Medical teaching plan generation method and device, equipment, storage medium and program product

Abstract

The invention relates to a medical teaching plan generation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining personalized configuration parameters including student level parameters, teaching target parameters and teaching plan language parameters; obtaining N medical teaching plan contents matched with the personalized configuration parameters from a preset database; searching real-time teaching plan contents from at least one preset website according to keywords corresponding to the personalized configuration parameters; adding the N medical teaching plan contents, the real-time teaching plan contents and the personalized configuration parameters into a first prompt template to obtain a first input text; the first prompt template comprises prompt words used for prompting the preset large model to execute a medical teaching plan generation task; and inputting the first input text into a preset large model to obtain a target medical teaching plan output by the preset large model. By adopting the method, the personalized medical teaching plan can be efficiently, accurately and automatically generated, and the frontier of the teaching content is ensured.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, computer device, computer-readable storage medium and computer program product for generating medical teaching plans. Background Art

[0002] Due to the rapid development and updating of medical knowledge, traditional medical education methods face many challenges. Teachers and other professionals need to spend a lot of time and energy to write and update teaching plans to ensure the accuracy and cutting-edge nature of teaching content, which makes it difficult to meet the growing teaching needs and diversified teaching content.

[0003] The current teaching plan generation system provides some automation functions, relying on the manual editing of professionals, using predefined templates and rules to generate teaching plans. This method of generating content has low flexibility and personalization, making it difficult to meet the needs of different teaching scenarios, and has a slow update speed, making it difficult to keep up with the rapid update of medical knowledge. Summary of the invention

[0004] Based on this, it is necessary to provide a medical teaching plan generation method, device, computer equipment, computer readable storage medium and computer program product to address the above-mentioned technical problems, which can efficiently, accurately and automatically generate personalized medical teaching plans to ensure the cutting-edge nature of teaching content.

[0005] In a first aspect, the present application provides a method for generating a medical teaching plan, comprising:

[0006] Obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters;

[0007] Obtain N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database;

[0008] Searching for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters;

[0009] Adding N pieces of medical teaching plan content, real-time teaching plan content and personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting a preset large model to perform a medical teaching plan generation task;

[0010] The first input text is input into the preset large model to obtain the target medical teaching plan output by the preset large model.

[0011] In one embodiment, obtaining N pieces of medical teaching plan content matching the personalized configuration parameters from a preset database includes:

[0012] Vectorize the personalized configuration parameters to obtain a configuration vector;

[0013] N teaching plan vectors with the highest similarity to the configuration vector are selected from a preset database, and N pieces of medical teaching plan content corresponding to the N teaching plan vectors are obtained; the preset database stores teaching plan vectors obtained by vectorizing multiple pieces of medical teaching plan content.

[0014] In one embodiment, after inputting the first input text into the preset large model to obtain the target medical teaching plan output by the preset large model, the method further includes:

[0015] The target medical teaching plans were checked for grammar, content, and accuracy;

[0016] If the target medical teaching plan fails any one of the grammar check, content check and accuracy check, the failed check item and the target medical teaching plan are added to the second prompt template to obtain a second input text; the second prompt template includes prompt words for prompting the preset macro model to pay attention to the failed check item and perform the medical teaching plan regeneration task;

[0017] The second input text is input into the preset large model to obtain a medical teaching plan regenerated by the preset large model.

[0018] In one embodiment, the target medical teaching case is subjected to a content check, including:

[0019] Obtain keywords of knowledge points corresponding to teaching objectives;

[0020] Perform keyword matching on the target medical teaching plan according to the knowledge point keywords to obtain keyword matching results;

[0021] Segment the target medical teaching plan to obtain multiple paragraphs;

[0022] Detect the text similarity between any two paragraphs among multiple paragraphs;

[0023] Determine the duplicate content detection result based on the text similarity and the set threshold;

[0024] Based on the keyword matching results and duplicate content detection results, the target medical teaching plans were checked for content.

[0025] In one embodiment, the target medical teaching plan is checked for accuracy, including:

[0026] Extract medical knowledge points from target medical teaching plans;

[0027] Acquire the knowledge entity and first related information corresponding to the medical knowledge point from the medical knowledge graph;

[0028] searching a medical database for second relevant information according to the knowledge entity;

[0029] Based on the first relevant information and the second relevant information, the accuracy of the teaching plan content corresponding to the medical knowledge point in the target medical teaching plan is checked.

[0030] In one embodiment, after inputting the first input text into the preset large model to obtain the target medical teaching plan output by the preset large model, the method further includes:

[0031] Evaluate the translation scores of the target medical teaching plans;

[0032] If the translation score is less than the set score, the parameters of the preset large model are adjusted;

[0033] The target medical teaching plan is input into the adjusted preset large model for re-translation to obtain the medical teaching plan regenerated by the preset large model.

[0034] In a second aspect, the present application also provides a medical teaching plan generating device, comprising:

[0035] Configuration module, used to obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters;

[0036] A matching module, used to obtain N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database;

[0037] A search module, used to search for real-time teaching plan content from at least one preset website according to keywords corresponding to the personalized configuration parameters;

[0038] A prompt module, used for adding N pieces of medical teaching plan content, real-time teaching plan content and personalized configuration parameters into a first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting a preset large model to perform a medical teaching plan generation task;

[0039] A generation module is used to input the first input text into a preset large model to obtain a target medical teaching plan output by the preset large model.

[0040] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0041] Obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters;

[0042] Obtain N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database;

[0043] Searching for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters;

[0044] Adding N pieces of medical teaching plan content, real-time teaching plan content and personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting a preset large model to perform a medical teaching plan generation task;

[0045] The first input text is input into the preset large model to obtain the target medical teaching plan output by the preset large model.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0047] Obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters;

[0048] Obtain N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database;

[0049] Searching for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters;

[0050] Adding N pieces of medical teaching plan content, real-time teaching plan content and personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting a preset large model to perform a medical teaching plan generation task;

[0051] The first input text is input into the preset large model to obtain the target medical teaching plan output by the preset large model.

[0052] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0053] Obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters;

[0054] Obtain N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database;

[0055] Searching for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters;

[0056] Adding N pieces of medical teaching plan content, real-time teaching plan content and personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting a preset large model to perform a medical teaching plan generation task;

[0057] The first input text is input into the preset large model to obtain the target medical teaching plan output by the preset large model.

[0058] The above-mentioned medical lesson plan generation method, device, computer equipment, computer-readable storage medium and computer program product obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching target parameters and lesson plan language parameters; obtain N medical lesson plan contents matching the personalized configuration parameters from a preset database; search for real-time lesson plan contents from at least one preset website according to the keywords corresponding to the personalized configuration parameters; add N medical lesson plan contents, real-time lesson plan contents and personalized configuration parameters to the first prompt template to obtain the first input text; the first prompt template includes prompt words for prompting the preset large model to perform the medical lesson plan generation task; input the first input text into the preset large model to obtain the target medical lesson plan output by the preset large model. Through the above-mentioned method, based on the semantic understanding ability of the preset large model, a lesson plan with strict logic and reasonable structure can be generated; personalized lesson plans suitable for different student groups can be provided according to different teaching student levels, teaching goals and teaching languages; it has the ability to track the latest medical trends in real time, automatically update the lesson plan content, and ensure the cutting-edge nature and accuracy of the teaching content. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 An application environment diagram of a method for generating a medical teaching plan in an embodiment;

[0061] Figure 2 A schematic diagram of a process of generating a medical teaching plan in one embodiment;

[0062] Figure 3 A schematic diagram of a flow chart of a method for generating a medical teaching plan in another embodiment;

[0063] Figure 4 is a structural block diagram of a medical teaching plan generating device in one embodiment;

[0064] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] 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.

[0066] The medical teaching plan generation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through a communication network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. In one implementation, the terminal 102 receives the personalized configuration parameters input by the user, transmits the personalized configuration parameters to the server 104 through the communication network, and the server 104 generates a target lesson plan according to the personalized configuration parameters, and returns the generated target lesson plan to the terminal 102 through the communication network for the user to view. Optionally, the user can also directly enter the personalized configuration parameters on the server 104. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, smart watches, etc. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0067] In an exemplary embodiment, Figure 2 As shown, a method for generating a medical teaching plan is provided, and the method is applied to Figure 1 The server 104 in FIG. 1 is used as an example to illustrate, including:

[0068] Step 202, obtaining personalized configuration parameters; the personalized configuration parameters include student level parameters, teaching goal parameters and lesson plan language parameters.

[0069] Among them, the server receives or obtains the personalized configuration parameters input by the user from the terminal. In one implementation, a questioning program is designed, and by running the questioning program, the user is guided to input the student level parameter, the teaching goal parameter and the lesson plan language parameter. The student level parameter is used to characterize the student's professional knowledge mastery or the current grade, the teaching goal parameter is used to characterize the expected teaching effect or the expected knowledge points mastered by the student, and the lesson plan language parameter is used to characterize the language of the expected output lesson plan, for example, Chinese, English, etc.

[0070] Step 204, obtaining N pieces of medical teaching plan content matching the personalized configuration parameters from a preset database.

[0071] Among them, data is collected from data sources such as medical databases, literature, and teaching materials, and preprocessed to obtain multiple medical teaching plan contents and store them in a preset database. Optionally, the server is provided with Axios (a network request library based on Promise), sends HTTP requests to obtain web page content, uses cheerio library to parse web page data (HTML data), extracts required data through selectors, and stores them in an array. Optionally, the preprocessing step includes removing special characters and extra spaces.

[0072] In one implementation, a corresponding knowledge point label and difficulty label are pre-set for each medical teaching plan content in a preset database. When matching, the difficulty label that matches the student level parameter and the knowledge point label that matches the teaching goal parameter are determined, and N medical teaching plan contents corresponding to the difficulty label and the knowledge point label are selected.

[0073] In one implementation, the preset database is a vector database. After obtaining the medical teaching plan content from the web page, the text content is vectorized and stored in the vector database. When matching, the vectorized personalized configuration parameters are matched with the vectors in the vector database for similarity, and the N medical teaching plan contents corresponding to the N vectors with the highest similarity are selected.

[0074] Step 206, searching for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters.

[0075] Among them, keywords are extracted from personalized configuration parameters using TF-IDF (term frequency-inverse document frequency), pre-trained network models, etc. The server is configured in advance with at least one credible medical knowledge point acquisition website. Search on a preset website based on the extracted keywords to extract real-time teaching plan content. Optionally, a search time period for real-time content is set (for example, within one year). After searching based on keywords, screening is performed according to the search time period to determine cutting-edge medical knowledge points as real-time teaching plan content.

[0076] Step 208, adding N medical teaching plan contents, real-time teaching plan contents and personalized configuration parameters to the first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting the preset large model to perform the medical teaching plan generation task.

[0077] Among them, the first prompt template is a large model prompt template constructed in advance, including multiple placeholders and prompt words for prompting the preset large model to perform the medical lesson plan generation task, and N medical lesson plan contents, real-time lesson plan contents and personalized configuration parameters are filled into the corresponding placeholders respectively to obtain the first input text. The first input text is used to guide the preset large model to understand the lesson plan content and personalized configuration parameters, and perform the medical lesson plan generation task. The prompt words used to prompt the preset large model to perform the medical lesson plan generation task, for example: "Your task is to generate a medical lesson plan that matches the following personalized configuration parameters based on the following N medical lesson plan contents and real-time lesson plan contents."

[0078] Step 210, input the first input text into the preset large model to obtain the target medical teaching plan output by the preset large model.

[0079] The preset large model refers to a large model (Large Model) or a large language model (LargeLanguage Model) set in advance. The large model, which can also be called a foundation model or a foundation model (i.e., FoundationModels), refers to a neural network model containing ultra-large-scale parameters. The preset large model of this embodiment can be any open source large model, and this embodiment does not limit this.

[0080] The first input text is input into the preset large model, and the semantic understanding ability of the preset large model is used to analyze N medical teaching plan contents and real-time teaching plan contents, determine and organize the medical knowledge points that are consistent with the student level parameters and teaching target parameters, install the teaching plan language parameters for translation, and finally generate the target medical teaching plan that matches the personalized configuration parameters.

[0081] In one implementation, the preset large model is set on a remote server, and the local server sets the API key and URL (Uniform Resource Locator) through the remote server's API (Application Programming Interface), configures the request header information, configures the first input text and generated parameters in the request body, and sends a POST request to the remote server. After receiving the API response, the text content in the generated parameters is extracted to obtain the target medical teaching plan.

[0082] In the above-mentioned medical lesson plan generation method, personalized configuration parameters are obtained; personalized configuration parameters include student level parameters, teaching goal parameters and lesson plan language parameters; N medical lesson plan contents matching the personalized configuration parameters are obtained from a preset database; according to the keywords corresponding to the personalized configuration parameters, real-time lesson plan contents are searched from at least one preset website; N medical lesson plan contents, real-time lesson plan contents and personalized configuration parameters are added to the first prompt template to obtain the first input text; the first prompt template includes prompt words for prompting the preset large model to perform the medical lesson plan generation task; the first input text is input into the preset large model to obtain the target medical lesson plan output by the preset large model. Through the above-mentioned method, based on the semantic understanding ability of the preset large model, a lesson plan with strict logic and reasonable structure can be generated; personalized lesson plans suitable for different student groups can be provided according to different teaching student levels, teaching goals and teaching languages; it has the ability to track the latest medical trends in real time, automatically update the lesson plan content, and ensure the cutting-edge nature and accuracy of the teaching content.

[0083] In an exemplary embodiment, step 204 includes: vectorizing the personalized configuration parameters to obtain a configuration vector; selecting N teaching plan vectors with the highest similarity to the configuration vector from a preset database, and obtaining N medical teaching plan contents corresponding to the N teaching plan vectors; the preset database stores teaching plan vectors obtained by vectorizing multiple medical teaching plan contents.

[0084] Among them, one hot model, bag of words model, N-gram model, word-vector model (Word2vec) and other methods can be used for vectorization processing. In the specific implementation, the preset database is a vector database, which stores the lesson plan vectors obtained by vectorizing the content of the medical lesson plan in advance. When matching with the personalized configuration parameters, the configuration vectors corresponding to the personalized configuration parameters are matched with the lesson plan vectors in the preset database for similarity, and the N lesson plan vectors with the highest similarity are selected. Optionally, the similarity between vectors is calculated by cosine similarity, Euclidean distance and other methods, which is not limited in this embodiment. N is an integer greater than 1 and can be set by the user according to actual needs.

[0085] In this embodiment, by calculating the similarity between the configuration vector and the lesson plan vector, the lesson plan content that best meets the learner's needs can be quickly found without manual screening and evaluation.

[0086] In an exemplary embodiment, Figure 3 As shown, after step 210, the method further includes:

[0087] Step 302, performing grammar check, content check and accuracy check on the target medical teaching plan.

[0088] Among them, after the target medical teaching plan is generated, a multi-level review mechanism is added to improve the accuracy of the output content of the preset large model, including: grammar check, content check and accuracy check. Optionally, a natural language processing tool (such as language detection and grammar check tool) can be used to perform a grammar check on the target medical teaching plan to confirm the correctness of the grammar and spelling of the generated target medical teaching plan. The content check is to check whether the content of the target medical teaching plan meets expectations and whether the content is redundant. The accuracy check is to check whether the knowledge points in the target medical teaching plan are accurate.

[0089] Step 304: If the target medical lesson plan fails any one of the grammar check, content check, and accuracy check, the failed check item and the target medical lesson plan are added to the second prompt template to obtain a second input text; the second prompt template includes prompt words for prompting the preset large model to pay attention to the failed check item and perform the medical lesson plan regeneration task.

[0090] Among them, the second prompt template is a large model prompt template constructed in advance, including multiple placeholders and prompt words for prompting the preset large model to pay attention to the failed inspection items and perform the medical lesson plan regeneration task. The failed inspection items and the target medical lesson plan are filled in the corresponding placeholders respectively to obtain the second input text. The second input text is used to guide the preset large model to understand the failed inspection items and the target medical lesson plan, and perform the medical lesson plan regeneration task. The failed inspection items, that is, the inspection items that the target medical lesson plan fails to pass, refer to any one of the grammar check, content check and accuracy check. If the grammar check fails, the grammar check items and the target medical lesson plan are added to the second prompt template; if the content check fails, the content check items and the target medical lesson plan are added to the second prompt template; if the accuracy check fails, the accuracy check items and the target medical lesson plan are added to the second prompt template.

[0091] Step 306, input the second input text into the preset large model to obtain a medical teaching plan regenerated by the preset large model.

[0092] Among them, the second input text is input into the preset large model, and the semantic understanding ability of the preset large model is used to analyze the target medical lesson plan, pay attention to the failed inspection items, and execute the medical lesson plan regeneration task, so that the regenerated target medical lesson plan can avoid the inspection items that failed the previous time, thereby improving the quality of the regenerated medical lesson plan.

[0093] In this embodiment, the accuracy of the output content of the preset large model can be improved through grammar checking, content checking and accuracy checking.

[0094] In an exemplary embodiment, a content check is performed on a target medical teaching plan, including: obtaining knowledge point keywords corresponding to the teaching objectives; performing keyword matching on the target medical teaching plan based on the knowledge point keywords to obtain keyword matching results; segmenting the target medical teaching plan to obtain multiple paragraphs; detecting text similarity between any two paragraphs in the multiple paragraphs; determining a duplicate content detection result based on the text similarity and a set threshold; and performing a content check on the target medical teaching plan based on the keyword matching results and the duplicate content detection results.

[0095] Among them, the teaching objective parameters are processed by using text analysis tools, natural language processing tools, etc., to assist in extracting knowledge point keywords, and a keyword matching algorithm is used to determine whether the target medical teaching plan content contains the specified knowledge points and themes, and obtain the keyword matching results. If the target medical teaching plan includes the knowledge point keyword, it indicates that the target medical teaching plan contains the specified knowledge points and themes, and the keyword matching result is a successful keyword matching; if the target medical teaching plan does not match the knowledge point keyword, it indicates that the target medical teaching plan does not cover the specified knowledge points and themes, and the keyword matching result is a failed keyword matching.

[0096] The target medical teaching plan is segmented using natural language processing tools to obtain multiple paragraphs, and the repeated parts in the target medical teaching plan are detected using text similarity algorithms (such as cosine similarity and Jaccard similarity). The threshold is set as the similarity critical value set in advance based on experience. If the text similarity between two paragraphs in the target medical teaching plan is higher than the set threshold, it indicates that there are repeated paragraphs in the target medical teaching plan, and the repeated content detection result is that there is repeated content; if the text similarity between any two paragraphs in the target medical teaching plan is less than or equal to the set threshold, it indicates that there are no repeated paragraphs in the target medical teaching plan, and the repeated content detection result is that there is no repeated content.

[0097] If the keyword matching result is a successful keyword match and the duplicate content detection result is that there is no duplicate content, then it is determined that the target medical teaching plan has passed the content check; if the keyword matching result is a failed keyword match, or the duplicate content detection result is that there is duplicate content, then it is determined that the target medical teaching plan has not passed the content check.

[0098] In this embodiment, keyword matching is used to ensure that the content in the medical teaching plan is closely related to the teaching objectives, and repeated content in the medical teaching plan is detected, which helps to streamline the teaching content, thereby improving the quality of the medical teaching plan output by the preset large model.

[0099] In an exemplary embodiment, a target medical lesson plan is checked for accuracy, including: extracting medical knowledge points from the target medical lesson plan; acquiring knowledge entities and first related information corresponding to the medical knowledge points from a medical knowledge graph; querying second related information in a medical database based on the knowledge entities; and checking the accuracy of lesson plan content corresponding to the medical knowledge points in the target medical lesson plan based on the first related information and the second related information.

[0100] Among them, text analysis tools, natural language processing tools, etc. are used to process the target medical teaching plan to assist in extracting medical knowledge points, such as disease names, drug names, etc. The medical knowledge points are used as query conditions to search in the medical knowledge graph, and the knowledge entities and first related information matching the query conditions are returned. The first related information can be the definition, symptoms, treatment methods, etc. of the medical knowledge points stored in the medical knowledge graph. For the retrieved knowledge entities, the corresponding second related information is further queried in the medical database. The second related information can be the literature, guidelines or cases corresponding to the knowledge entity.

[0101] Using automated tools, the teaching plan contents corresponding to the medical knowledge points in the target medical teaching plan are compared with the first relevant information and the second relevant information respectively to verify the accuracy of the target medical teaching plan.

[0102] It is understandable that the medical knowledge graph includes accurate medical knowledge, covering diseases, drugs, tests, surgical operations, etc., and contains structured information such as entities, relationships, attributes, etc. The medical database is a database constructed by integrating authoritative medical database resources, drug instructions, clinical guidelines and other open data in advance as a basis for auxiliary verification.

[0103] In this embodiment, since medical knowledge is changing rapidly and is highly specialized, by checking the target medical teaching plan through the medical knowledge graph and the medical database, possible erroneous information in the teaching plan can be discovered in a timely manner, ensuring that students can receive the latest and scientifically recognized medical knowledge.

[0104] In an exemplary embodiment, after step 210, the method further includes: evaluating the translation score of the target medical teaching plan; if the translation score is less than the set score, adjusting the parameters of the preset large model; inputting the target medical teaching plan into the adjusted preset large model for re-translation to obtain a medical teaching plan regenerated by the preset large model.

[0105] Among them, the translation score of the target medical teaching plan is evaluated based on the automatic evaluation index. If the translation score is less than the set score, it indicates that the translation quality is unqualified, the parameters of the large model are adjusted, and the translation steps are repeated. The set score is the evaluation standard set in advance according to the automatic evaluation index. For example, the BLEU (Bilingual Evaluation Understudy) score must reach 0.7 or above to be considered as a qualified translation, and the ROUGE-L (Recall-Oriented Understudy for Gisting Evaluation) score must reach 0.8 or above to be considered as a qualified translation.

[0106] If the translation score is greater than or equal to the set score, the target medical teaching plan is output and returned to the terminal that issued the personalized configuration parameters.

[0107] The parameters of the large model refer to the various adjustable settings and values ​​used in large-scale machine learning models, namely weights and biases. Optionally, in combination with the medical knowledge graph and medical database, the parameters of the preset large model can be adjusted through domain adaptation or fine-tuning to improve the model's understanding of medical terms and context.

[0108] In this embodiment, by continuously iterating and optimizing the parameters of the large model, the accuracy, fluency and professionalism of the translation of medical teaching plans can be significantly improved.

[0109] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0110] Based on the same inventive concept, the embodiment of the present application also provides a medical teaching plan generation device for implementing the medical teaching plan generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more medical teaching plan generation device embodiments provided below can refer to the limitations of the medical teaching plan generation method above, and will not be repeated here.

[0111] In an exemplary embodiment, Figure 4As shown, a medical teaching plan generating device is provided, comprising:

[0112] Configuration module 402 is used to obtain personalized configuration parameters; personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters.

[0113] The matching module 404 is used to obtain N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database.

[0114] The search module 406 is used to search for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters.

[0115] The prompt module 408 is used to add N medical teaching plan contents, real-time teaching plan contents and personalized configuration parameters to the first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting the preset large model to perform the medical teaching plan generation task.

[0116] The generation module 410 is used to input the first input text into the preset large model to obtain the target medical teaching plan output by the preset large model.

[0117] In the above-mentioned medical teaching plan generation device, personalized configuration parameters are obtained; personalized configuration parameters include student level parameters, teaching target parameters and teaching plan language parameters; N medical teaching plan contents matching the personalized configuration parameters are obtained from a preset database; according to the keywords corresponding to the personalized configuration parameters, real-time teaching plan contents are searched from at least one preset website; N medical teaching plan contents, real-time teaching plan contents and personalized configuration parameters are added to the first prompt template to obtain the first input text; the first prompt template includes prompt words for prompting the preset large model to perform the medical teaching plan generation task; the first input text is input into the preset large model to obtain the target medical teaching plan output by the preset large model. Through the above-mentioned method, based on the semantic understanding ability of the preset large model, a teaching plan with strict logic and reasonable structure can be generated; personalized teaching plans suitable for different student groups can be provided according to different teaching student levels, teaching goals and teaching languages; it has the ability to track the latest medical trends in real time, automatically update the teaching plan content, and ensure the cutting-edge nature and accuracy of the teaching content.

[0118] In an exemplary embodiment, the matching module 404 is also used to vectorize the personalized configuration parameters to obtain a configuration vector; select N teaching plan vectors with the highest similarity to the configuration vector from a preset database, and obtain N medical teaching plan contents corresponding to the N teaching plan vectors; the preset database stores teaching plan vectors obtained by vectorizing multiple medical teaching plan contents.

[0119] In an exemplary embodiment, the medical teaching plan generating device further includes a checking module and a regeneration module;

[0120] A checking module is used to perform grammar checking, content checking and accuracy checking on the target medical teaching plan respectively;

[0121] A regeneration module is used for adding the failed check items and the target medical lesson plan to a second prompt template to obtain a second input text if the target medical lesson plan fails any one of the grammar check, content check and accuracy check; the second prompt template includes prompt words for prompting the preset large model to pay attention to the failed check items and perform the medical lesson plan regeneration task; the second input text is input into the preset large model to obtain a medical lesson plan regenerated by the preset large model.

[0122] In an exemplary embodiment, the inspection module is also used to obtain knowledge point keywords corresponding to the teaching objectives; perform keyword matching on the target medical teaching plan based on the knowledge point keywords to obtain keyword matching results; segment the target medical teaching plan to obtain multiple paragraphs; detect the text similarity between any two paragraphs in the multiple paragraphs; determine the duplicate content detection result based on the text similarity and a set threshold; and perform content inspection on the target medical teaching plan based on the keyword matching results and the duplicate content detection results.

[0123] In an exemplary embodiment, the inspection module is also used to extract medical knowledge points from the target medical teaching plan; obtain knowledge entities and first related information corresponding to the medical knowledge points from the medical knowledge graph; query second related information in the medical database based on the knowledge entities; and perform accuracy inspection on the teaching plan content corresponding to the medical knowledge points in the target medical teaching plan based on the first related information and the second related information.

[0124] In an exemplary embodiment, the medical teaching plan generating device further includes a translation module;

[0125] The translation module is used to evaluate the translation score of the target medical teaching plan; if the translation score is less than the set score, the parameters of the preset large model are adjusted; the target medical teaching plan is input into the adjusted preset large model for re-translation to obtain the medical teaching plan regenerated by the preset large model.

[0126] Each module in the above-mentioned medical teaching plan generation device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0127] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store teaching plan vectors or medical data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a medical teaching plan generation method is implemented.

[0128] Those skilled in the art will understand that Figure 5 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.

[0129] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the following steps when executing the computer program: obtaining personalized configuration parameters; the personalized configuration parameters include student level parameters, teaching target parameters and lesson plan language parameters; obtaining N medical lesson plan contents matching the personalized configuration parameters from a preset database; searching for real-time lesson plan contents from at least one preset website according to keywords corresponding to the personalized configuration parameters; adding the N medical lesson plan contents, the real-time lesson plan contents and the personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes a prompt word for prompting a preset large model to perform a medical lesson plan generation task; inputting the first input text into the preset large model to obtain a target medical lesson plan output by the preset large model.

[0130] In one embodiment, when the processor executes the computer program, the following steps are also implemented: vectorizing the personalized configuration parameters to obtain a configuration vector; selecting N teaching plan vectors with the highest similarity to the configuration vector from a preset database, and obtaining N medical teaching plan contents corresponding to the N teaching plan vectors; the preset database stores teaching plan vectors obtained by vectorizing multiple medical teaching plan contents.

[0131] In one embodiment, the processor also implements the following steps when executing the computer program: performing grammar check, content check and accuracy check on the target medical lesson plan; if the target medical lesson plan fails any one of the grammar check, content check and accuracy check, adding the failed check item and the target medical lesson plan to a second prompt template to obtain a second input text; the second prompt template includes prompt words for prompting the preset large model to pay attention to the failed check item and perform the medical lesson plan regeneration task; the second input text is input into the preset large model to obtain a medical lesson plan regenerated by the preset large model.

[0132] In one embodiment, the processor also implements the following steps when executing the computer program: obtaining knowledge point keywords corresponding to the teaching objectives; performing keyword matching on the target medical teaching plan based on the knowledge point keywords to obtain keyword matching results; segmenting the target medical teaching plan to obtain multiple paragraphs; detecting the text similarity between any two paragraphs in the multiple paragraphs; determining the duplicate content detection result based on the text similarity and a set threshold; and performing content checking on the target medical teaching plan based on the keyword matching results and the duplicate content detection results.

[0133] In one embodiment, when the processor executes the computer program, the following steps are also implemented: extracting medical knowledge points from the target medical teaching plan; obtaining knowledge entities and first related information corresponding to the medical knowledge points from the medical knowledge graph; querying second related information in the medical database based on the knowledge entities; and checking the accuracy of the teaching plan content corresponding to the medical knowledge points in the target medical teaching plan based on the first related information and the second related information.

[0134] In one embodiment, when the processor executes the computer program, the following steps are also implemented: evaluating the translation score of the target medical teaching plan; if the translation score is less than the set score, adjusting the parameters of the preset large model; inputting the target medical teaching plan into the adjusted preset large model for re-translation, and obtaining a medical teaching plan regenerated by the preset large model.

[0135] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: obtaining personalized configuration parameters; the personalized configuration parameters include student level parameters, teaching target parameters and lesson plan language parameters; obtaining N medical lesson plan contents matching the personalized configuration parameters from a preset database; searching for real-time lesson plan contents from at least one preset website according to keywords corresponding to the personalized configuration parameters; adding the N medical lesson plan contents, the real-time lesson plan contents and the personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes a prompt word for prompting a preset large model to perform a medical lesson plan generation task; inputting the first input text into the preset large model to obtain a target medical lesson plan output by the preset large model.

[0136] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: vectorizing the personalized configuration parameters to obtain a configuration vector; selecting N teaching plan vectors with the highest similarity to the configuration vector from a preset database, and obtaining N medical teaching plan contents corresponding to the N teaching plan vectors; the preset database stores teaching plan vectors obtained by vectorizing multiple medical teaching plan contents.

[0137] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: performing grammar check, content check and accuracy check on the target medical lesson plan respectively; if the target medical lesson plan fails any one of the grammar check, content check and accuracy check, adding the failed check item and the target medical lesson plan to a second prompt template to obtain a second input text; the second prompt template includes prompt words for prompting the preset large model to pay attention to the failed check item and perform the medical lesson plan regeneration task; the second input text is input into the preset large model to obtain a medical lesson plan regenerated by the preset large model.

[0138] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining knowledge point keywords corresponding to the teaching objectives; performing keyword matching on the target medical teaching plan based on the knowledge point keywords to obtain keyword matching results; segmenting the target medical teaching plan to obtain multiple paragraphs; detecting the text similarity between any two paragraphs in the multiple paragraphs; determining the duplicate content detection result based on the text similarity and a set threshold; and performing content checking on the target medical teaching plan based on the keyword matching results and the duplicate content detection results.

[0139] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: extracting medical knowledge points from the target medical teaching plan; obtaining knowledge entities and first related information corresponding to the medical knowledge points from the medical knowledge graph; querying second related information in the medical database based on the knowledge entities; and checking the accuracy of the teaching plan content corresponding to the medical knowledge points in the target medical teaching plan based on the first related information and the second related information.

[0140] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: evaluating the translation score of the target medical teaching plan; if the translation score is less than the set score, adjusting the parameters of the preset large model; inputting the target medical teaching plan into the adjusted preset large model for re-translation, and obtaining a medical teaching plan regenerated by the preset large model.

[0141] In one embodiment, a computer program product is provided, including a computer program, which implements the following steps when executed by a processor: obtaining personalized configuration parameters; the personalized configuration parameters include student level parameters, teaching target parameters and lesson plan language parameters; obtaining N medical lesson plan contents matching the personalized configuration parameters from a preset database; searching for real-time lesson plan contents from at least one preset website according to keywords corresponding to the personalized configuration parameters; adding the N medical lesson plan contents, the real-time lesson plan contents and the personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes a prompt word for prompting a preset large model to perform a medical lesson plan generation task; and inputting the first input text into the preset large model to obtain a target medical lesson plan output by the preset large model.

[0142] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: vectorizing the personalized configuration parameters to obtain a configuration vector; selecting N teaching plan vectors with the highest similarity to the configuration vector from a preset database, and obtaining N medical teaching plan contents corresponding to the N teaching plan vectors; the preset database stores teaching plan vectors obtained by vectorizing multiple medical teaching plan contents.

[0143] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: performing grammar check, content check and accuracy check on the target medical lesson plan respectively; if the target medical lesson plan fails any one of the grammar check, content check and accuracy check, adding the failed check item and the target medical lesson plan to a second prompt template to obtain a second input text; the second prompt template includes prompt words for prompting the preset large model to pay attention to the failed check item and perform the medical lesson plan regeneration task; the second input text is input into the preset large model to obtain a medical lesson plan regenerated by the preset large model.

[0144] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining knowledge point keywords corresponding to the teaching objectives; performing keyword matching on the target medical teaching plan based on the knowledge point keywords to obtain keyword matching results; segmenting the target medical teaching plan to obtain multiple paragraphs; detecting the text similarity between any two paragraphs in the multiple paragraphs; determining the duplicate content detection result based on the text similarity and a set threshold; and performing content checking on the target medical teaching plan based on the keyword matching results and the duplicate content detection results.

[0145] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: extracting medical knowledge points from the target medical teaching plan; obtaining knowledge entities and first related information corresponding to the medical knowledge points from the medical knowledge graph; querying second related information in the medical database based on the knowledge entities; and checking the accuracy of the teaching plan content corresponding to the medical knowledge points in the target medical teaching plan based on the first related information and the second related information.

[0146] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: evaluating the translation score of the target medical teaching plan; if the translation score is less than the set score, adjusting the parameters of the preset large model; inputting the target medical teaching plan into the adjusted preset large model for re-translation, and obtaining a medical teaching plan regenerated by the preset large model.

[0147] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0148] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0149] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0150] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for generating a medical teaching plan, characterized in that: The method comprises: Obtaining personalized configuration parameters; the personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters; Acquire N pieces of medical teaching plan content that match the personalized configuration parameters from a preset database; Searching for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters; Adding the N pieces of medical teaching plan content, the real-time teaching plan content and the personalized configuration parameters to a first prompt template to obtain a first input text; the first prompt template includes prompt words for prompting a preset large model to perform a medical teaching plan generation task; The first input text is input into the preset large model to obtain the target medical teaching plan output by the preset large model.

2. The method according to claim 1, characterized in that The step of obtaining N pieces of medical teaching plan contents matching the personalized configuration parameters from a preset database includes: Vectorizing the personalized configuration parameters to obtain a configuration vector; Select N teaching plan vectors with the highest similarity to the configuration vector from the preset database, and obtain N medical teaching plan contents corresponding to the N teaching plan vectors; the preset database stores teaching plan vectors obtained by vectorizing multiple medical teaching plan contents.

3. The method according to claim 1, characterized in that: After inputting the first input text into the preset large model to obtain the target medical teaching plan output by the preset large model, the method further includes: Conducting grammar checking, content checking, and accuracy checking on the target medical teaching plan; If the target medical teaching plan fails any one of the grammar check, the content check and the accuracy check, the failed check item and the target medical teaching plan are added to a second prompt template to obtain a second input text; the second prompt template includes prompt words for prompting the preset large model to pay attention to the failed check item and perform the task of regenerating the medical teaching plan; The second input text is input into the preset large model to obtain a medical teaching plan regenerated by the preset large model.

4. The method according to claim 3, characterized in that Conduct content checks on the target medical teaching plans, including: Obtaining knowledge point keywords corresponding to the teaching objectives; Perform keyword matching on the target medical teaching plan according to the knowledge point keywords to obtain keyword matching results; Segmenting the target medical teaching plan to obtain a plurality of paragraphs; Detecting text similarity between any two paragraphs in the plurality of paragraphs; Determining a duplicate content detection result based on the text similarity and a set threshold; The target medical teaching plan is subjected to a content check based on the keyword matching result and the duplicate content detection result.

5. The method according to claim 3, characterized in that: The target medical teaching plan is checked for accuracy, including: extracting medical knowledge points from the target medical teaching plan; Acquire the knowledge entity and first related information corresponding to the medical knowledge point from the medical knowledge graph; querying a medical database for second relevant information according to the knowledge entity; The accuracy of the teaching plan content corresponding to the medical knowledge point in the target medical teaching plan is checked based on the first relevant information and the second relevant information.

6. The method according to any one of claims 1 to 5, characterized in that After inputting the first input text into the preset large model to obtain the target medical teaching plan output by the preset large model, the method further includes: Evaluating the translation score of the target medical teaching plan; If the translation score is less than the set score, adjusting the parameters of the preset large model; The target medical teaching plan is input into the adjusted preset large model for re-translation to obtain the medical teaching plan regenerated by the preset large model.

7. A medical teaching plan generating device, characterized in that: The device comprises: A configuration module, used to obtain personalized configuration parameters; the personalized configuration parameters include student level parameters, teaching goal parameters and teaching plan language parameters; A matching module, used for acquiring N pieces of medical teaching plan content matching the personalized configuration parameters from a preset database; A search module, used to search for real-time teaching plan content from at least one preset website according to the keywords corresponding to the personalized configuration parameters; A prompt module, used for adding the N pieces of medical teaching plan content, the real-time teaching plan content and the personalized configuration parameters into a first prompt template to obtain a first input text; the first prompt template includes a prompt word for prompting a preset large model to perform a medical teaching plan generation task; A generation module is used to input the first input text into the preset large model to obtain a target medical teaching plan output by the preset large model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images