Mental health assessment method and device, computer equipment and storage medium

Abstract

The invention relates to the field of intelligent medical treatment, and provides a mental health assessment method, device and equipment and a computer storage medium, and the method comprises the steps: outputting an assessment question based on a decision tree, and obtaining video information and voice information of a user for answering the assessment question; performing feature extraction on the video information to obtain expression information, and performing feature extraction on the voice information to obtain mood information and text information; determining a to-be-evaluated vector corresponding to the user according to the expression information, the tone information and the text information; and determining a mental health assessment result of the user according to the to-be-assessed vector. The psychological health of the user is assessed by acquiring the answer of the user to the assessment question, so that the cost required by psychological health inquiry is reduced, and the intelligent level of psychological health assessment is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical care, and in particular to a method, apparatus, computer device and storage medium for evaluating mental health. Background Art

[0002] As people pay more and more attention to mental health, the need to identify and deal with some common psychological problems has also increased. Scientific mental health assessment not only helps to detect and diagnose symptoms early, but also provides individuals with accurate treatment plans. However, the current mental health assessment methods have certain limitations. Traditional mental health assessments mainly rely on face-to-face consultations with psychologists, which undoubtedly increases the demand for professional human resources. How to improve the intelligence level of mental health assessment has become an urgent problem to be solved. Summary of the invention

[0003] The main purpose of this application is to provide a mental health assessment method, device, equipment and computer storage medium, aiming to improve the intelligence level of mental health assessment.

[0004] In a first aspect, the present application provides a method for evaluating mental health, the method comprising the following steps:

[0005] Outputting evaluation questions based on the decision tree, and obtaining video information and voice information of the user's answers to the evaluation questions;

[0006] Extracting features from the video information to obtain expression information, and extracting features from the voice information to obtain tone information and text information;

[0007] Determine a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information;

[0008] Determine a psychological health assessment result of the user according to the vector to be assessed.

[0009] In a second aspect, the present application further provides a mental health assessment device, the mental health assessment device comprising:

[0010] A feedback acquisition module, used to output evaluation questions based on the decision tree, and acquire video information and voice information of the user's answer to the evaluation questions;

[0011] A feature extraction module is used to extract features from the video information to obtain expression information, and to extract features from the voice information to obtain tone information and text information;

[0012] An information vectorization module, used to determine a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information;

[0013] An information evaluation module is used to determine the user's psychological health evaluation result according to the vector to be evaluated.

[0014] In a third aspect, the present application also provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the mental health assessment method as described above is implemented.

[0015] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the mental health assessment method as described above is implemented.

[0016] The present application provides a method, device, equipment and computer storage medium for evaluating mental health. The present application outputs evaluation questions based on a decision tree, and obtains video information and voice information of the user's answers to the evaluation questions; extracts features from the video information to obtain expression information, extracts features from the voice information to obtain tone information and text information; determines the vector to be evaluated corresponding to the user based on the expression information, the tone information and the text information; and determines the mental health evaluation result of the user based on the vector to be evaluated. Since the user's mental health is evaluated by obtaining the user's answers to the evaluation questions, the cost required for mental health consultation is reduced, and the intelligence level of mental health evaluation is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a method for evaluating mental health provided in one embodiment of the present application;

[0019] Figure 2 A usage scenario diagram of a method for evaluating mental health provided in one embodiment of the present application;

[0020] Figure 3 A schematic block diagram of a mental health assessment device provided in one embodiment of the present application;

[0021] Figure 4 The present invention is a block diagram showing the structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0024] The embodiments of the present application provide a method, apparatus, computer device and computer-readable storage medium for evaluating mental health.

[0025] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0026] Please refer to Figure 1 , Figure 1 A flowchart of a method for evaluating mental health provided in an embodiment of the present application. The method for evaluating mental health can be used in a terminal or a server to output evaluation questions for evaluating the user's mental health and obtain video information and voice information answered by the user, automatically output the user's mental health evaluation results, and improve the intelligence level of mental health evaluation. Among them, the terminal can be an electronic device such as a mobile phone, a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, and a wearable device; the server can be an independent server, a server cluster, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

[0027] Please refer to Figure 2 , Figure 2 This is a usage scenario diagram provided by an embodiment of the present application. Figure 2 As shown, the embodiments of the present application can be applied to a terminal device, which outputs assessment questions in the form of voice or text, obtains video information and voice information of the user's answers to the assessment questions, and outputs the user's mental health assessment results based on the video information and voice information.

[0028] like Figure 1As shown, the mental health assessment method includes steps S101 to S104.

[0029] Step S101: output evaluation questions based on the decision tree, and obtain video information and voice information of the user's answers to the evaluation questions.

[0030] Exemplarily, the decision tree is suitable as an auxiliary tool in the process of disease diagnosis, helping doctors to gradually narrow down the possible range of diseases and finally make a diagnosis based on the patient's symptoms, signs, laboratory test results and other information. Therefore, in the mental health assessment method provided in the embodiment of the present application, the assessment questions are output based on the decision tree so that the mental health assessment can be performed based on the user's answers to the assessment questions.

[0031] For example, since mental illness is usually more complex and diverse than physical illness, diagnosis often needs to consider multiple aspects such as psychological state, behavior patterns, life events, personal history, etc., and the assessment of mental health also needs to collect comprehensive information from multiple aspects of the user. Therefore, the process of the user answering the assessment questions is recorded in audio and video, and the video information and voice information of the user answering the assessment questions are obtained, so that the mental health assessment can be performed later based on the video information and voice information.

[0032] In some implementations, outputting the evaluation question based on the decision tree and obtaining video information and voice information of the user answering the evaluation question includes:

[0033] Outputting the key indicator question corresponding to the root node of the decision tree, and determining the user's answer tendency to the key indicator question according to the voice information;

[0034] The branch node to which the decision tree jumps is determined according to the answer tendency, and the evaluation question corresponding to the branch node is output.

[0035] For example, when conducting a medical consultation through a decision tree, it is first necessary to determine the main symptoms of the patient at the root node, so as to divide the patient into different subsets according to the characteristics determined at the root node, thereby jumping to the branch node and outputting the preset evaluation questions of each branch node. In the mental health evaluation method provided in the embodiment of the present application, the subsets corresponding to the branch nodes can be, for example, depressive symptoms and bipolar symptoms, respectively, and the severity of the user's depressive symptoms or bipolar symptoms is determined at each branch node according to the user's answer to the evaluation question, thereby realizing intelligent evaluation of mental health.

[0036] For example, the branch node to jump to can be determined based on the user's answer at the previous node (such as the root node), for example, the branch node corresponding to the user's answer tendency at the previous node is pre-set, where the answer tendency can include a positive tendency or a negative tendency. The answer tendency can be determined by an algorithm based on a sentiment dictionary, or by a pre-trained machine learning model, a deep learning model, or a large language model, which is not limited here.

[0037] In some implementations, the method further includes, before outputting the evaluation question based on the decision tree and obtaining the video information and voice information of the user answering the evaluation question:

[0038] Acquire a sample data set, the sample data set including patient status information corresponding to a sample user, preset evaluation questions received by the sample user, and answer information of the sample user to the preset evaluation questions;

[0039] The evaluation problem corresponding to each node of the decision tree is determined according to the information gain corresponding to each preset evaluation problem.

[0040] For example, the assessment questions preset in the decision tree are crucial to the accuracy of the mental health assessment results. The questions asked to patients in existing face-to-face consultations follow certain standards. For example, DSM-5 (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition) provides a set of detailed standards for the diagnosis of depression. A decision tree for mental health assessment can be constructed based on the sample data set collected in the face-to-face consultation.

[0041] Exemplarily, the sample data set includes patient status information corresponding to the sample user, preset evaluation questions received by the sample user, and answer information of the sample user to the preset evaluation questions, wherein the patient status information may be a diagnosis result of the sample user based on the answer information, so as to classify the sample user according to the patient status information, and thus construct a decision tree capable of classifying users based on the classification ability of the sample users by the preset evaluation questions, wherein the classification ability of the sample users by the preset evaluation questions may be reflected in the form of information gain.

[0042] In some implementations, determining the evaluation question corresponding to each node of the decision tree according to the information gain corresponding to each preset evaluation question includes:

[0043] Determining the initial information entropy of the sample data set according to the patient status information corresponding to the target sample data set, wherein the target sample data set is the sample data set corresponding to the current node, and the patient status information includes: at least one of a depressive state, a manic state, and a healthy state;

[0044] Determining the expected information entropy of each of the preset evaluation questions according to the answer tendency corresponding to the answer information of the sample users and the patient status information corresponding to each answer tendency;

[0045] Determining the information gain of each of the preset evaluation questions according to the difference between the initial information entropy and the expected information entropy;

[0046] The preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

[0047] Exemplarily, information entropy is an indicator to measure the purity of a sample set. If all samples in a sample set belong to the same category, the entropy is 0; if samples are evenly distributed in multiple categories, the entropy is maximum; information gain is used to reflect the magnitude of the reduction in information entropy. It can be understood that the process of outputting evaluation questions from a decision tree needs to be able to gradually reduce the information entropy of sample users. Therefore, the information gain of each preset evaluation question can be calculated separately, so as to determine the evaluation question corresponding to each node of the decision tree according to the information gain.

[0048] For example, the answer information of the sample users may contain a rich amount of information, but the most intuitive information is whether the sample users' answer tendencies are affirmative, neutral, or negative. The sample users are classified according to their answer tendencies, and the information entropy of the patient status information of each category of sample users is calculated to obtain the expected information entropy of this preset evaluation problem.

[0049] Exemplarily, the initial information entropy of the sample data set is determined according to the patient status information corresponding to the sample data set. It can be understood that the target sample data set is the sample data in the sample data set that is transferred to the current node after being classified by the previous node. For example, after the root node divides the sample data set into positive tendency, neutral tendency and negative tendency, the sample users with positive tendency, neutral tendency and negative tendency are respectively transferred to the corresponding next node. Taking the positive tendency as an example, the current node needs to determine the preset evaluation questions when asking questions to the sample users classified as positive tendency at the root node. At this time, the target sample data set is the sample data determined as positive tendency at the root node.

[0050] Exemplarily, the answer tendency may be obtained by analyzing the answer information of sample users according to a natural language analysis model, which will not be elaborated here.

[0051] Exemplarily, the initial information entropy can be calculated by the following formula:

[0052]

[0053] Among them, S represents the target sample data set, n represents the number of categories, and the probability of each category is p i .

[0054] Exemplarily, the expected information entropy can be calculated by the following formula:

[0055]

[0056] Among them, v represents the patient status information of the sample user. Each patient status information splits the target sample dataset into subsets S v , calculate each subset S v The information entropy H(S v ), and multiply it by the proportion of the subset to the target sample data set The expected value of the entropy after the feature split is obtained by summing all subsets. Information gain is the difference between the initial information entropy and the expected information entropy, that is, IG(S,A) = H(S)-Expected Entropy.

[0057] Exemplarily, the preset evaluation question with the largest information gain has the best classification ability for sample users, so the preset evaluation question is used as the evaluation question corresponding to the current node, and the preset evaluation question is removed from the sample data set to avoid the preset evaluation question from appearing repeatedly in the decision tree.

[0058] In some implementations, determining the preset evaluation problem with the maximum information gain as the evaluation problem corresponding to the current node includes:

[0059] When the depth of the current node is less than the preset depth, the number of samples corresponding to the current node is greater than the preset number, or the information entropy of the patient status information corresponding to each sample data corresponding to the current node is greater than the preset information entropy, the preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

[0060] Exemplarily, in the process of building a decision tree, reasonably determining when to stop splitting a node can prevent the overfitting of the decision tree. Specifically, the decision tree can be prevented from becoming too complex by limiting the maximum depth of the tree. If the depth of the decision tree reaches a preset threshold, the splitting is stopped; or, if the number of samples in the target sample data set corresponding to the current node is lower than a certain threshold (such as 5 or 10), it is considered that the samples of the current node are too few to be effectively split, so the splitting is stopped; or, when the information entropy of the target sample data set is less than or equal to the preset information entropy, it indicates that the samples in the current node are already "pure" enough, that is, there are no more features that can be used to improve the classification or the splitting has little improvement on the model, and the splitting can be stopped. On the contrary, if any of the above conditions is not met, the decision tree continues to be split, and the evaluation problem of the current node is determined according to the information gain.

[0061] Step S102: extract features from the video information to obtain expression information, and extract features from the voice information to obtain tone information and text information.

[0062] For example, in addition to the content of the user's answer to the assessment question, the user's expression and tone can also reflect certain information. For example, patients with depression often show reduced facial expressions, especially smiles or expressions showing other positive emotions, and they show lack of or slow reactions to topics or events that usually cause emotional reactions. Depressed patients may speak slower due to slow thinking, their pitch may change less, and their volume may decrease. Therefore, in order to determine the user's risk of illness through mental health assessment, in addition to text information, it is also necessary to extract the user's expression information from the video information and the user's tone information from the voice information.

[0063] For example, the expression information can be obtained by separating the user's expression from the video background by performing face detection and facial feature point location on the video information. Alternatively, the user's tone of voice information can be obtained by acquiring the pitch, volume, speech speed, etc. of the human voice in the voice information. Alternatively, the content in the voice information can be converted into text by speech recognition ASR (Automatic Speech Recognition) to obtain text information.

[0064] Step S103: determining a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information.

[0065] Exemplarily, the information of the three modes is vectorized through an encoder to obtain the user's vector to be evaluated, so that the vector to be evaluated can be processed by the model.

[0066] In some implementations, determining the to-be-evaluated vector corresponding to the user according to the expression information, the tone information, and the text information includes:

[0067] Converting the expression information into an expression vector, converting the tone information into a tone vector, and converting the text information into a text vector;

[0068] The expression vector, the tone vector and the text vector are concatenated to obtain the vector to be evaluated.

[0069] Exemplarily, the corresponding encoders are used for vectorization of expression information, tone information and text information, respectively, to obtain expression vectors, tone vectors and text vectors, and the above three vectors are concatenated based on a certain order to obtain the vector to be evaluated. For example, the expression information, tone information and text information can be vectorized by convolutional neural network CNN, recurrent neural network RNN, and topic model LDA, but it is not limited to this and is not limited here.

[0070] Step S104: determining a psychological health assessment result of the user according to the vector to be assessed.

[0071] For example, a model capable of evaluating the user's mental health based on the vector to be evaluated can be trained through patient samples in face-to-face consultations.

[0072] In some implementations, determining the user's mental health assessment result according to the vector to be assessed includes:

[0073] Inputting the vector to be evaluated into a preset deep learning model to obtain a multi-dimensional vector for representing a preset symptom;

[0074] The mental health assessment index of each symptom of the user is determined according to the multi-dimensional vector.

[0075] Exemplarily, the deep learning model can identify the emotional tendency in the information and analyze the multimodal information according to the context information. Therefore, the deep learning model can be used to evaluate the mental health. Specifically, the deep learning model used in the mental health evaluation method provided in the embodiment of the present application can be, for example, a BERT model, but is certainly not limited to this and is not limited here.

[0076] For example, the sample data set collected during the face-to-face consultation process can be converted into a sample vector, and the label of the sample vector can be determined according to the doctor's diagnosis results, and the BERT model can be trained so that the BERT model can determine the user's mental health assessment results based on the vector to be evaluated corresponding to the user during the consultation process.

[0077] Exemplarily, the mental health assessment result is output in the form of a mental health assessment index, indicating that the user is at risk of depression or bipolar symptoms, which serves as a basis for auxiliary diagnosis to improve the intelligence and accuracy of the diagnosis.

[0078] The mental health assessment method provided in the above embodiment outputs assessment questions based on a decision tree, and obtains video information and voice information of the user's answers to the assessment questions; extracts features from the video information to obtain expression information, and extracts features from the voice information to obtain tone information and text information; determines the vector to be assessed corresponding to the user based on the expression information, the tone information, and the text information; and determines the mental health assessment result of the user based on the vector to be assessed. Since the user's mental health is assessed by obtaining the user's answers to the assessment questions, the cost required for mental health consultation is reduced, and the intelligence level of mental health assessment is improved.

[0079] See also Figure 3 , Figure 3 It is a schematic diagram of a mental health assessment device provided in one embodiment of the present application. The mental health assessment device can be configured in a server or a terminal to execute the aforementioned mental health assessment method.

[0080] like Figure 3 As shown, the mental health assessment device includes: a feedback acquisition module 110, a feature extraction module 120, an information quantization module 130, and an information assessment module 140.

[0081] A feedback acquisition module 110, configured to output evaluation questions based on the decision tree, and acquire video information and voice information of the user's answers to the evaluation questions;

[0082] A feature extraction module 120 is used to extract features from the video information to obtain expression information, and to extract features from the voice information to obtain tone information and text information;

[0083] An information vectorization module 130, configured to determine a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information;

[0084] The information evaluation module 140 is used to determine the psychological health evaluation result of the user according to the vector to be evaluated.

[0085] Exemplarily, the feedback acquisition module 110, in the process of outputting the evaluation question based on the decision tree and acquiring the video information and voice information of the user's answer to the evaluation question, is used to implement:

[0086] Outputting the key indicator question corresponding to the root node of the decision tree, and determining the user's answer tendency to the key indicator question according to the voice information;

[0087] The branch node to which the decision tree jumps is determined according to the answer tendency, and the evaluation question corresponding to the branch node is output.

[0088] Exemplarily, before implementing the process of outputting the evaluation question based on the decision tree and obtaining the video information and voice information of the user answering the evaluation question, the mental health evaluation device is used to implement:

[0089] Acquire a sample data set, the sample data set including patient status information corresponding to a sample user, preset evaluation questions received by the sample user, and answer information of the sample user to the preset evaluation questions;

[0090] The evaluation problem corresponding to each node of the decision tree is determined according to the information gain corresponding to each preset evaluation problem.

[0091] Exemplarily, the mental health assessment device, in the process of determining the assessment questions corresponding to each node of the decision tree according to the information gain corresponding to each preset assessment question, is used to implement:

[0092] Determining the initial information entropy of the sample data set according to the patient status information corresponding to the target sample data set, wherein the target sample data set is the sample data set corresponding to the current node, and the patient status information includes: at least one of a depressive state, a manic state, and a healthy state;

[0093] Determining the expected information entropy of each of the preset evaluation questions according to the answer tendency corresponding to the answer information of the sample users and the patient status information corresponding to each answer tendency;

[0094] Determining the information gain of each of the preset evaluation questions according to the difference between the initial information entropy and the expected information entropy;

[0095] The preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

[0096] Exemplarily, the mental health assessment device, in the process of determining the preset assessment problem with the maximum information gain as the assessment problem corresponding to the current node, is used to implement:

[0097] When the depth of the current node is less than the preset depth, the number of samples corresponding to the current node is greater than the preset number, or the information entropy of the patient status information corresponding to each sample data corresponding to the current node is greater than the preset information entropy, the preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

[0098] Exemplarily, the information vectorization module 130, in the process of determining the vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information, is used to implement:

[0099] Converting the expression information into an expression vector, converting the tone information into a tone vector, and converting the text information into a text vector;

[0100] The expression vector, the tone vector and the text vector are concatenated to obtain the vector to be evaluated.

[0101] Exemplarily, the information evaluation module 140, in the process of determining the psychological health evaluation result of the user according to the vector to be evaluated, is used to implement:

[0102] Inputting the vector to be evaluated into a preset deep learning model to obtain a multi-dimensional vector for representing a preset symptom;

[0103] The mental health assessment index of each symptom of the user is determined according to the multi-dimensional vector.

[0104] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0105] The method and apparatus of the present application can be used in many general or special computing system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0106] Exemplarily, the above method and apparatus may be implemented in the form of a computer program. The computer program may be implemented in Figure 4 Runs on the computer device shown.

[0107] See also Figure 4 , Figure 4 A schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device may be a server or a terminal.

[0108] like Figure 4As shown, the computer device includes a processor, a memory, and a network interface connected via a system bus, wherein the memory may include a storage medium and an internal memory.

[0109] The storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any mental health assessment method.

[0110] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0111] The internal memory provides an environment for the operation of the computer program in the storage medium. When the computer program is executed by the processor, the processor can execute any mental health assessment method.

[0112] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0113] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0114] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0115] Outputting evaluation questions based on the decision tree, and obtaining video information and voice information of the user's answers to the evaluation questions;

[0116] Extracting features from the video information to obtain expression information, and extracting features from the voice information to obtain tone information and text information;

[0117] Determine a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information;

[0118] Determine a psychological health assessment result of the user according to the vector to be assessed.

[0119] In some implementations, the processor, in the process of outputting the evaluation question based on the decision tree and obtaining the video information and voice information of the user answering the evaluation question, is used to implement:

[0120] Outputting the key indicator question corresponding to the root node of the decision tree, and determining the user's answer tendency to the key indicator question according to the voice information;

[0121] The branch node to which the decision tree jumps is determined according to the answer tendency, and the evaluation question corresponding to the branch node is output.

[0122] In some implementations, before implementing the output of the evaluation question based on the decision tree and obtaining the video information and voice information of the user answering the evaluation question, the processor is used to implement:

[0123] Acquire a sample data set, the sample data set including patient status information corresponding to a sample user, preset evaluation questions received by the sample user, and answer information of the sample user to the preset evaluation questions;

[0124] The evaluation problem corresponding to each node of the decision tree is determined according to the information gain corresponding to each preset evaluation problem.

[0125] In some implementations, the processor, in the process of determining the evaluation problem corresponding to each node of the decision tree according to the information gain corresponding to each preset evaluation problem, is used to implement:

[0126] Determining the initial information entropy of the sample data set according to the patient status information corresponding to the target sample data set, wherein the target sample data set is the sample data set corresponding to the current node, and the patient status information includes: at least one of a depressive state, a manic state, and a healthy state;

[0127] Determining the expected information entropy of each of the preset evaluation questions according to the answer tendency corresponding to the answer information of the sample users and the patient status information corresponding to each answer tendency;

[0128] Determining the information gain of each of the preset evaluation questions according to the difference between the initial information entropy and the expected information entropy;

[0129] The preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

[0130] In some implementations, the processor, in the process of determining the preset evaluation problem with the maximum information gain as the evaluation problem corresponding to the current node, is configured to implement:

[0131] When the depth of the current node is less than the preset depth, the number of samples corresponding to the current node is greater than the preset number, or the information entropy of the patient status information corresponding to each sample data corresponding to the current node is greater than the preset information entropy, the preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

[0132] In some implementations, the processor, in the process of determining the vector to be evaluated corresponding to the user according to the expression information, the tone information, and the text information, is configured to implement:

[0133] Converting the expression information into an expression vector, converting the tone information into a tone vector, and converting the text information into a text vector;

[0134] The expression vector, the tone vector and the text vector are concatenated to obtain the vector to be evaluated.

[0135] In some implementations, the processor, in the process of determining the user's psychological health assessment result according to the vector to be assessed, is configured to implement:

[0136] Inputting the vector to be evaluated into a preset deep learning model to obtain a multi-dimensional vector for representing a preset symptom;

[0137] The mental health assessment index of each symptom of the user is determined according to the multi-dimensional vector.

[0138] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned description of the mental health assessment can refer to the corresponding process in the aforementioned mental health assessment control method embodiment, and will not be repeated here.

[0139] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. The computer program includes program instructions. The method implemented when the program instructions are executed can refer to the various embodiments of the mental health assessment method of the present application.

[0140] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0141] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0142] It should also be understood that the term "and / or" used in the specification of this application and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0143] The serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for assessing mental health, characterized in that: The method comprises: Outputting evaluation questions based on the decision tree, and obtaining video information and voice information of the user's answers to the evaluation questions; Extracting features from the video information to obtain expression information, and extracting features from the voice information to obtain tone information and text information; Determine a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information; Determine a psychological health assessment result of the user according to the vector to be assessed.

2. The method for evaluating mental health according to claim 1, characterized in that: The step of outputting the evaluation question based on the decision tree and obtaining the video information and voice information of the user answering the evaluation question includes: Outputting the key indicator question corresponding to the root node of the decision tree, and determining the user's answer tendency to the key indicator question according to the voice information; The branch node to which the decision tree jumps is determined according to the answer tendency, and the evaluation question corresponding to the branch node is output.

3. The method for evaluating mental health according to claim 1, characterized in that: Before outputting the evaluation question based on the decision tree and obtaining the video information and voice information of the user answering the evaluation question, the method further includes: Acquire a sample data set, the sample data set including patient status information corresponding to a sample user, preset evaluation questions received by the sample user, and answer information of the sample user to the preset evaluation questions; The evaluation problem corresponding to each node of the decision tree is determined according to the information gain corresponding to each preset evaluation problem.

4. The method for evaluating mental health according to claim 3, characterized in that: Determining the evaluation problem corresponding to each node of the decision tree according to the information gain corresponding to each preset evaluation problem includes: Determining the initial information entropy of the sample data set according to the patient status information corresponding to the target sample data set, wherein the target sample data set is the sample data set corresponding to the current node, and the patient status information includes: at least one of a depressive state, a manic state, and a healthy state; Determining the expected information entropy of each of the preset evaluation questions according to the answer tendency corresponding to the answer information of the sample users and the patient status information corresponding to each answer tendency; Determining the information gain of each of the preset evaluation questions according to the difference between the initial information entropy and the expected information entropy; The preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

5. The method for evaluating mental health according to claim 4, characterized in that: The step of determining the preset evaluation problem with the maximum information gain as the evaluation problem corresponding to the current node includes: When the depth of the current node is less than the preset depth, the number of samples corresponding to the current node is greater than the preset number, or the information entropy of the patient status information corresponding to each sample data corresponding to the current node is greater than the preset information entropy, the preset evaluation problem with the maximum information gain is determined as the evaluation problem corresponding to the current node.

6. The method for evaluating mental health according to claim 1, characterized in that: The determining the to-be-evaluated vector corresponding to the user according to the expression information, the tone information and the text information includes: Converting the expression information into an expression vector, converting the tone information into a tone vector, and converting the text information into a text vector; The expression vector, the tone vector and the text vector are concatenated to obtain the vector to be evaluated.

7. The method for evaluating mental health according to claim 1, characterized in that: The determining the psychological health assessment result of the user according to the vector to be assessed includes: Inputting the vector to be evaluated into a preset deep learning model to obtain a multi-dimensional vector for representing a preset symptom; The mental health assessment index of each symptom of the user is determined according to the multi-dimensional vector.

8. A mental health assessment device, characterized in that: The mental health assessment device comprises: A feedback acquisition module, used to output evaluation questions based on the decision tree, and acquire video information and voice information of the user's answer to the evaluation questions; A feature extraction module is used to extract features from the video information to obtain expression information, and to extract features from the voice information to obtain tone information and text information; An information vectorization module, used to determine a vector to be evaluated corresponding to the user according to the expression information, the tone information and the text information; An information evaluation module is used to determine the user's psychological health evaluation result according to the vector to be evaluated.

9. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, the steps of the mental health assessment method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the mental health assessment method according to any one of claims 1 to 7 are implemented.

Images