Method, device, electronic device and storage medium for generating medical consultation template

Abstract

The present embodiment relates to the field of artificial intelligence technology, and in particular to a method, device, electronic device and storage medium for generating a medical consultation template. The method comprises: obtaining a plurality of original medical consultation templates; each original medical consultation template comprises medical consultation question data, question option data and initial diagnosis result data, the medical consultation question data comprises a plurality of original medical consultation questions, the question option data comprises a plurality of original question options, and each original medical consultation question corresponds to at least two original question options; then the original medical consultation questions, original question options and initial diagnosis result data in the original medical consultation template are trained and processed to obtain the question dimension score value and the option dimension score value; then the original medical consultation questions in the original medical consultation template are screened and processed according to the question dimension score value and the option dimension score value to obtain the target medical consultation template. The technical solution of the embodiment of the present application can improve the patient's medical consultation experience and enhance the role of online medical consultation services.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, device, electronic device and storage medium for generating a medical consultation template. Background Art

[0002] In recent years, with the rapid development of Internet technology, the traditional medical field has gradually entered the Internet era, and more and more medical institutions are providing online consultation services through the Internet. Online consultation services refer to the process in which doctors communicate with patients through the Internet as a medium to achieve consultation.

[0003] In the related art, in online consultation services, doctors often use some consultation templates to provide consultation services to patients. However, the existing consultation templates have the defect of too many consultation questions, which makes patients unwilling to answer and results in low consultation efficiency. Summary of the invention

[0004] The main purpose of the embodiments of the present disclosure is to propose a method, device, electronic device and storage medium for generating a medical consultation template, which can improve the patient's medical consultation experience and enhance the role of online medical consultation services.

[0005] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present disclosure proposes a method for generating a medical consultation template, comprising:

[0006] Acquire multiple original medical consultation templates; wherein each of the original medical consultation templates includes medical consultation question data, question option data, and initial diagnosis result data, the medical consultation question data includes multiple original medical consultation questions, the question option data includes multiple original question options, and each of the original medical consultation questions corresponds to at least two of the original question options;

[0007] Input the original medical question and the initial diagnosis result data into a preset question learning model, adjust the question learning model parameters in the question learning model according to the original medical question and the initial diagnosis result data, and use the adjusted question learning model parameters as the question dimension score value; wherein the question dimension score value is used to characterize the importance of the original medical question in the original medical question template from the question dimension;

[0008] Inputting the initial diagnosis result data and the original question options into a preset option learning model, adjusting the option learning model parameters in the option learning model according to the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values; wherein the option score value is used to characterize the importance of the original question options in the original medical question;

[0009] The option score values ​​are classified and matched according to a preset mapping classification rule to obtain an option dimension score value; wherein the mapping classification rule is used to characterize the corresponding relationship between each of the original medical question and the original question option, and the option dimension score value is used to characterize the importance of the original medical question in the original medical question template from the option dimension;

[0010] Screening the original medical questions in the original medical question template according to the question dimension score value and the option dimension score value to obtain a target medical question;

[0011] A target medical inquiry template is constructed according to the target medical inquiry topic.

[0012] In some embodiments, the question learning model includes a question conversion model and a question classification model;

[0013] The step of adjusting the question learning model parameters in the question learning model by using the original medical question and the initial diagnosis result data, and using the adjusted question learning model parameters as question dimension score values, includes:

[0014] Obtaining the co-occurrence relationship between the original medical questions to obtain the question co-occurrence relationship;

[0015] Constructing a question co-occurrence pair of the original medical question based on the question co-occurrence relationship;

[0016] Training the question conversion model through the question co-occurrence pair to obtain a trained question conversion model;

[0017] Convert each of the original medical questions using the trained question conversion model to obtain a corresponding medical question vector;

[0018] The question classification model is updated according to the medical question vector and the initial diagnosis result data, and the question classification model parameters in the updated question classification model are used as the question dimension score values.

[0019] In some embodiments, the question conversion model includes a question encoding model and a coding conversion model;

[0020] The trained question conversion model is used to convert each of the original medical questions to obtain a corresponding medical question vector, including:

[0021] Encoding each of the original medical questions through the question coding model to obtain a corresponding medical question code;

[0022] The coding conversion model is used to perform vector conversion processing on the medical question coding to obtain the medical question vector.

[0023] In some embodiments, the updating of the question classification model according to the medical question vector and the initial diagnosis result data, and using the question classification model parameters in the updated question classification model as the question dimension score value, includes:

[0024] Fine-tuning the parameters of the question classification model according to the medical question vector and the initial diagnosis result data;

[0025] Obtain the weight of each question vector in the question classification model after parameter fine-tuning to obtain the question weight value;

[0026] The question weight value is normalized, and the normalized question weight value is used as the question dimension score value.

[0027] In some embodiments, the option learning model includes an option conversion model and an option classification model;

[0028] The step of adjusting the option learning model parameters in the option learning model by using the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values, comprises:

[0029] Obtaining the co-occurrence relationship between the options of the original question to obtain the option co-occurrence relationship;

[0030] Constructing option co-occurrence pairs of the original question options based on the option co-occurrence relationships;

[0031] Training the option conversion model through the option co-occurrence pairs to obtain a trained option conversion model;

[0032] Convert each of the original question options using the trained option conversion model to obtain a corresponding question option vector;

[0033] The option classification model is updated according to the question option vector and the initial diagnosis result data, and the option classification model parameters in the updated option classification model are used as the option score values.

[0034] In some embodiments, the updating of the option classification model according to the question option vector and the initial diagnosis result data, and using the option classification model parameters in the updated option classification model as the option score value, includes:

[0035] Fine-tuning the parameters of the option classification model according to the question option vector and the initial diagnosis result data;

[0036] Obtain the weight of each question option vector in the option classification model after parameter fine-tuning, and obtain the option weight value;

[0037] The option weight value is normalized, and the normalized option weight value is used as the option score value.

[0038] In some embodiments, the screening of the original medical questions in the original medical question template according to the question dimension score value and the option dimension score value to obtain the target medical question includes:

[0039] Obtain the preset medical question weights and the preset question option weights;

[0040] Get the preset score threshold;

[0041] The question dimension score values ​​are weighted according to the question weights to obtain question weight values;

[0042] The option dimension score values ​​are weighted according to the question option weights to obtain option weight values;

[0043] Obtaining a target score value according to the question weighted value and the option weighted value;

[0044] If the target score value is greater than or equal to the score threshold, the original medical question corresponding to the target score value is obtained from the original medical question template to obtain the target medical question title.

[0045] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present disclosure provides a device for generating a medical inquiry template, comprising:

[0046] An acquisition module is used to acquire a plurality of original medical consultation templates; wherein each of the original medical consultation templates includes medical consultation question data, question option data and initial diagnosis result data, the medical consultation question data includes a plurality of original medical consultation questions, the question option data includes a plurality of original question options, and each of the original medical consultation questions corresponds to at least two of the original question options;

[0047] A first processing module is used to input the original medical question and the initial diagnosis result data into a preset question learning model, adjust the question learning model parameters in the question learning model according to the original medical question and the initial diagnosis result data, and use the adjusted question learning model parameters as question dimension score values; wherein the question dimension score value is used to characterize the importance of the original medical question in the original medical question template from the question dimension;

[0048] A second processing module is used to input the initial diagnosis result data and the original question options into a preset option learning model, adjust the option learning model parameters in the option learning model according to the initial diagnosis result data and the original question options, and use the adjusted option learning model parameters as option score values; wherein the option score value is used to characterize the importance of the original question options in the original medical question;

[0049] A classification matching module, used to perform classification matching processing on the option score values ​​according to a preset mapping classification rule to obtain an option dimension score value; wherein the mapping classification rule is used to characterize the corresponding relationship between each of the original medical question and the original question option, and the option dimension score value is used to characterize the importance of the original medical question in the original medical question template from the option dimension;

[0050] A screening module, used for screening the original medical questions in the original medical question template according to the score value of the question dimension and the score value of the option dimension to obtain a target medical question;

[0051] A construction module is used to construct a target medical inquiry template according to the target medical inquiry topic.

[0052] To achieve the above objective, a third aspect of an embodiment of the present disclosure provides an electronic device, including:

[0053] at least one memory;

[0054] at least one processor;

[0055] at least one program;

[0056] The program is stored in the memory, and the processor executes at least one program to implement:

[0057] A method as described in any one of the embodiments of the first aspect.

[0058] To achieve the above-mentioned purpose, a fourth aspect of an embodiment of the present disclosure provides a storage medium, wherein the storage medium is a computer-readable storage medium, and the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute:

[0059] A method as described in any one of the embodiments of the first aspect.

[0060] The consultation template generation method, device, electronic device and storage medium proposed in the embodiments of the present disclosure perform importance analysis on the original consultation questions and original question options in the original consultation template to obtain question dimension score values ​​for characterizing the importance of the original consultation questions in the original consultation template and option score values ​​for characterizing the importance of the original question options in the original consultation template. Then, the option dimension score values ​​are obtained according to the option score values ​​and preset mapping classification rules. The original consultation questions in the original consultation template are screened and processed according to the question dimension score values ​​and the option dimension score values ​​to obtain the target consultation template, thereby optimizing the original consultation template, reducing the original consultation questions in the original consultation template, improving the patient's consultation experience, enhancing the role of the online consultation service, and improving the effectiveness of the online consultation service. In addition, in the entire target consultation template generation process, no human participation is required, which frees up manpower and improves efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a flow chart of a method for generating a medical consultation template provided in an embodiment of the present application;

[0062] Figure 2 yes Figure 1 Flow chart of the specific method of step S200;

[0063] Figure 3 yes Figure 2 Flow chart of the specific method of step S240;

[0064] Figure 4 yes Figure 2 Flow chart of the specific method of step S250;

[0065] Figure 5 yes Figure 1 Flow chart of the specific method of step S300;

[0066] Figure 6 yes Figure 5 Flow chart of the specific method of step S350;

[0067] Figure 7 yes Figure 1 Flow chart of the specific method of step S500;

[0068] Figure 8 It is a module block diagram of the device for generating a medical inquiry template provided in an embodiment of the present application;

[0069] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0071] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0072] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0073] First, some nouns involved in this application are analyzed:

[0074] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0075] Medical cloud: Medical cloud refers to the use of cloud computing to create a medical and health service cloud platform based on new technologies such as cloud computing, mobile technology, multimedia, 4G communication, big data, and the Internet of Things, combined with medical technology, to achieve the sharing of medical resources and the expansion of medical scope. Because of the application of cloud computing technology, medical cloud improves the efficiency of medical institutions and facilitates residents to seek medical treatment. For example, hospital appointment registration, electronic medical records, medical insurance, etc. are all products of the combination of cloud computing and the medical field. Medical cloud also has the advantages of data security, information sharing, dynamic expansion, and overall layout.

[0076] One-Hot Encoding: One-hot encoding is also called one-bit valid code. Its method is to use N-bit state register to encode N states. Each state has its own independent register bit, and at any time, only one of them is valid. That is, only one bit is 1, and the rest are zero values. One-hot encoding is widely used in machine learning algorithms such as regression, classification, and clustering. In machine learning algorithms such as regression, classification, and clustering, the calculation of distance or similarity between features is very important. The commonly used distance or similarity calculations are similarity calculations in Euclidean space, and the calculation of cosine similarity is based on Euclidean space. Using one-hot encoding, the values ​​of discrete features are extended to Euclidean space, and a certain value of discrete features corresponds to a certain point in Euclidean space. Using one-hot encoding for discrete features will make the distance calculation between features more reasonable.

[0077] Unsupervised learning: Using unlabeled data to learn the distribution of data or the relationship between data is called unsupervised learning. The biggest difference between supervised learning and unsupervised learning is whether the data is labeled. The most commonly used scenarios for unsupervised learning are clustering and dimensionality reduction.

[0078] Logistic Regression: Logistic regression is a generalized linear regression analysis model, which is often used in data mining, automatic disease diagnosis, economic forecasting and other fields. For example, it explores the risk factors that cause diseases and predicts the probability of disease occurrence based on risk factors. In logistic regression, the weights of independent variables can be obtained through logistic regression analysis, so that the classification and prediction of results can be achieved. Logistic regression is a classification model, which is often used for binary classification. Its essence is: assuming that the data follows this distribution, and then using maximum likelihood estimation to estimate the parameters.

[0079] Neural network model (NN): A neural network is a mathematical model or computational model that imitates the structure and function of a biological neural network. A neural network is calculated by connecting a large number of artificial neurons. In most cases, an artificial neural network can change its internal structure based on external information and is an adaptive system. Modern neural networks are a nonlinear statistical data modeling tool that is often used to model complex relationships between inputs and outputs, or to explore data patterns. A neural network is a computational model consisting of a large number of nodes (or "neurons") and connections between them. Each node represents a specific output function, called an activation function. The connection between each two nodes represents a weighted value for the signal passing through the connection, called a weight, which is equivalent to the memory of an artificial neural network. The output of the network varies depending on the network's connection method, weight value, and activation function. The network itself is usually an approximation of a certain algorithm or function in nature, or it may be an expression of a logical strategy.

[0080] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0081] The method for generating a medical consultation template provided in the embodiment of the present application can be applied to artificial intelligence. Basic artificial intelligence technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. Artificial intelligence software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0082] In recent years, with the development of Internet technology, the traditional medical field has gradually entered the Internet era, and more and more medical institutions are providing medical services through the Internet. Among them, online consultation services have received more and more attention. Online consultation refers to the process in which doctors use the Internet as a medium to communicate with patients to achieve consultation. Different from traditional offline consultation, online consultation services can provide patients with high-quality medical services more conveniently and quickly. In order to improve the efficiency of online doctors' consultation, doctors generally prepare consultation templates for different departments in advance based on medical knowledge and consultation practice experience. The consultation template consists of at least a dozen or as many as dozens of consultation questions and question options (single choice or multiple choice). The consultation template is usually thrown out by the system before the conversation between the doctor and the patient. After the patient answers all the template questions, the template will provide the doctor with a preliminary diagnosis result. Based on the questions and answers collected by the template and the preliminary diagnosis results, the doctor will ask more questions or directly give the diagnosis result. However, there are some shortcomings in the existing consultation template. First, the number of questions in the consultation templates prepared by doctors is too large, which reduces the patient's online consultation experience. Some patients even give up the consultation due to too many questions. Second, the questions for different diseases have different focuses. Using the same set of consultation templates for different diseases will reduce the effectiveness of the consultation.

[0083] Based on this, the embodiments of the present application provide a method, device, electronic device and storage medium for generating a medical consultation template, which can optimize the existing templates, reduce the medical consultation questions in the existing medical consultation templates, improve the patient's medical consultation experience, enhance the role of online medical consultation services, and improve the efficiency of online medical consultation services.

[0084] The method, device, electronic device and storage medium for generating a medical inquiry template provided by the embodiments of the present disclosure are specifically described through the following embodiments. First, the method for generating a medical inquiry template in the embodiments of the present disclosure is described.

[0085] The method for generating a medical consultation template provided in the embodiment of the present application relates to the field of artificial intelligence technology. The method for generating a medical consultation template provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, or a smart watch, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, or can be configured as a cloud server that provides basic cloud computing services such as 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 big data and artificial intelligence platforms; the software can be an application that implements the method for generating a medical consultation template, etc., but is not limited to the above forms.

[0086] The disclosed embodiments can be used in numerous general or special computer 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 computer devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. 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, and the like 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.

[0087] Reference Figure 1 In the first aspect, some embodiments of the present application provide a method for generating a medical inquiry template, including step S100, step S200, step S300, step S400, step S500 and step S600. It should be understood that the method for generating a medical inquiry template of the embodiments of the present application includes but is not limited to steps S100 to S600. These six steps are described in detail below.

[0088] Step S100: Acquire multiple original medical consultation templates; wherein each original medical consultation template includes medical consultation question data, question option data and initial diagnosis result data, the medical consultation question data includes multiple original medical consultation questions, the question option data includes multiple original question options, and each original medical consultation question corresponds to at least two original question options;

[0089] In step S100, the original medical consultation template can be a general medical consultation template or a specific medical consultation template. A specific medical consultation template is a medical consultation template used by a specific department; a general medical consultation template refers to a medical consultation template that can be used by all departments without distinguishing between departments. The original medical consultation template can be obtained through a medical cloud server, or by connecting to a hospital's database system, or by other acquisition methods.

[0090] It should be noted that for each original medical consultation template obtained, the original medical consultation template must include initial diagnosis result data, medical question data and question option data. The medical question data includes multiple original medical questions, and the question option data includes multiple original question options. The original medical questions refer to the medical questions on the original medical consultation template, and the original question options refer to the options corresponding to each medical question. Each original medical question corresponds to at least two original question options; the initial diagnosis result data refers to the preliminary diagnosis result made by the doctor based on the questions answered by the patient on the original medical consultation template.

[0091] It should be noted that, in this embodiment, multiple original medical question templates refer to obtaining multiple original medical question templates of the same type. For example, if the original medical question template of type A needs to be optimized, multiple original medical question templates of type A need to be obtained. If both the original medical question templates of type A and type B need to be optimized, multiple original medical question templates of type A need to be obtained; and multiple original medical question templates of type B need to be obtained. If two or more types of original medical question templates need to be optimized, the subsequent optimization screening process and training processing process should be carried out separately.

[0092] Step S200: inputting the original medical question and the initial diagnosis result data into a preset question learning model, adjusting the question learning model parameters in the question learning model according to the original medical question and the initial diagnosis result data, and using the adjusted question learning model parameters as the question dimension score value; wherein the question dimension score value is used to characterize the importance of the original medical question in the original medical question template from the question dimension;

[0093] In step S200, the original medical question and the initial diagnosis result data are input into a preset question learning model for training. Specifically, the initial diagnosis result data is used as the target label (result) and the original medical question is used as the input to fine-tune the parameters in the question learning model. Then, the adjusted parameters of the question learning model are used as the question dimension score value, which is used to characterize the importance of the original medical question in the original medical template.

[0094] Step S300: inputting the initial diagnosis result data and the original question options into a preset option learning model, adjusting the option learning model parameters in the option learning model according to the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values; wherein the option score value is used to characterize the importance of the original question options in the original medical question;

[0095] Similar to the aforementioned step S200, in step S300, the parameters in the option learning model are fine-tuned with the initial diagnosis result data as the target label (result) and the original question options as input. Then, the adjusted parameters of the option learning model are used as the option score value, which is used to characterize the importance of the original question options in the original medical question.

[0096] Step S400: classify and match the option score values ​​according to the preset mapping classification rules to obtain the option dimension score values; wherein the mapping classification rules are used to characterize the corresponding relationship between each original medical question and the original question options, and the option dimension score values ​​are used to characterize the importance of the original medical question in the original medical question template from the option dimension;

[0097] In step S400, the mapping classification rule is the corresponding relationship between each original medical question and the original question options. For example, the original medical question contains 3 options, then from the option dimension, the option dimension score value is the sum of the 3 option score values.

[0098] Step S500: Screening the original medical questions in the original medical question template according to the question dimension score value and the option dimension score value to obtain the target medical question;

[0099] In step S500, after obtaining the question dimension score and the option dimension score according to the above steps, it is possible to roughly determine the percentage of each original medical question in the initial diagnosis result data from the question dimension, and the percentage of each original medical question in the initial diagnosis result data from the option dimension. Therefore, the original medical questions in the original medical template can be screened according to the question dimension score and the option dimension score, so as to obtain the target medical question.

[0100] The specific screening process is as follows: the problem dimension score and the option dimension score are added together, and then the target score corresponding to each original medical question is obtained, and then the target scores are sorted from large to small according to the added target scores, and the original medical questions with high rankings are selected as the target medical questions; or the problem dimension score and the option dimension score are added together, and then the target score corresponding to each original medical question is obtained, and then a score threshold is set, and the original medical questions corresponding to the target scores exceeding the score threshold are selected as the target medical questions.

[0101] Step S600: constructing a target medical inquiry template according to the target medical inquiry topic.

[0102] In step S600, all the obtained target medical questions and the corresponding original question options are combined to obtain a target medical question template. The target medical question template is a template obtained by optimizing the original medical question template.

[0103] The medical consultation template generation method of the embodiment of the present application performs an importance analysis on the original medical consultation questions and the original question options in the original medical consultation template to obtain a question dimension score value for characterizing the importance of the original medical consultation questions in the original medical consultation template and an option score value for characterizing the importance of the original question options in the original medical consultation template. Then, the option dimension score value is obtained according to the option score value and a preset mapping classification rule. The original medical consultation questions in the original medical consultation template are screened and processed according to the question dimension score value and the option dimension score value to obtain a target medical consultation template, thereby optimizing the original medical consultation template, reducing the original medical consultation questions in the original medical consultation template, improving the patient's medical consultation experience, enhancing the role of the online medical consultation service, and improving the effectiveness of the online medical consultation service. In addition, in the entire target medical consultation template generation process, no human participation is required, which frees up manpower and improves efficiency.

[0104] Reference Figure 2 In some embodiments of the present application, the question learning model includes a question conversion model and a question classification model. Step S200 includes step S210, step S220, step S230, step S240 and step S250. It should be understood that step S200 includes but is not limited to step S210 to step S250. Figure 2 These five steps are introduced in detail.

[0105] Step S210: Obtain the co-occurrence relationship between the original medical questions to obtain the question co-occurrence relationship;

[0106] Step S220: constructing question co-occurrence pairs of the original medical questions based on the question co-occurrence relationship;

[0107] Specifically, in step S210 to step S220, the problem co-occurrence relationship, that is, the relationship in which the original medical questions co-occur, is specifically described in detail using Table 1 as an example, where Table 1 is a problem co-occurrence relationship table provided in an embodiment of the present application.

[0108]

[0109] Table 1

[0110] In Table 1, numbers 1, 2, and 3 represent that the original medical consultation template has 3 medical consultation data, numbers Q1 to Q5 represent that the original medical consultation template has 5 original medical consultation questions, 0 in each original medical consultation question indicates that the patient did not answer the question or did not have the question, and 1 indicates that the patient answered the question, and the initial diagnosis result data is the preliminary diagnosis result obtained by the doctor based on the original medical consultation questions.

[0111] According to the question co-occurrence relationship in Table 1, question co-occurrence pairs can be constructed. For example, for the medical consultation data numbered 1, the existing question co-occurrence pairs are: (Q1, Q3), (Q1, Q4), (Q3, Q1), (Q3, Q4), (Q4, Q1), (Q4, Q3).

[0112] It should be noted that in this embodiment, in order to preserve data distribution, repeated question co-occurrence pairs should be retained, and question co-occurrence pairs have a front-to-back order. For example, (Q1, Q3) and (Q3, Q1) are different question co-occurrence pairs.

[0113] Step S230: training the question conversion model through question co-occurrence to obtain a trained question conversion model;

[0114] Step S240: converting each original medical question through the trained question conversion model to obtain a corresponding medical question vector;

[0115] According to the large number of question co-occurrence pairs constructed in the above steps, in step S230 and step S240, the first question of each question co-occurrence pair is used as the input of the question conversion model, and the second question is used as the output of the question conversion model, and the question conversion model is trained by unsupervised learning. After training, each original medical question is converted by the question conversion model to obtain the corresponding medical question vector.

[0116] It should be noted that the question conversion model can be obtained by the following steps: first, a neural network with multiple layers of fully connected hidden layers is constructed, and then the neural network is trained and processed by unsupervised learning to obtain a question conversion model; during the training process, the first question of each question co-occurrence pair is used as the input of the question conversion model, and the second question is used as the output of the question conversion model. For example, for the question co-occurrence pair (Q1, Q3), during training, Q1 is the input of the question conversion model, and Q3 is the output of the question conversion model. Of course, other methods can also be used to obtain the question conversion model, and this application does not make specific restrictions on this.

[0117] Step S250: updating the question classification model according to the medical question vector and the initial diagnosis result data, and using the question classification model parameters in the updated question classification model as the question dimension score value.

[0118] In step S250, the medical question vector obtained above is used as the input of the question classification model, and the initial diagnosis result data is used as the output of the question classification model. The question classification model is updated and trained, and then the updated parameters are obtained as the question dimension score value.

[0119] It should be noted that, in this embodiment, for each piece of medical consultation data, the original medical consultation question that appears is represented by a medical consultation question vector, and the original medical consultation question that does not appear is represented by a zero vector.

[0120] Reference Figure 3 In some embodiments of the present application, the question conversion model includes a question encoding model and a coding conversion model. Step S240 includes but is not limited to step S241 and step 242. Figure 3 These two steps are introduced in detail.

[0121] Step S241: encoding each original medical question through the question encoding model to obtain the corresponding medical question code;

[0122] In step S241, the question encoding model is a one-hot encoding model, and the original medical question is one-hot encoded by the one-hot encoding model to obtain the corresponding medical question encoding, so that each original medical question is represented by a one-hot encoding.

[0123] Step S242: Perform vector conversion processing on the medical question code through the coding conversion model to obtain the medical question vector.

[0124] In step S242, the encoding conversion model can be obtained by the following steps: first, a neural network with multiple layers of fully connected hidden layers is constructed, and then the neural network is trained by unsupervised learning to obtain the encoding conversion model. Then, each question code is input into the encoding conversion model to obtain the corresponding question vector, and the question vector is represented by V Q (Q i ) represents, i represents the i-th one.

[0125] Reference Figure 4 In some embodiments of the present application, step S250 includes but is not limited to step S251, step S252, and step S253. Figure 4 These three steps are introduced in detail.

[0126] Step S251: fine-tuning the parameters of the question classification model according to the question vector and the initial diagnosis result data as the output of the question classification model;

[0127] Step S252: Obtain the weight of each question vector in the question classification model after parameter fine-tuning to obtain the question weight value;

[0128] Step S253: normalize the question weight value, and use the normalized question weight value as the question dimension score value.

[0129] Specifically, in this embodiment, for each piece of medical consultation data, the original medical consultation question that appears is represented by the corresponding medical consultation question vector, and the question that does not appear is represented by the zero vector. For each original medical consultation question, a preset initial parameter w is given to represent the weight. Then the question classification model can be represented by formula (1), which is specifically as follows:

[0130] T Q =w0+w1gV Q (Q1)+w2gV Q (Q2)+...+w N V Q (Q N ) (1)

[0131] In formula (1), T Q represents the initial diagnosis result data, and the initial parameter w is the parameter of the question classification model. When the initial diagnosis result data is known, the initial parameter w of each original question is fine-tuned through a large amount of consultation data, so as to realize the training and updating of the question classification model. Then, the weight of each question vector in the question classification model after parameter fine-tuning is obtained to obtain the question weight value, that is, the parameter w corresponding to each original question is obtained, and then the obtained parameter w is normalized to obtain the corresponding question dimension score value. The specific calculation process is shown in formula (2), which is:

[0132]

[0133] Through formula (2), the problem dimension score value corresponding to each original medical question is obtained.

[0134] Reference Figure 5 In some embodiments of the present application, the option learning model includes an option conversion model and an option classification model. Step S300 includes but is not limited to step S310, step S320, step S330, step S340 and step S350. Figure 5 These five steps are introduced in detail.

[0135] Step S310: Obtain the co-occurrence relationship between the original question options to obtain the option co-occurrence relationship;

[0136] Step S320: constructing option co-occurrence pairs of original question options based on option co-occurrence relationships;

[0137] Specifically, in step S310 to step S320, the option co-occurrence relationship is the relationship in which the original question options co-occur, and is specifically described in detail using Table 2 as an example. Table 2 is an option co-occurrence relationship table provided in an embodiment of the present application.

[0138]

[0139] Table 2

[0140] In Table 2, numbers 1, 2, and 3 represent that the original medical consultation template has 3 medical consultation data, numbers Q1 to Q5 represent that the original medical consultation template has 5 original medical consultation questions, q1, q2, and q3 represent the original question options corresponding to each original medical consultation question, respectively. Each original medical consultation question has multiple options, corresponding to multiple choices or single choices. 0 in each original medical consultation question indicates that the patient does not have this option or did not select this option, and 1 indicates that the patient selected this option. The initial diagnosis result data is the preliminary diagnosis result obtained by the doctor based on the original medical consultation question and the original question options.

[0141] Similarly, multiple option co-occurrence pairs can be constructed according to the option co-occurrence relationship in Table 2. For example, for the consultation data numbered 1, the option co-occurrence pairs are: (q1,1, q3,1), (q1,1, q4,1), (q1,1, q4,3), (q3,1, q1,1), (q3,1, q4,1), (q3,1, q4,3), (q4,1, q3,1), (q4,3, q1,1), (q4,3, q3,1).

[0142] It should be noted that in this embodiment, in order to preserve the data distribution, repeated option co-occurrence pairs should still be retained, and option co-occurrence pairs have a front-to-back order. The original question options in the same question do not construct option co-occurrence pairs. For example, (q1,1, q3,1) and (q3,1, q1,1) are different option co-occurrence pairs, and there will be no option co-occurrence pairs in the form of (q4,1, q4,3).

[0143] Step S330: training the option conversion model through option co-occurrence pairs to obtain a trained option conversion model;

[0144] Step S340: convert each original question option through the trained option conversion model to obtain a corresponding question option vector;

[0145] Based on the large number of option co-occurrence pairs constructed in the above steps, similar to the above step S230, in step S330, the first original question option in each option co-occurrence pair is used as the input of the option conversion model, and the second original question option is used as the output of the option conversion model, and the option conversion model is trained by unsupervised learning. After training, each original question option is processed by one-hot encoding to obtain the question option code, and then converted by the trained option conversion model to obtain the question option vector, which is used as V q (qi,j) represents.

[0146] Step S350: updating the option classification model according to the question option vector and the initial diagnosis result data, and using the option classification model parameters in the updated option classification model as the option score value.

[0147] Reference Figure 6 In some embodiments of the present application, step S350 includes but is not limited to step S351, step S352 and step S353. Figure 6 These three steps are described in detail.

[0148] Step S351: fine-tuning the parameters of the option classification model according to the question option vector and the initial diagnosis result data;

[0149] Step S352: Obtain the weight of each question option vector in the option classification model after parameter fine-tuning to obtain the option weight value;

[0150] Step S353: normalize the option weight value, and use the normalized option weight value as the option score value.

[0151] Specifically, in this embodiment, for each piece of medical consultation data, the original question options that appear are represented by the corresponding question option vector, and the original question options that do not appear are represented by the zero vector. For each original question option, a preset initial parameter w is given. i,j represents the weight, then the option classification model can be expressed by formula (3), which is as follows:

[0152] T q =w0+w 1,1 V q (q1,1)+w 1,2 V q (q1,2)+w 1,3 V q (q1,3)+w 2,1 V q (q2,1)+...+w N,M V q (qN,M)(3)

[0153] In formula (3), T q Represents the initial diagnosis result data, the initial parameter w i,j is the parameter of the option classification model. When the initial diagnosis result data is known, the initial parameter w of each original question option is calculated through a large amount of consultation data. i,j Fine-tune the option classification model to achieve training and updating. Then obtain the weight of each question option vector in the option classification model after parameter fine-tuning to obtain the option weight value, that is, obtain the parameter w corresponding to each original question option. i,j , and then get the parameter w i,j After normalization, the corresponding option score value is obtained, and its calculation formula is shown in formula (4):

[0154]

[0155] The option score value corresponding to each original question option is obtained through formula (4).

[0156] Reference Figure 7 In some embodiments of the present application, step S500 includes but is not limited to step S510, step S520, step S530, step S540, step S550 and step S560. Figure 7 These six steps are introduced in detail.

[0157] Step S510: Obtaining preset medical question weights and preset question option weights;

[0158] Step S520: Obtaining a preset score threshold;

[0159] Step S530: weighting the question dimension score values ​​according to the question weights to obtain the question weighted values;

[0160] Step S540: weighting the option dimension score values ​​according to the question option weights to obtain the option weighted values;

[0161] Step S550: Obtaining a target score value according to the question weight value and the option weight value;

[0162] Step S560: If the target score value is greater than or equal to the score threshold, the original medical question corresponding to the target score value is obtained from the original medical question template to obtain the target medical question.

[0163] Specifically, in this embodiment, the weight of the question refers to the weight on the question dimension, expressed as w Q Indicates that the question option weight refers to the weight on the option dimension, expressed as w qIndicates. Both the weight of the question and the weight of the question options are preset. Then the target score value is obtained by formula (5), which is:

[0164] w=w Q gw i '+w q gw i ” (5)

[0165] In formula (5), w represents the target score value, w i " represents the score value of the option dimension. The target score value of each original question is obtained by formula (5), and then the original questions with target scores greater than or equal to the score threshold are screened out from the original question template to obtain the target question, which is used as part of the target question template. The original questions with target scores less than the score threshold do not contribute much to the initial diagnosis result data, so they are screened out and do not need to be included in the target question template.

[0166] Reference Figure 8 In the second aspect, some embodiments of the present application also provide a medical inquiry template generating device, including an acquisition module 710, a first processing module 720, a second processing module 730, a classification matching model 740, a screening module 750 and a construction module 760.

[0167] Acquisition module 710 is used to acquire multiple original medical consultation templates; wherein each original medical consultation template includes medical consultation question data, question option data and initial diagnosis result data, the medical consultation question data includes multiple original medical consultation questions, the question option data includes multiple original question options, and each original medical consultation question corresponds to at least two original question options.

[0168] The first processing module 720 is used to input the original medical question and the initial diagnosis result data into a preset question learning model, adjust the question learning model parameters in the question learning model according to the original medical question and the initial diagnosis result data, and use the adjusted question learning model parameters as the question dimension score value; wherein the question dimension score value is used to characterize the importance of the original medical question in the original medical question template from the question dimension.

[0169] The second processing module 730 is used to input the initial diagnosis result data and the original question options into a preset option learning model, adjust the option learning model parameters in the option learning model according to the initial diagnosis result data and the original question options, and use the adjusted option learning model parameters as option score values; wherein the option score value is used to characterize the importance of the original question options in the original medical question.

[0170] The classification matching model 740 is used to perform classification matching processing on the option score values ​​according to a preset mapping classification rule to obtain the option dimension score value; wherein the mapping classification rule is used to characterize the corresponding relationship between each original medical question and the original question option, and the option dimension score value is used to characterize the importance of the original medical question in the original medical question template from the option dimension;

[0171] The screening module 750 is used to screen the original medical questions in the original medical question template according to the question dimension score value and the option dimension score value to obtain the target medical question.

[0172] The construction module 760 is used to construct a target medical inquiry template according to the target medical inquiry topic.

[0173] The medical consultation template generating device of the embodiment of the present application performs an importance analysis on the original medical consultation questions and the original question options in the original medical consultation template to obtain a question dimension score value for characterizing the importance of the original medical consultation questions in the original medical consultation template and an option score value for characterizing the importance of the original question options in the original medical consultation template. Then, the option dimension score value is obtained according to the option score value and a preset mapping classification rule. The original medical consultation questions in the original medical consultation template are screened and processed according to the question dimension score value and the option dimension score value to obtain a target medical consultation template, thereby optimizing the original medical consultation template, reducing the original medical consultation questions in the original medical consultation template, improving the patient's medical consultation experience, enhancing the role of the online medical consultation service, and improving the effectiveness of the online medical consultation service. In addition, in the entire target medical consultation template generation process, no human participation is required, which frees up manpower and improves efficiency.

[0174] It should be noted that the medical inquiry template generating device of the embodiment of the present application corresponds to the aforementioned medical inquiry template generating method. For the specific screening process, training process, etc., please refer to the aforementioned medical inquiry template generating method, which will not be repeated here.

[0175] The present disclosure also provides an electronic device, including:

[0176] at least one memory;

[0177] at least one processor;

[0178] at least one program;

[0179] The program is stored in the memory, and the processor executes at least one program to implement the above-mentioned method for generating a medical consultation template in the present disclosure. The electronic device can be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), a car computer, etc.

[0180] Combine the following Fig. 9 The electronic device according to the embodiment of the present application is introduced in detail.

[0181] like Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0182] The processor 810 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present disclosure;

[0183] The memory 820 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 820 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 820, and the processor 810 calls and executes the method for generating a medical consultation template in the embodiment of the present disclosure;

[0184] Input / output interface 830, used to implement information input and output;

[0185] The communication interface 840 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0186] bus 850 , which transmits information between the various components of the device (e.g., processor 810 , memory 820 , input / output interface 830 , and communication interface 840 );

[0187] The processor 810 , the memory 820 , the input / output interface 830 and the communication interface 840 are connected to each other in communication within the device via a bus 850 .

[0188] The embodiment of the present disclosure further provides a storage medium, which is a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned medical consultation template generation method.

[0189] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0190] The embodiments described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems.

[0191] Those skilled in the art will appreciate that the technical solutions shown in the figures do not limit the embodiments of the present disclosure and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0192] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0193] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0194] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0195] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0196] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0197] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0198] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0199] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or 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 multiple instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of each embodiment method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store programs.

[0200] The preferred embodiments of the present disclosure are described above with reference to the accompanying drawings, but the scope of the rights of the present disclosure is not limited thereto. Any modification, equivalent substitution and improvement made by those skilled in the art without departing from the scope and essence of the present disclosure should be within the scope of the rights of the present disclosure.

Claims

1. A method for generating a medical inquiry template, characterized in that: include: Acquire multiple original medical consultation templates; wherein each of the original medical consultation templates includes medical consultation question data, question option data, and initial diagnosis result data, the medical consultation question data includes multiple original medical consultation questions, the question option data includes multiple original question options, and each of the original medical consultation questions corresponds to at least two of the original question options; Input the original medical question and the initial diagnosis result data into a preset question learning model, adjust the question learning model parameters in the question learning model according to the original medical question and the initial diagnosis result data, and use the adjusted question learning model parameters as the question dimension score value; wherein the question dimension score value is used to characterize the importance of the original medical question in the original medical question template from the question dimension; Inputting the initial diagnosis result data and the original question options into a preset option learning model, adjusting the option learning model parameters in the option learning model according to the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values; wherein the option score value is used to characterize the importance of the original question options in the original medical question; The option score values ​​are classified and matched according to a preset mapping classification rule to obtain an option dimension score value; wherein the mapping classification rule is used to characterize the corresponding relationship between each of the original medical question and the original question option, and the option dimension score value is used to characterize the importance of the original medical question in the original medical question template from the option dimension; Screening the original medical questions in the original medical question template according to the question dimension score value and the option dimension score value to obtain a target medical question; Constructing a target medical consultation template according to the target medical consultation topic; The step of adjusting the question learning model parameters in the question learning model by using the original medical question and the initial diagnosis result data, and using the adjusted question learning model parameters as the question dimension score values, includes: Obtaining the co-occurrence relationship between the original medical questions to obtain the question co-occurrence relationship; Constructing a question co-occurrence pair of the original medical question based on the question co-occurrence relationship; The step of adjusting the option learning model parameters in the option learning model by using the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values, includes: Obtaining the co-occurrence relationship between the options of the original question to obtain the option co-occurrence relationship; Option co-occurrence pairs of the original question options are constructed based on the option co-occurrence relationships.

2. The method according to claim 1, characterized in that The problem learning model includes a problem conversion model and a problem classification model; The adjusting and processing of the question learning model parameters in the question learning model by using the original medical question and the initial diagnosis result data, and using the adjusted question learning model parameters as the question dimension score value, further includes: Training the question conversion model through the question co-occurrence pair to obtain a trained question conversion model; Convert each of the original medical questions using the trained question conversion model to obtain a corresponding medical question vector; The question classification model is updated according to the medical question vector and the initial diagnosis result data, and the question classification model parameters in the updated question classification model are used as the question dimension score values.

3. The method according to claim 2, characterized in that The problem conversion model includes a problem encoding model and a coding conversion model; The trained question conversion model is used to convert each of the original medical questions to obtain a corresponding medical question vector, including: Encoding each of the original medical questions through the question coding model to obtain a corresponding medical question code; The coding conversion model is used to perform vector conversion processing on the medical question coding to obtain the medical question vector.

4. The method according to claim 2, characterized in that: The updating process of the question classification model according to the medical question vector and the initial diagnosis result data, and using the question classification model parameters in the updated question classification model as the question dimension score value, includes: Fine-tuning the parameters of the question classification model according to the medical question vector and the initial diagnosis result data; Obtain the weight of each question vector in the question classification model after parameter fine-tuning to obtain the question weight value; The question weight value is normalized, and the normalized question weight value is used as the question dimension score value.

5. The method according to claim 1, characterized in that The option learning model includes an option conversion model and an option classification model; The adjusting and processing of the option learning model parameters in the option learning model by using the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values, further includes: Training the option conversion model through the option co-occurrence pairs to obtain a trained option conversion model; Convert each of the original question options using the trained option conversion model to obtain a corresponding question option vector; The option classification model is updated according to the question option vector and the initial diagnosis result data, and the option classification model parameters in the updated option classification model are used as the option score values.

6. The method according to claim 5, characterized in that The updating process of the option classification model according to the question option vector and the initial diagnosis result data, and using the option classification model parameters in the updated option classification model as the option score value, includes: Fine-tuning the parameters of the option classification model according to the question option vector and the initial diagnosis result data; Obtain the weight of each question option vector in the option classification model after parameter fine-tuning, and obtain the option weight value; The option weight value is normalized, and the normalized option weight value is used as the option score value.

7. The method according to any one of claims 1 to 6, characterized in that: The screening process of the original medical question in the original medical question template according to the question dimension score value and the option dimension score value to obtain the target medical question includes: Obtain the preset medical question weights and the preset question option weights; Get the preset score threshold; The question dimension score values ​​are weighted according to the question weights to obtain question weight values; The option dimension score values ​​are weighted according to the question option weights to obtain option weight values; Obtaining a target score value according to the question weighted value and the option weighted value; If the target score value is greater than or equal to the score threshold, the original medical question corresponding to the target score value is obtained from the original medical question template to obtain the target medical question title.

8. A device for generating a medical inquiry template, characterized in that: include: An acquisition module is used to acquire a plurality of original medical consultation templates; wherein each of the original medical consultation templates includes medical consultation question data, question option data and initial diagnosis result data, the medical consultation question data includes a plurality of original medical consultation questions, the question option data includes a plurality of original question options, and each of the original medical consultation questions corresponds to at least two of the original question options; A first processing module is used to input the original medical question and the initial diagnosis result data into a preset question learning model, adjust the question learning model parameters in the question learning model according to the original medical question and the initial diagnosis result data, and use the adjusted question learning model parameters as question dimension score values; wherein the question dimension score value is used to characterize the importance of the original medical question in the original medical question template from the question dimension; A second processing module is used to input the initial diagnosis result data and the original question options into a preset option learning model, adjust the option learning model parameters in the option learning model according to the initial diagnosis result data and the original question options, and use the adjusted option learning model parameters as option score values; wherein the option score value is used to characterize the importance of the original question options in the original medical question; A classification matching module, used to perform classification matching processing on the option score values ​​according to a preset mapping classification rule to obtain an option dimension score value; wherein the mapping classification rule is used to characterize the corresponding relationship between each of the original medical question and the original question option, and the option dimension score value is used to characterize the importance of the original medical question in the original medical question template from the option dimension; A screening module, used for screening the original medical questions in the original medical question template according to the score value of the question dimension and the score value of the option dimension to obtain a target medical question; A construction module, used to construct a target medical consultation template according to the target medical consultation topic; The step of adjusting the question learning model parameters in the question learning model by using the original medical question and the initial diagnosis result data, and using the adjusted question learning model parameters as the question dimension score values, includes: Obtaining the co-occurrence relationship between the original medical questions to obtain the question co-occurrence relationship; Constructing a question co-occurrence pair of the original medical question based on the question co-occurrence relationship; The step of adjusting the option learning model parameters in the option learning model by using the initial diagnosis result data and the original question options, and using the adjusted option learning model parameters as option score values, includes: Obtaining the co-occurrence relationship between the options of the original question to obtain the option co-occurrence relationship; Option co-occurrence pairs of the original question options are constructed based on the option co-occurrence relationships.

9. An electronic device, characterized in that: include: at least one memory; at least one processor; at least one program; The program is stored in the memory, and the processor executes at least one program to implement: A method as claimed in any one of claims 1 to 7.

10. A storage medium, the storage medium being a computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to cause a computer to execute: A method as claimed in any one of claims 1 to 7.

Images