] In this embodiment, the sandbox scene creation module 121 realizes the psychological projection of the subject through an electronic sandbox based on sandbox theory. The module enables the subject to freely choose various toys of interest and place them in the sandbox through system guidance, while operating the terrain of the sandbox, including digging sand, piling sand and smoothing the terrain, to express the individual's psychological state.
[0108] In this embodiment, the sand table theme theory module 122 identifies the creative content of the user in the electronic sand table by setting a theme theory model to obtain the sand table theme induction result, and the theme theory model is obtained through pre-experimental training.
[0109] The intelligent psychological assessment system based on the electronic sandbox provided by the embodiment of the present invention obtains the position of the object of interest placed by the user in the sandbox of the electronic sandbox and the target operation on the sandbox terrain through the sandbox scene creation module 121; the position of the object of interest and the target operation are subject-identified and summarized through the sandbox theme theory module 122 to obtain the sandbox theme summary, thereby providing reliable data support for subsequent sandbox theme classification and psychological stratification evaluation.
[0110] In some embodiments, the intelligent psychological assessment system based on the electronic sandbox also includes: an external test implementation module 140, which is used to determine the external reference value corresponding to the user's psychological assessment result according to the SCL-90 scale, and evaluate the user's symptom degree in different mental health dimensions based on the external reference value.
[0111] In this embodiment, the SCL-90 scale is used to determine the true value (external reference value) of the user in psychological dimensions such as depression, anxiety, and compulsion, which can provide authoritative mental health assessment data for evaluating the degree of symptoms of individuals in different mental health dimensions, and thus provide an external reference for the evaluation results output by the system.
[0112] In some embodiments, the intelligent psychological assessment system based on the electronic sandbox also includes: a reliability and validity analysis module 150, which performs reliability verification on the user's psychological assessment results and their external reference values according to the α coefficient and the intra-group correlation coefficient, and performs validity verification on the user's psychological assessment results and their external reference values according to the target validity; wherein the target validity includes at least one of structural validity, convergent validity, discriminant validity and criterion-related validity.
[0113] In this embodiment, the α coefficient and the intra-group correlation coefficient are obtained through multiple simulation experiments. The value range of the α coefficient is from 0 to 1. The higher the value, the better the internal consistency. The α coefficient is greater than 0.7.
[0114] In this embodiment, construct validity, convergent validity, discriminant validity and criterion-related validity can also be obtained through simulation experiments or reference tables.
[0115] In this embodiment, the output results of the intelligent psychological assessment system based on the electronic sandbox are evaluated through reliability testing and validity testing to improve the reliability and validity of the output results.
[0116] The electronic sandbox-based intelligent psychological assessment system provided in the embodiment of the present invention performs reliability verification and validity verification on the system through the reliability and validity analysis module 150, thereby realizing early warning of mental health disorders.
[0117] Figure 5 This is the second structural diagram of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention. Figure 5 In the illustrated embodiment, the intelligent psychological assessment system based on the electronic sandbox includes an ability module 110, a task module 120, an evidence module 130, an external test implementation module 140 and a reliability and validity analysis module 150; wherein the task module 120 includes a sandbox scene creation module 121 and a sandbox theme theory module 122, and the evidence module 130 includes a sandbox theme identification module 131 and a psychological stratification assessment module 132.
[0118] Figure 6 It is a flow chart of the operation mechanism of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention. Figure 6In the illustrated embodiment, the operation mechanism of the intelligent psychological assessment system based on the electronic sand table includes the establishment of assessment tools, the testing process and the reliability and validity analysis; wherein, in the stage of establishing the assessment tool, three core modules are set: the ability model (to assess different dimensions of the user's mental health, such as depression, anxiety and compulsion, etc.), the evidence model (including the sand table theme identification module 131 and the psychological stratification assessment module 132) and the task model (the sand table scene creation module 121 and the sand table theme theory module 122); in the stage of the testing process, the tool for creating the sand table scene is the electronic psychological sand table (i.e., the electronic sand table), and the goal is to obtain the subject's The tool for sand tray theme identification is the sand tray theme identification model, and the corresponding goal is to summarize the sand tray theme according to the sand tray scene; the tool for psychological indicator evaluation is the psychological hierarchical evaluation model, and the corresponding goal is to obtain the scores of depression, anxiety, and obsessive-compulsive dimensions according to the sand tray theme; the tool for external test implementation module 140 is the standardized scale evaluation, and the corresponding goal is to determine the true value of the psychological evaluation results through the existing mental health evaluation tools; in the reliability and validity analysis stage, the reliability is verified by the alpha coefficient and the intra-group correlation coefficient, and the validity is verified by the structural validity, convergent validity, discriminant validity and criterion-related validity.
[0119] The following describes the operating method of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention. The operating method of the intelligent psychological assessment system based on the electronic sandbox described below and the intelligent psychological assessment system based on the electronic sandbox described above can correspond to each other.
[0120] Figure 7 It is a flow chart of the operation method of the intelligent psychological assessment system based on the electronic sandbox provided by the present invention. Figure 7 As shown, the operation method of the intelligent psychological assessment system based on the electronic sandbox includes the following steps:
[0121] Step 710: Evaluate the target emotion of the user through the capability module to obtain the evaluation result; preliminarily identify and summarize the content created by the user through the electronic sand table through the task module to obtain the sand table theme summary result; the target emotions include healthy emotions and negative emotions, and the negative emotions include at least one of depression, anxiety, compulsion and anger.
[0122] In this step, healthy emotions can include joy, cheerfulness, and peace, etc.
[0123] In this embodiment, the ability module can be configured as the visitor's health score in the three psychological dimensions of depression, anxiety, and compulsion. The ability module comprehensively examines the three dimensions of depression, anxiety, and compulsion to conduct a comprehensive assessment of the visitor's mental health status.
[0124] In this embodiment, the capability module can recognize the creative content left by the user on the electronic sandbox and determine the target emotion. For example, the sandbox image is input into the emotion recognition model set by the capability module to obtain the corresponding target emotion.
[0125] In this embodiment, the capability module may also directly identify the user portrait to determine the user's current target emotion, where the user portrait includes the user's facial expressions and body movements.
[0126] In this embodiment, the task module is used to perform the evaluation task of the electronic sandbox, and the evaluation task includes: sandbox scene creation and sandbox theme theory; for example, the sandbox scene creation module allows the subject to create freely in the electronic sandbox, while the sandbox theme theory module realizes the recognition and induction of themes through artificial intelligence technology.
[0127] In this embodiment, a sand table theme theoretical model may be set in the task module, and the theoretical model is used to preliminarily identify the content created by the user in the electronic sand table to obtain classification information related to the sand table theme.
[0128] In this embodiment, the task module includes: a sandbox scene creation module, which is used to obtain the position of the object of interest placed by the user in the sandbox of the electronic sandbox and the target operation on the sandbox terrain; the target operation includes at least one of digging sand, piling sand and smoothing the terrain; a sandbox theme theory module, which is used to perform theme identification and induction on the position of the object of interest and the target operation to obtain the sandbox theme induction result.
[0129] Step 720: Determine the sand table theme through the evidence module according to the external psychological semantics and the basic semantic perception of the sand tools extracted from the sand table theme induction results, and process the sand table theme and the evaluation results in turn through semantic similarity measurement and hierarchical clustering to obtain the user's psychological evaluation results; wherein the evaluation results are used to indicate the degree of the user's symptoms in multiple psychological dimensions.
[0130] In this step, the evidence module is configured to construct the association between the above-mentioned ability model and task model through the electronic sand tray, that is, from the sand tray theme score to the psychological diagnostic elements, and then to the association between the three mental health dimensions of depression, anxiety, and compulsion. Finally, artificial intelligence technology is used to further identify and summarize the sand tray theme to obtain a more accurate sand tray theme.
[0131] In this embodiment, an evidence module is designed through the hierarchical analysis method of a psychoanalyst, and a hierarchical identification of sand tray themes (HIST) model is constructed with the help of artificial intelligence technology. The sand tray theme is quickly and accurately identified through steps such as basic semantic perception of sand tools, introduction of external psychological semantics, and the final theme classification task. Then, the scores of the three dimensions of depression, anxiety, and compulsion are obtained through the sand tray theme. For example, the sand tray theme and target emotion are automatically integrated and summarized as diagnostic elements through artificial intelligence technology, and presented in the form of a database to facilitate the analysis of other psychological dimensions, comprehensively display related symptoms, and provide convenience for in-depth research.
[0132] In this embodiment, the evidence module includes: a sand table theme recognition module, which is used to extract the corresponding basic semantic perception of sand tools from the sand table theme induction results through a pre-trained convolutional neural network, and integrate the introduced external psychological semantics and basic semantic perception of sand tools through self-attention operation and Transformer network, and classify the sand table theme according to the integrated features to obtain the sand table theme recognition results; a psychological stratification evaluation module, which is used to perform word segmentation and extract theme features on the sand table theme recognition results to obtain feature phrases, use a text pre-training model to extract multiple semantic features from the feature phrases, calculate the semantic similarity between multiple semantic features, and cluster the feature phrases according to the distance information corresponding to the semantic similarity to obtain the user psychological evaluation results.
[0133] The operating method of the intelligent psychological assessment system based on the electronic sand table provided by the embodiment of the present invention evaluates the target emotion of the user through the ability module to obtain the assessment result, preliminarily identifies and summarizes the content created by the user through the electronic sand table through the task module to obtain the sand table theme summary result, determines the sand table theme according to the external psychological semantics and the basic semantic perception of the sand tool extracted from the sand table theme summary result through the evidence module, and processes the sand table theme and the assessment result through semantic similarity measurement and hierarchical clustering in turn to obtain the user psychological assessment result, thereby improving the reliability and efficiency of the psychological assessment and providing an effective tool for the assessment of mental health problems.
[0134] In some embodiments, after obtaining the user's psychological evaluation results, the method further includes: determining the external reference value corresponding to the user's psychological evaluation results according to the SCL-90 scale through an external test implementation module, and evaluating the user's symptom degree in different mental health dimensions based on the external reference value.
[0135] In this embodiment, the SCL-90 scale is used to determine the true value (external reference value) of the user in psychological dimensions such as depression, anxiety, and compulsion, which can provide authoritative mental health assessment data for evaluating the degree of symptoms of individuals in different mental health dimensions, and thus provide an external reference for the evaluation results output by the system.
[0136] In some embodiments, after obtaining the user psychological evaluation results, the method also includes: performing reliability verification on the user psychological evaluation results and their external reference values according to the α coefficient and the intra-group correlation coefficient through a reliability and validity analysis module, and performing validity verification on the user psychological evaluation results and their external reference values according to the target validity; wherein the target validity includes at least one of structural validity, convergent validity, discriminant validity and criterion-related validity.
[0137] In this embodiment, the α coefficient and the intra-group correlation coefficient are obtained through multiple simulation experiments. The value range of the α coefficient is from 0 to 1. The higher the value, the better the internal consistency. The α coefficient is greater than 0.7.
[0138] In this embodiment, construct validity, convergent validity, discriminant validity and criterion-related validity can also be obtained through simulation experiments or reference tables.
[0139] In this embodiment, the output results of the intelligent psychological assessment system based on the electronic sandbox are evaluated through reliability testing and validity testing to improve the reliability and validity of the output results.
[0140] The operating method of the intelligent psychological assessment system based on the electronic sandbox provided in the embodiment of the present invention performs reliability verification and validity verification on the system through a reliability and validity analysis module, thereby realizing early warning of mental health disorders.
[0141] Figure 8 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 complete mutual communication through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the operation method of the intelligent psychological evaluation system based on the electronic sand table, the method comprising: evaluating the target emotion of the user through the ability module to obtain the evaluation result; preliminarily identifying and summarizing the content created by the user through the electronic sand table through the task module to obtain the sand table theme summary result; the target emotion includes healthy emotion and bad emotion, and the bad emotion includes at least one of depression, anxiety, compulsion and anger; determining the sand table theme according to the external psychological semantics and the basic semantic perception of the sand tool extracted from the sand table theme summary result through the evidence module, and obtaining the user psychological evaluation result by sequentially performing the semantic similarity measurement and hierarchical clustering on the sand table theme and the evaluation result; wherein the evaluation result is used to represent the symptom degree of the user in multiple psychological dimensions.
[0142] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0143] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the operation method of the intelligent psychological assessment system based on the electronic sandbox provided by the above-mentioned methods, and the method includes: evaluating the user's target emotion through the ability module to obtain an evaluation result; preliminarily identifying and summarizing the content created by the user through the electronic sandbox through the task module to obtain a sandbox theme summary result; the target emotions include healthy emotions and bad emotions, and the bad emotions include at least one of depression, anxiety, compulsion and anger; determining the sandbox theme according to the external psychological semantics and the basic semantic perception of the sand tool extracted from the sandbox theme summary result through the evidence module, and obtaining the user's psychological assessment result by processing the sandbox theme and the evaluation result through semantic similarity measurement and hierarchical clustering in turn; wherein the assessment result is used to represent the degree of the user's symptoms in multiple psychological dimensions.
[0144] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the operating method of the intelligent psychological assessment system based on the electronic sandbox provided by the above-mentioned methods, the method comprising: evaluating the user's target emotion through the ability module to obtain an evaluation result; preliminarily identifying and summarizing the content created by the user through the electronic sandbox through the task module to obtain a sandbox theme summary result; the target emotions include healthy emotions and bad emotions, and the bad emotions include at least one of depression, anxiety, compulsion and anger; determining the sandbox theme through the evidence module according to external psychological semantics and the basic semantic perception of the sand tool extracted from the sandbox theme summary result, and obtaining the user's psychological assessment result by processing the sandbox theme and the evaluation result through semantic similarity measurement and hierarchical clustering in turn; wherein the assessment result is used to represent the degree of the user's symptoms in multiple psychological dimensions.
[0145] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0146] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. An intelligent psychological assessment system based on an electronic sandbox, characterized in that: include: A capability module, the capability module is used to evaluate the target emotion of the user and obtain an evaluation result; the target emotion includes healthy emotion and negative emotion, and the negative emotion includes at least one of depression, anxiety, compulsion and anger; A task module, which is used to preliminarily identify and summarize the content created by the user through the electronic sandbox to obtain the sandbox theme summary result; An evidence module, the evidence module is used to determine the sand table theme according to external psychological semantics and the basic semantic perception of the sand tool extracted from the sand table theme induction result, and to process the sand table theme and the evaluation result in turn through semantic similarity measurement and hierarchical clustering to obtain the user psychological evaluation result; wherein the evaluation result is used to represent the symptom degree of the user in multiple psychological dimensions; the evidence module includes: A sand table theme recognition module, which is used to extract the corresponding basic semantic perception of sand tools from the sand table theme induction result through a pre-trained convolutional neural network, integrate the introduced external psychological semantics and the basic semantic perception of sand tools through a self-attention operation and a Transformer network, and classify the sand table theme according to the integrated features to obtain a sand table theme recognition result; The sandbox theme recognition module includes a basic semantic perception module, an external knowledge introduction module and a theme comprehensive classification module; The basic semantic perception module is used to extract basic semantic information from the sandbox image, and use the residual backbone network to encode the given sandbox image to obtain the corresponding feature vector; The external knowledge introduction module is used to encode the psychological attribute knowledge of each sand tool target into a high-level semantic vector, integrate the high-level semantic vector into the model using the object category index, and weight it in combination with the existence probability of the sand tool object to obtain a comprehensive feature vector; The specific calculation process is expressed as follows: ; ; ; in, Represents the psychological attribute knowledge of the sand tool target, Indicates that Transformer The encoded high-level semantic vector, It is a high-level semantic vector after weighted integration by the self-attention mechanism. It is a comprehensive feature vector that combines the basic semantic vector and the weighted feature vector; It is the feature vector corresponding to the basic semantic information extracted by the pre-trained convolutional neural network ResNet; The comprehensive theme classification module is used to integrate basic semantic information and external knowledge information, and finally classify the sandbox themes.
2. The intelligent psychological assessment system based on electronic sandbox according to claim 1 is characterized in that: The evidence module also includes: A psychological stratification assessment module is used to perform word segmentation and extract topic features on the sandbox theme recognition results to obtain feature phrases, extract multiple semantic features from the feature phrases using a text pre-training model, calculate the semantic similarity between the multiple semantic features, and cluster the feature phrases according to the distance information corresponding to the semantic similarity to obtain the user psychological assessment results.
3. The intelligent psychological assessment system based on electronic sandbox according to claim 1 is characterized in that: The task modules include: A sandbox scene creation module, the sandbox scene creation module is used to obtain the position of the object of interest placed by the user in the sandbox of the electronic sandbox and the target operation on the sandbox terrain; the target operation includes at least one of digging sand, piling sand and smoothing the terrain; The sandbox theme theory module is used to perform theme identification and induction on the position of the object of interest and the target operation to obtain the sandbox theme induction result.
4. The intelligent psychological assessment system based on electronic sandbox according to claim 1 is characterized in that: The system further comprises: The external test implementation module is used to determine the external reference value corresponding to the user's psychological evaluation result according to the SCL-90 scale, and evaluate the symptom degree of the user in different mental health dimensions according to the external reference value.
5. The intelligent psychological assessment system based on the electronic sandbox according to any one of claims 1 to 4, characterized in that: The system further comprises: A reliability and validity analysis module, which performs reliability verification on the user's psychological evaluation results and their external reference values according to the alpha coefficient and the intra-group correlation coefficient, and performs validity verification on the user's psychological evaluation results and their external reference values according to the target validity; wherein the target validity includes at least one of structural validity, convergent validity, discriminant validity and criterion-related validity.
6. A method for operating an intelligent psychological assessment system based on an electronic sandbox corresponding to the intelligent psychological assessment system based on an electronic sandbox as claimed in claim 1, characterized in that: include: The user's target emotions are evaluated through the capability module to obtain the evaluation results; The content created by the user through the electronic sand table is preliminarily identified and summarized by the task module to obtain the sand table theme summary result; the target emotions include healthy emotions and bad emotions, and the bad emotions include at least one of depression, anxiety, compulsion and anger; The sand tray theme is determined by the evidence module according to the external psychological semantics and the basic semantic perception of the sand tools extracted from the sand tray theme induction result, and the sand tray theme and the evaluation result are processed by semantic similarity measurement and hierarchical clustering in turn to obtain the user psychological evaluation result; wherein, the evaluation result is used to represent the symptom degree of the user in multiple psychological dimensions.
7. The method for operating the intelligent psychological assessment system based on the electronic sandbox according to claim 6, characterized in that: After obtaining the user psychological evaluation result, the method further includes: The external test implementation module determines the external reference value corresponding to the user's psychological evaluation result according to the SCL-90 scale, and evaluates the symptom degree of the user in different mental health dimensions according to the external reference value.
8. The method for operating the intelligent psychological assessment system based on the electronic sandbox according to any one of claims 6 to 7, characterized in that: After obtaining the user psychological evaluation result, the method further includes: The reliability and validity analysis module performs reliability verification on the user's psychological assessment results and their external reference values according to the α coefficient and the intra-group correlation coefficient, and performs validity verification on the user's psychological assessment results and their external reference values according to the target validity; wherein the target validity includes at least one of structural validity, convergent validity, discriminant validity and criterion-related validity.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the operating method of the intelligent psychological assessment system based on the electronic sandbox as described in any one of claims 6 to 8 is implemented.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the operating method of the intelligent psychological assessment system based on the electronic sandbox as claimed in any one of claims 6 to 8 is implemented.
Citation Information
Patent Citations
Virtual sand table creation operation method and application thereof
CN110910490A
Psychological assessment device, method and system based on electronic psychological sand tables
CN111524578A
Psychological sand table analysis method and system
CN111724881A
Multi-modal mental health analysis method based on sand table image, text and video
CN114171198A
Sand table game quantitative analysis system, method and equipment
CN114203283A
