Reasonable drug use report generation method and system combined with artificial intelligence

Abstract

The invention relates to the technical field of data processing and report generation, in particular to a reasonable drug use report generation method and system combined with artificial intelligence. The method comprises the following steps of: selecting a report type; constructing a report analysis model and analyzing report data; intensively analyzing report data by utilizing a domain knowledge graph; generating a structured report; according to feedback information of the user, an algorithm for report analysis and report generation is optimized; editing a data part and a conclusion part of the report by utilizing the two rich text editing areas; rejecting report contents which do not need to be displayed; and exporting a final editing result. In combination with an artificial intelligence technology, the method is used for automatically analyzing report data and generating a report, so that the analysis efficiency and accuracy are improved; by introducing a self-learning mechanism, a report analysis and generation algorithm is optimized, and the report generation quality is improved; by providing a friendly interface and a rich editing function, personalized requirements of different users are met, and the method has a wide application prospect in the medical field.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and report generation, and in particular to a method and system for generating a rational drug use report combining artificial intelligence. Background Art

[0002] In the information age, reports are crucial to business fluency and decision-making quality. A large number of complex reports are generated every day in the medical field, but the traditional manual generation method is time-consuming and error-prone, especially inefficient when processing large amounts of complex data. Therefore, realizing the automation and intelligent generation of reports and improving efficiency and accuracy have become problems that need to be solved urgently.

[0003] In the prior art, reports are usually generated using methods that use preset templates and fixed rules. Such methods often lack intelligent and automated processing capabilities, do not have the ability to automatically parse complex report data and understand its connotation, and cannot dynamically generate reports with clear structure and accurate content based on the specific content of the data. In addition, such methods are often not versatile enough and are difficult to flexibly adapt to the operating habits and report generation needs of different users. For example, in the medical field, different doctors or departments may need to generate reports of different formats and contents according to their own needs, and existing report generation methods often cannot meet these personalized needs. Therefore, there is an urgent need to propose a method that can automatically parse report data, generate structured reports, and has intelligent and automated capabilities.

[0004] In view of the shortcomings of the existing technology, the present invention proposes a method for generating a rational drug use report in combination with artificial intelligence. This method can automatically parse report data and generate reports by using artificial intelligence technology, thereby improving parsing efficiency and accuracy; by introducing a self-learning mechanism, the parsing and generation algorithms of the report are optimized to improve the quality of report generation; by providing a friendly interface and rich editing functions, it adapts to the personalized needs of different users and meets a wide range of needs in the medical field. Summary of the invention

[0005] In view of the defects in the prior art, the present invention provides a method and system for generating a rational medication report in combination with artificial intelligence.

[0006] In a first aspect, the present invention provides a method for generating a rational drug use report in combination with artificial intelligence, comprising: selecting a report category for a report to be generated; constructing a report parsing model based on the report category; parsing report data through the report parsing model to obtain basic parsing data; using a domain knowledge graph to perform enhanced parsing on the basic parsing data to obtain enhanced parsing data; generating a structured report based on the enhanced parsing data; based on the structured report, optimizing the report parsing and report generation algorithms according to user feedback information to obtain a structured improved report; using two rich text editing areas to edit the data part and the conclusion part of the structured improved report to obtain an editing result, the rich text editing area supports pre-editing and saving rich text content as a template, the template including a private template and a shared template; eliminating content that does not need to be displayed in the editing result to obtain a final editing result; exporting the final editing result to obtain a rational drug use report in combination with artificial intelligence. The present invention can generate high-quality structured reports by accurately parsing report data and strengthening the analysis by using domain knowledge graphs; the algorithm is continuously optimized through user feedback, and the report content is further improved to meet diverse needs; the editing convenience and personalized customization capabilities of the report are enhanced by introducing two rich text editing areas and supporting flexible configuration of private and shared templates; the final editing results are refined and accurate by eliminating content that does not need to be displayed; the rational drug use report obtained by the export function can not only effectively integrate artificial intelligence technology, but also show a high degree of readability and practicality, thereby providing support for clinical decision-making and promoting the improvement of rational drug use levels.

[0007] Optionally, the construction of a report parsing model based on the report category includes: preprocessing the report data based on the report category to obtain optimized report data; constructing a report parsing model based on the optimized report data, wherein the report parsing model includes a report intelligence model and a report deep analysis model. The present invention obtains optimized report data with clear structure and accurate content by preprocessing the report data based on the report category, thereby providing a reliable basis for subsequent analysis; by using the optimized report data to construct a report parsing model, the integrated application of the report intelligence model and the report deep analysis model is realized. The report intelligence model can accurately identify and understand the key information in the report, while the report deep analysis model can deeply explore the intrinsic connection and potential value between the data, thereby comprehensively improving the depth and breadth of report analysis.

[0008] Optionally, parsing the report data through the report parsing model to obtain basic parsing data includes: performing text parsing on the text data in the optimized report data through the report intelligence model to obtain basic text parsing data, wherein the text parsing includes semantic analysis, named entity recognition, and contextual relationship understanding; performing numerical parsing on the numerical data in the optimized report data through the report deep analysis model to obtain basic numerical parsing data, wherein the numerical parsing includes feature extraction, pattern recognition, and predictive analysis. The present invention parses the text data in the optimized report data through the report intelligence model to obtain basic text parsing data containing rich semantic information, accurate named entities, and clear contextual relationships, providing support for a comprehensive understanding of the report content; parses the numerical data in the optimized report data through the report deep analysis model, extracts key features, identifies data patterns, and performs predictive analysis of potential associations, thereby obtaining basic numerical parsing data, providing a basis for quantitative analysis and trend prediction of data.

[0009] Optionally, the report intelligent model satisfies the following expression:

[0010] in, Indicates the explanatory text of the report data. represents a natural language parsing function, Represents the original text data in the report. Represents the lexical features of the data in the report, represents the syntax tree, represents the noise reduction function, The context information for the report. is domain knowledge; the report analysis model satisfies the following expression:

[0011] in, Indicates the result of deep analysis. represents a deep learning model function, Represents the original data set in the report. represents the feature extraction function, represents the data preprocessing function, Represents unstructured data, Indicates data type. The present invention converts the original text data in the report into accurate interpretation text through the report intelligent model, making full use of the natural language parsing function to deeply analyze the vocabulary features and syntax trees, and combining context information and domain knowledge to reduce noise interference, thereby ensuring that the report content is accurately and comprehensively understood and expressed; through the report deep analysis model, key information is extracted from the original data set of the report, and a deep analysis result is constructed. This process relies on the learning ability of the deep learning model, improves data quality through feature extraction functions, and balances data preprocessing functions at the same time, combined with optimized neural network architecture and training parameters, to achieve accurate analysis and deep mining of report data.

[0012] Optionally, generating a structured report according to the enhanced analytical data includes: selecting a report template according to the enhanced analytical data; filling in the enhanced analytical data based on the report template; verifying the enhanced analytical data to obtain a verification result; and generating a structured report according to the verification result. The present invention selects a report template by enhancing analytical data to ensure that the report structure matches the data content, which not only improves the professionalism and readability of the report, but also enables the report to more accurately reflect the information and trends behind the data; by filling in the enhanced analytical data based on the selected report template, the data can be accurately located and effectively displayed; by verifying the enhanced analytical data, a reliable verification result can be obtained, providing a guarantee for generating a structured report.

[0013] Optionally, based on the structured report and according to user feedback information, optimizing the report parsing and report generation algorithms to obtain a structured improvement report includes: based on the structured report, collecting user feedback information; forming a feedback analysis report based on the feedback information; determining algorithm optimization requirements for report parsing and report generation through the feedback analysis report, and forming an algorithm optimization requirements list; designing an algorithm optimization plan based on the algorithm optimization requirements list; obtaining an optimized algorithm for report parsing and report generation using the algorithm optimization plan; and obtaining a structured improvement report through the optimization algorithm. The present invention obtains users' direct evaluation and suggestions on the effects of report parsing and report generation by collecting feedback information from users, thereby providing a basis for algorithm optimization; a feedback analysis report is formed through the feedback information, and the optimization requirements of the report parsing and report generation algorithms are determined accordingly, the specific goals and directions of the algorithm improvement are clarified, and a clear list of algorithm optimization requirements is formed, thereby improving the pertinence and efficiency of the optimization work; an optimization algorithm for report parsing and report generation is obtained by designing an algorithm optimization scheme, and the optimization algorithm, while retaining the advantages of the original algorithm, makes targeted improvements on the problems reported by users, thereby making report parsing more accurate and report generation more intelligent; and a structured improvement report generated by applying the optimization algorithm is used to improve the quality and practicality of the report.

[0014] In a second aspect, the present invention provides a rational drug use report generation system combined with artificial intelligence, comprising a report selection module, a report parsing module, a report generation module, a self-learning module, a rich text editing module, a reset and rejection module, an export module and a template management module; the report parsing module is connected to the report selection module, the self-learning module and the report generation module; the report generation module is connected to the report parsing module, the self-learning module, the rich text editing module, the reset and rejection module, the export module and the template management module; the report selection module is used to select the report category for which a report needs to be generated; the report parsing module is used to receive the report category information transmitted by the report selection module, And pass the parsed report data to the report generation module and the self-learning module; the report generation module is used to receive the parsed data, and generate a final report in combination with the rich text content edited by the user, template information, and reset instructions; the self-learning module is used to collect user feedback and operation data to improve the parsing and generation algorithms; the rich text editing module is used to pass the rich text content edited by the user to the report generation module; the reset and elimination module is used to delete specific content in the report according to user instructions; the export module is used to export the generated report as a PDF or Word document, and the PDF document can add a custom watermark; the template management module is used to provide rich text information pre-edited and saved by the user. The present invention can ensure the accuracy and standardization of medication reports, accelerate the report generation process and improve its quality by utilizing intelligent report parsing and automatic generation technology; with the help of a self-learning module, the system can continuously optimize the parsing and generation algorithms based on user feedback and actual operation data, achieve self-adjustment and improvement, and thus continuously improve the intelligence level of report generation; through comprehensive editing and exporting functions, rich text content can be freely edited, templates can be flexibly selected, and specific parts of the report can be reset or deleted instantly, meeting the diverse needs of users and enhancing the practicality of the report and the convenience of operation.

[0015] Optionally, the report parsing module includes a data preprocessing unit, a feature extraction unit and a semantic understanding unit; the data preprocessing unit is used to clean and format the report data; the feature extraction unit uses deep learning technology to extract key information in the report; the semantic understanding unit uses natural language processing technology to parse the logical relationship and contextual meaning between data. The present invention can efficiently clean and format the report data through the data preprocessing unit, ensure the accuracy and consistency of the data, lay the foundation for subsequent feature extraction and semantic understanding, and improve the efficiency and accuracy of the entire report parsing process; through the feature extraction unit, it can accurately extract key information in the report, and the key information is crucial for rational drug use decisions; through the semantic understanding unit, it can deeply understand the logical relationship and contextual meaning between data, so as to accurately parse the complex information in the report, avoid decision-making errors caused by misunderstanding or omission of information, provide more reliable and intelligent report parsing services for rational drug use, and enhance the practicality and intelligence level of the system.

[0016] Optionally, the self-learning module adopts a reinforcement learning algorithm to achieve optimization through the following steps: Monitor and record the user's feedback behavior during use, including editing, content deletion, and report regeneration requests; analyze the feedback behavior, identify the key factors affecting report quality and user experience, and obtain analysis results; adjust the deep learning model parameters and natural language processing strategies according to the analysis results. The present invention can continuously monitor and record various feedback behaviors of users during use, including editing, content deletion, and report regeneration requests, thereby accumulating rich user interaction data and providing a basis for subsequent analysis and optimization; through in-depth analysis of these feedback behaviors, it can accurately identify the key factors affecting report quality and user experience, such as missing information, logical errors, or irregular formats, and obtain corresponding analysis results, providing direction and basis for optimizing report generation algorithms; through the analysis results, it can automatically adjust the deep learning model parameters and natural language processing strategies, and continuously optimize the report parsing and generation algorithms, thereby improving the accuracy and readability of the report, meeting the growing personalized needs of users, and improving the intelligence level of the system and user satisfaction.

[0017] Optionally, the template management module includes a version control unit and a permission management unit; the version control unit allows the user to track and compare different versions of templates; the permission management unit sets access and editing permissions for private templates and shared templates according to user roles. The present invention uses a version control unit to easily track and compare different versions of templates, effectively manage the iteration process of templates, ensure the accuracy and consistency of template content, and also facilitate backtracking of historical versions, quickly restore or draw on previous designs, improve work efficiency and flexibility of template management; through the permission management unit, strict access and editing permissions are set according to user roles to protect the security and privacy of private templates, prevent unauthorized access and modification, and promote the rational use of shared templates. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of a method for generating a rational medication report in combination with artificial intelligence according to an embodiment of the present invention; Figure 2 The present invention is a schematic diagram of the structure of a rational medication report generation system combined with artificial intelligence according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0020] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0021] See also Figure 1 The embodiment of the present invention provides a method for generating a rational medication report in combination with artificial intelligence, the method comprising the following steps: S1. Select the report category to be generated.

[0022] In one embodiment, the report category for the report to be generated is selected by following the steps below: S11. Clarify the purpose and requirements of the report.

[0023] In this embodiment, first, the purpose of the report is clarified, whether it is used for internal decision-making, external reporting or other specific needs.

[0024] Furthermore, understand who the audience of the report is and what data and information they will be interested in.

[0025] S12. Evaluate report category suitability.

[0026] In this embodiment, the existing rational drug use report forms are reviewed to understand their categories, structures and contents.

[0027] Furthermore, the report category to be generated is determined based on the report purpose and audience needs.

[0028] S13. Select the specific report template and format.

[0029] In this embodiment, a standard report template in the industry is referenced.

[0030] Furthermore, the format and layout of the report can be customized according to specific needs to ensure that the report is easy to read and understand while containing all the necessary information.

[0031] S14. Verify and optimize report categories.

[0032] In this embodiment, the selected report is reviewed to ensure the accuracy and completeness of the report.

[0033] Furthermore, collect feedback from the audience to understand their satisfaction with the report and their suggestions for improvement.

[0034] Further, the report categories are optimized and adjusted based on the feedback.

[0035] S2. Based on the report category, construct a report parsing model.

[0036] Among them, S2 includes the following steps: S21. Based on the report category, pre-process the report data to obtain optimized report data.

[0037] In one embodiment, based on the report category selected in step S1, the report data is preprocessed, and the preprocessing process is as follows: S211. Clarify preprocessing objectives.

[0038] In this embodiment, data quality requirements are clarified according to the specific requirements of the report category, including data integrity, accuracy, consistency, and timeliness.

[0039] Furthermore, according to the data quality requirements, identify the data preprocessing tasks that need to be performed, such as missing value processing, outlier processing, and data format conversion.

[0040] S212. Data cleaning.

[0041] In this embodiment, the missing values ​​of the data are processed.

[0042] Specifically, identify and count the location and number of missing values; select an appropriate filling strategy based on business logic or statistical methods, such as using the mean, median, mode, or model-based predicted values ​​to fill missing values; for missing values ​​that cannot be reasonably filled, consider deleting related records or performing other processing.

[0043] Furthermore, outliers in the data are processed.

[0044] Specifically, statistical methods or business logic are used to identify outliers; based on actual conditions, outliers are corrected, deleted, or treated as special cases.

[0045] S213.Data conversion and formatting.

[0046] In this embodiment, the data is converted and formatted.

[0047] Specifically, ensure that the data types of all fields are consistent with the report category requirements, such as converting string types to numeric types, or converting date types to a unified format; standardize the formats of date, time, and currency fields to ensure data consistency and readability; for categorical variables, consider using one-hot encoding or label encoding methods for encoding.

[0048] S214. Data integration and consolidation.

[0049] In this embodiment, multiple data sources are merged, wherein duplicate records are removed when merging the data sources to ensure the uniqueness of the data.

[0050] Furthermore, the merged data is checked for consistency to ensure that there are no conflicts or contradictions between the data from different data sources.

[0051] S215. Data verification and validation.

[0052] In this embodiment, the data is checked for completeness, including whether all necessary fields are filled.

[0053] Furthermore, the accuracy of the data can be verified by comparing it with historical data, business logic or external data sources.

[0054] Furthermore, check whether the data between different fields are consistent, such as whether the relationship between the drug prescription amount and the drug prescription quantity is reasonable.

[0055] S216. Generate optimization report data.

[0056] In this embodiment, the pre-processed data is exported into a format suitable for report generation or data analysis, such as CSV, Excel or database file.

[0057] Further, according to the requirements of the report category, a report generation tool or software is used to generate the final report.

[0058] Furthermore, the generated reports are reviewed to ensure that the data is accurate, the format is standardized, and the information is complete.

[0059] S22. Based on the optimized report data, a report parsing model is constructed.

[0060] In one embodiment, based on the optimized report data, a report parsing model is constructed using a basic large model of mathematical expression, such as a Zero reasoning model. The report parsing model includes a report intelligence model and a report deep analysis model.

[0061] Specifically, the report intelligent model satisfies the following expression:

[0062] in, Indicates the explanatory text of the report data. represents a natural language parsing function, Represents the original text data in the report. Represents the lexical features of the data in the report, represents the syntax tree, represents the noise reduction function, which is used to reduce the noise, and its result is One of the inputs of the function, generating the explanatory text of the report data , The context information for the report. is domain knowledge; the report analysis model satisfies the following expression:

[0063] in, Indicates the result of deep analysis. represents a deep learning model function, Represents the original data set in the report. represents the feature extraction function, represents the data preprocessing function, Represents unstructured data, Indicates a data type. Represents the report dataset Preprocessing is performed, and the preprocessed data is extracted by the feature extraction function Processing, finally the feature extraction results, unstructured data and data type or task type Passed as input to the deep learning model function To obtain in-depth analysis results . Such a logical process is more in line with the conventional process of data processing and machine learning models. The intelligent report model can receive the original text data in the report, use the natural language parsing function, and combine the vocabulary features and syntax trees to generate an explanatory text for the report data. At the same time, through the noise reduction function, combined with the contextual information and domain knowledge of the report, the accuracy and relevance of the explanatory text are ensured. The deep analysis model of the report focuses on in-depth analysis of the original data set in the report. The expressiveness of data features is improved through feature enhancement factors, and a deep learning model is used to combine data preprocessing functions, unstructured data and training parameters to output a structured data set. This data set can provide clearer and more valuable data analysis results, assisting decision makers in making more accurate decisions.

[0064] In another embodiment, a local version of the large language model of Tongyi Qianwen is first deployed, and the report structure and content are fine-tuned based on the optimized report data.

[0065] Furthermore, by designing different prompt word projects, we construct the report intelligent speech model and the report deep analysis model respectively.

[0066] S3. Analyze the report data through the report analysis model to obtain basic analysis data.

[0067] Among them, S3 includes the following steps: S31. Through the report intelligence model, text data in the optimized report data is parsed to obtain basic text parsing data.

[0068] In one embodiment, the report intelligent model constructed in step S2 utilizes natural language technology to parse the text data in the optimized report data to obtain basic text parsing data, wherein the text parsing includes semantic analysis, named entity recognition, and contextual relationship understanding.

[0069] Specifically, the semantic analysis includes the following steps: First, build a vocabulary database that contains the definition, synonyms, and antonyms of each word; Furthermore, word embedding algorithms, such as Word2Vec, are used to map words into a low-dimensional vector space and calculate the similarity between them to capture the association between words; Furthermore, syntactic analysis is performed to break down sentences into different components, such as subject, predicate, and object, to achieve understanding of sentence structure; Furthermore, semantic role labeling is performed to associate each word in the sentence with its semantic role, wherein the semantic role includes agent, patient, time, and place; Furthermore, semantic relationship extraction is performed to extract the relationships and connections between different entities from the text.

[0070] The semantic analysis model includes a word embedding algorithm, a syntactic analyzer, a semantic role tagger, and a relation extraction algorithm. The objective function of the word embedding algorithm is to minimize the following loss function: , in, represents the loss function, Represents the total number of words in the text, represents the size of the context window, Indicates the current word. Represents words in context, Indicates that given the current word When the context word Probability of occurrence.

[0071] The named entity recognition comprises the following steps: First, determine the boundaries of entities in the text, that is, word segmentation. It should be noted that in Chinese text, since there is no space as a separator between words, word segmentation is a key step in named entity recognition.

[0072] Furthermore, a single entity is identified and the category to which the entity belongs is determined, such as a person's name, a place name, or an organization name.

[0073] Furthermore, composite entities, i.e. entities composed of multiple words, are identified.

[0074] Furthermore, commonly used named entity recognition models include conditional random fields, support vector machines, hidden Markov models, and deep learning models such as recurrent neural networks, long short-term memory networks, and transformer models. The objective function of the conditional random field is usually to maximize the following log-likelihood function: , in, is the state characteristic function, is the transfer characteristic function, is the input characteristic function, represents the input sequence, represents the output sequence, represents the length of the sequence, represents the score of the forward pass, is the transition score, indicating the transition from state Transfer to state The probability of is the emission fraction, indicating that in a given state Next, observe probability.

[0075] The context understanding comprises the following steps: First, analyze the contextual information in the text to understand the specific meaning of the entity and the context in which it is located; Furthermore, context models and attention mechanism methods are used to achieve context understanding to provide more accurate and complete semantic analysis results.

[0076] Among them, the contextual relationship understanding model includes a context model and an attention mechanism.

[0077] The context vector is usually used to represent the context information of a certain position or entity in the text. It is obtained by weighted summing of the words in the context or using a more complex neural network structure. The expression of the context vector is as follows: , in, Indicates location The context vector of Indicates location Words for location The weight of Indicates location The word embedding vector of the word, Represents a text function.

[0078] Attention weight is used to measure the relevance or importance between different parts of the text. It is obtained by calculating the similarity score between the query vector and the context vector. Its expression is as follows:

[0079] in, Indicates location The query vector With location The word vector The attention weights between is a similarity calculation function. Represents a text function.

[0080] Contextual relation extraction refers to identifying and extracting the relationships between entities or associations between events from text. This usually involves further processing and analysis of the context vector.

[0081] For entity relationship extraction, a relationship classifier can be defined , which receives the context vectors of two entities as input and outputs the category of the relationship between them.

[0082]

[0083] in, Represents the relationship between two entities. represents the context vector of the entity, A context vector representing the entity.

[0084] S32. Perform numerical analysis on the numerical data in the optimized report data through the report deep analysis model to obtain basic numerical analysis data, wherein the numerical analysis includes feature extraction, pattern recognition and predictive analysis.

[0085] In one embodiment, the report deep analysis model constructed in step S2 utilizes deep learning technology to perform numerical analysis on the numerical data in the optimized report data to obtain basic numerical analysis data, wherein the numerical analysis includes feature extraction, pattern recognition, and predictive analysis.

[0086] Specifically, feature extraction is a key step in deep learning, which aims to extract useful information or features from raw data to facilitate subsequent analysis and prediction. In numerical data, feature extraction may involve data preprocessing, transformation or dimensionality reduction.

[0087] Suppose we have a numerical dataset ,in, It is The feature vector of data points. Feature extraction can be expressed as:

[0088] in, represents the extracted feature vector, Represents a feature extraction function, which may be a linear transformation, nonlinear transformation, principal component analysis (PCA), autoencoder or other complex deep learning models. In linear transformation, the following conditions are met:

[0089] in, Represents the weight matrix, which determines the importance of each original feature. Represents the bias vector, which is used to adjust the baseline of the output features.

[0090] Pattern recognition is another important step in deep learning, which aims to identify potential patterns or regularities from the extracted features. It usually involves classification, clustering or regression tasks.

[0091] Assume that we have extracted features , pattern recognition can be expressed as:

[0092] in, Represents data points The predicted label or category of Represents a pattern recognition function, which may be a classifier, clusterer, or regression model.

[0093] Predictive analysis involves using a trained model to make predictions about new data, usually by passing the new data through a feature extraction and pattern recognition phase.

[0094] For new data points , the predictive analysis can be expressed as:

[0095] in, represents the predicted probability value, represents the conversion function, which is used to convert the output into a probability distribution. , represents the bias vector, , represents the weight matrix, Represents a nonlinear activation function, which is used to introduce nonlinearity so that the model can learn complex feature representations; Used to adjust the baseline of extracted features, The following conditions must be met:

[0096] in, is the number of categories (for classification tasks) or the number of output variables (for regression tasks); The element determines the influence of each dimension in the original feature vector on the extracted features. The following conditions must be met:

[0097] in, It should be noted that for regression tasks, we can remove Function, use linear output directly or nonlinear transformation.

[0098] S4. Use the domain knowledge graph to perform enhanced analysis on the basic analysis data to obtain enhanced analysis data.

[0099] In one embodiment, first, domain knowledge related to rational drug use is collected and organized, including drug information, disease information, and related medical terms and concepts.

[0100] Furthermore, this knowledge is used to construct a domain knowledge graph, in which nodes represent entities and edges represent the relationships between entities.

[0101] Furthermore, the domain knowledge graph is used to enhance the analysis of basic analytical data.

[0102] Specifically, through matching and reasoning, key information that may be missing in the report can be identified and supplemented, such as automatically supplementing the dosage and usage information of the drug according to its name. At the same time, the relationship information in the graph is used to analyze the logical relationship and contextual meaning between the data, thereby improving the accuracy and depth of the analysis.

[0103] Furthermore, enhanced analytical data containing richer and more accurate information is obtained.

[0104] S5. Generate a structured report based on the enhanced parsed data.

[0105] In one embodiment, first, clarify the problem that the report aims to solve, the information it provides, or the decision it supports; understand who the main readers of the report are, their information needs, interests, and background knowledge.

[0106] Furthermore, the analysis results are sorted out and strengthened, including reviewing the text analysis results, summarizing the numerical analysis results and integrating the information.

[0107] Specifically, the review of text parsing results includes key findings of semantic analysis, named entity recognition and contextual relationship understanding; the summary of numerical parsing results includes the main conclusions of feature extraction, pattern recognition and predictive analysis; the integrated information includes combining the results of text parsing and numerical parsing to form a coherent and consistent information system.

[0108] Further, design the report structure, including determining the report framework, dividing chapters, and designing charts and visualizations.

[0109] Specifically, determining the report framework includes: designing the overall framework of the report according to the report objectives and audience needs; the overall framework includes the introduction, main text, conclusion and recommendations; dividing the chapters includes: dividing the report content into several chapters or sections, each chapter or section focusing on a specific topic or issue; designing charts and visualizations includes: selecting appropriate charts, graphics and visualization tools to display key data and findings, and enhance the readability and attractiveness of the report.

[0110] Furthermore, structured reports are automatically generated.

[0111] S6. Based on the structured report and according to user feedback, the report parsing and report generation algorithms are optimized to obtain a structured improved report.

[0112] Wherein, S6 further comprises the following steps: S61. Collect user feedback information based on the structured report.

[0113] In one embodiment, based on the structured report, multiple feedback collection channels are designed through online questionnaires, mail surveys, telephone interviews or face-to-face meetings to ensure that users can provide feedback conveniently.

[0114] Furthermore, in the feedback collection channel, clearly list the content that users may need to provide feedback on, such as the accuracy, readability, completeness, practicality, formatting of the report, and the user's specific suggestions or opinions.

[0115] Furthermore, a fixed feedback collection cycle is set, such as monthly or quarterly, to organize and analyze the collected user feedback to form a preliminary user feedback summary.

[0116] S62. Form a feedback analysis report based on the feedback information.

[0117] In one embodiment, user feedback is classified and summarized according to the dimensions of content, importance, and frequency of occurrence to form a preliminary framework of a feedback analysis report.

[0118] Furthermore, for quantifiable feedback, such as the accuracy and readability scores of the reports, quantitative evaluations are conducted to form specific quantitative indicators.

[0119] Furthermore, based on the results of classification and quantitative evaluation, a feedback analysis report is written to elaborate on the overall situation, main issues and suggestions of user feedback.

[0120] S63. Determine the optimization requirements of the report parsing and report generation algorithms through the feedback analysis report, and form an optimization requirements list.

[0121] In one embodiment, based on the feedback analysis report, key points and specific problems that need to be optimized in the report parsing and report generation algorithms are identified.

[0122] Furthermore, based on the identified optimization needs, clear optimization goals are set, such as improving the accuracy, readability, and generation efficiency of the report.

[0123] Furthermore, the optimization objectives and corresponding optimization requirements are organized into an algorithm optimization requirements list to provide a basis for the subsequent optimization solution design.

[0124] S64. Design an algorithm optimization solution based on the optimization requirements list.

[0125] In one embodiment, the optimization methods and successful cases of related algorithms in the industry are investigated, and their experiences and lessons are learned to provide reference for the design of algorithm optimization solutions.

[0126] Furthermore, based on the optimization requirements list and survey results, specific algorithm optimization strategies are designed, such as improving data preprocessing methods, optimizing feature extraction algorithms, and adjusting report generation templates.

[0127] Furthermore, a detailed implementation plan is formulated for the algorithm optimization solution, including optimization steps, time nodes, and personnel division of labor.

[0128] S65. Utilize the algorithm optimization solution to obtain an optimization algorithm for report parsing and report generation.

[0129] In one embodiment, according to the algorithm optimization solution, the original report parsing and report generation algorithm is coded and optimized.

[0130] Furthermore, the optimized algorithm is fully tested and verified to ensure that it can meet the optimization goals and requirements while ensuring the stability and reliability of the algorithm.

[0131] Furthermore, based on the test and verification results, the algorithm is iteratively optimized until the best effect is achieved.

[0132] S66. Obtain a structured improvement report through the optimization algorithm.

[0133] In one embodiment, a new structured and refined report is generated using an optimized report parsing and report generation algorithm.

[0134] Furthermore, the structured refinement report is provided to some users for evaluation to collect their feedback and suggestions.

[0135] Furthermore, the algorithm is continuously improved and optimized based on user evaluation feedback to ensure that the structured refinement report can continue to meet user needs and expectations.

[0136] S7. Use two rich text editing areas to edit the data part and the conclusion part of the structured improvement report to obtain the editing results.

[0137] Wherein, S7 further comprises the following steps: S71. Define two rich text editing areas.

[0138] In one embodiment, two rich text editing areas are set up, which are used to edit the data part and the conclusion part of the structured improvement report respectively. The rich text editing area supports common rich text formats, such as bold, italic, underline, color, font size, paragraph alignment, list, and picture insertion; it also supports pre-editing and saving rich text content as templates, and the templates include private templates and shared templates; the private template pre-edits and saves specific rich text content as a private template according to the personalized needs of the user. For example, a user can create a template containing specific titles, subtitles, paragraph formats, etc. according to his or her own report style. The shared template provides all users with a set of standard, reusable rich text templates, covering common report structures and format requirements, such as title pages, directories, main text, conclusions, and references.

[0139] S72. Use the two rich text editing areas to edit the data part and the conclusion part of the structured improvement report to obtain the editing result.

[0140] In one embodiment, when a user needs to edit a structured refinement report, he or she first enters the rich text editing area; the system provides a template selection interface, allowing the user to select a suitable private template or shared template from the template library as the editing starting point.

[0141] Furthermore, in the rich text editing area of ​​the data section, users can enter or paste the data content of the report according to the structure and format requirements of the selected template; users can use the rich text editing function to format and layout the data content to ensure the clarity and readability of the data.

[0142] Furthermore, in the rich text editing area of ​​the conclusion part, the user can enter or paste the corresponding conclusion content according to the data analysis results and conclusions of the report. Similarly, the user can use the rich text editing function to format and layout the conclusion content to highlight key information and conclusions.

[0143] Furthermore, after completing the editing, the user should save the editing results. The system should provide a save button or shortcut key to facilitate the user to save the edited rich text content to the database. At the same time, the system should support the user to save the currently edited rich text content as a new private template or shared template for subsequent reuse.

[0144] S8. Eliminate the content that does not need to be displayed in the editing result to obtain the final editing result.

[0145] In one embodiment, the editing result is opened, and the contents are examined item by item to determine whether each content needs to be retained. If a content does not meet the elimination criteria, that is, it needs to be retained, it is marked as "retained"; if a content meets the elimination criteria, that is, it does not need to be displayed, it is marked as "removed".

[0146] Furthermore, based on the review results, perform a removal operation. For content marked as "removed", you can use the delete function of the rich text editor to remove it from the editing results; when performing the removal operation, you should ensure that you do not accidentally delete content that needs to be retained, and try to keep the format and structure of the editing results unchanged.

[0147] Furthermore, after the elimination operation is completed, the editing results are reviewed and verified. Make sure that all content that does not need to be displayed has been correctly eliminated and that no content that needs to be retained has been mistakenly deleted. At the same time, check whether the format and structure of the editing results are still clear, consistent and easy to understand.

[0148] Further, according to the results of the review and verification, the editing results are finally sorted and formatted to obtain the final editing results. It should be noted that this step needs to ensure that all retained content is arranged in a logical order and format requirements.

[0149] S9. Export the final editing result to obtain a rational drug use report combined with artificial intelligence.

[0150] In one embodiment, the rich text editor or document processing software where the final editing result is located is first opened.

[0151] Further, select the "Save As" or "Export" option from the "File" menu.

[0152] Further, in the pop-up dialog box, select Word document (.docx or .doc) as the save type. Specify the save location and click the "Save" button.

[0153] In another embodiment, the software where the final editing result is located is opened first.

[0154] Next, select the "Print" option in the "File" menu. It should be noted that the "Print" here actually refers to exporting to PDF, not actual printing.

[0155] Further, select the installed PDF converter or printer driver in the printer list.

[0156] Further, click on the "Properties" or "Settings" button to adjust the PDF export options as needed.

[0157] Further, click "OK" to return to the print dialog box, and then click the "Print" or "Export to PDF" button.

[0158] Further, specify the save location and file name and click the “Save” button.

[0159] In yet another embodiment, a PDF document is first opened using a PDF editor.

[0160] Further, select the "Tools" or "Edit" menu in the software and find the "Watermark" or "Header and Footer" option.

[0161] Further, click the "Add" or "Create" button to enter the watermark editing interface.

[0162] Further, in the watermark editing interface, select "text" or "image" as the watermark type, and enter or select the watermark content.

[0163] Furthermore, you can adjust the watermark's location, size, transparency, rotation angle and other properties to meet your needs. If necessary, you can set the watermark to apply to specific pages of the document or the entire document.

[0164] See also Figure 2 , Figure 2 The present invention is a schematic diagram of the structure of a rational drug use report generation system combined with artificial intelligence in an embodiment of the present invention. The system includes an input module, a processing module, an output module, and a storage module, wherein the input module, the processing module, the output module, and the storage module are interconnected, wherein the storage module is used to store a computer program, the computer program includes program instructions, the processing module is configured to call the program instructions, and the system uses the method for generating a rational drug use report combined with artificial intelligence.

[0165] The input module includes a report selection module, a rich text editing module, a reset and rejection module, and a template management module; the processing module includes a report parsing module and a self-learning module; the output module includes a report generation module and an export module: the storage module has the characteristics of fast reading and writing speed, large capacity and high reliability, and is mainly used to store data input by the input module and the result data processed by the processing module, and can meet the needs of large data storage.

[0166] The report parsing module is connected to the report selection module, the self-learning module and the report generation module; the report generation module is connected to the report parsing module, the self-learning module, the rich text editing module, the reset and elimination module, the export module and the template management module.

[0167] The report selection module is used to select the report category for which a report needs to be generated.

[0168] The report parsing module is used to receive the report category information transmitted by the report selection module, and transmit the parsed report data to the report generation module and the self-learning module.

[0169] The report generation module is used to receive the parsed data and generate a final report in combination with the rich text content, template information, and reset instructions edited by the user.

[0170] The self-learning module is used to collect user feedback and operation data to improve the parsing and generation algorithms.

[0171] The rich text editing module is used to transfer the rich text content edited by the user to the report generating module.

[0172] The reset and elimination module is used to delete specific content in the report according to user instructions.

[0173] The export module is used to export the generated report into PDF or Word format, and can add a custom watermark to the PDF.

[0174] The template management module is used to provide template information edited and saved by the user for reference by the report generation module when generating a report.

[0175] The report parsing module includes a data preprocessing unit, a feature extraction unit and a semantic understanding unit; the data preprocessing unit is used to clean and format the report data; the feature extraction unit uses deep learning technology to extract key information in the report; the semantic understanding unit uses natural language processing technology to parse the logical relationship and contextual meaning between data.

[0176] The self-learning module adopts a reinforcement learning algorithm and achieves optimization through the following steps: monitoring and recording the user's feedback behavior during use, including editing, modification, content deletion and report regeneration requests; analyzing the feedback behavior, identifying the key factors affecting report quality and user experience, and obtaining analysis results; and adjusting the deep learning model parameters and natural language processing strategies based on the analysis results.

[0177] The template management module includes a version control unit and a permission management unit; the version control unit allows users to track and compare different versions of templates; the permission management unit sets access and editing permissions for private templates and shared templates according to user roles.

[0178] In summary, the present invention can effectively solve the problems of low efficiency and error-proneness of traditional report generation methods, insufficient intelligence and automation capabilities of existing systems, and poor versatility, and the inability to flexibly meet the needs of large data volumes, complex classifications, and diversified report generation in the medical field. By utilizing artificial intelligence technology, the report can be automatically parsed and generated to improve the efficiency and accuracy of report generation and reduce the error rate; through continuous optimization through a self-learning mechanism, the system can continuously learn and evolve to adapt to the personalized needs of different users; by providing a human-computer interaction-friendly interface and equipped with flexible editing and export functions, the report generation needs in different scenarios can be met; through the template management function, commonly used report templates can be preset to improve work efficiency, while supporting both private and shared modes to enhance flexibility of use.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for generating a rational drug use report combining artificial intelligence, characterized in that: The method comprises the following steps: Select the report category to be generated; Based on the report category, construct a report parsing model; Parsing report data through the report parsing model to obtain basic parsing data; Using the domain knowledge graph, the basic parsing data is subjected to enhanced parsing to obtain enhanced parsing data; generating a structured report based on the enhanced parsed data; Based on the structured report, according to user feedback information, the report parsing and report generation algorithms are optimized to obtain a structured and refined report; Using two rich text editing areas, edit the data part and the conclusion part of the structured improvement report to obtain the editing result, the rich text editing area supports pre-editing and saving rich text content as a template, and the template includes a private template and a shared template; Eliminate the content that does not need to be displayed in the editing result to obtain the final editing result; The final editing result is exported to obtain a rational medication report combined with artificial intelligence.

2. According to claim 1, a method for generating a rational drug use report combining artificial intelligence is characterized in that: The constructing of a report parsing model based on the report category includes: Based on the report category, preprocess the report data to obtain optimized report data; Based on the optimized report data, a report parsing model is constructed, and the report parsing model includes a report intelligence model and a report deep analysis model.

3. According to claim 1, a method for generating a rational drug use report combining artificial intelligence is characterized in that: The parsing of report data by the report parsing model to obtain basic parsing data includes: Through the report intelligent model, text data in the optimized report data is parsed to obtain basic text parsing data, wherein the text parsing includes semantic analysis, named entity recognition, and contextual relationship understanding; Through the report deep analysis model, the numerical data in the optimized report data is numerically analyzed to obtain basic numerical analysis data, and the numerical analysis includes feature extraction, pattern recognition and predictive analysis.

4. The method for generating a rational drug use report combining artificial intelligence according to claim 3, characterized in that: The report intelligent model satisfies the following expression: , in, Indicates the explanatory text of the report data. represents a natural language parsing function, Represents the original text data in the report. Represents the lexical features of the data in the report, represents the syntax tree, represents the noise reduction function, The context information for the report. is domain knowledge; the report analysis model satisfies the following expression: , in, Indicates the result of deep analysis. represents a deep learning model function, Represents the original data set in the report. represents the feature extraction function, represents the data preprocessing function, Represents unstructured data, Indicates a data type.

5. The method for generating a rational drug use report combining artificial intelligence according to claim 1, characterized in that: Generating a structured report according to the enhanced parsed data includes: Selecting a report template based on the enhanced analytical data; Filling the enhanced analytical data based on the report template; Verifying the enhanced parsing data to obtain a verification result; A structured report is generated based on the verification results.

6. The method for generating a rational drug use report combining artificial intelligence according to claim 1, characterized in that: Based on the structured report, according to the user's feedback information, the report parsing and report generation algorithms are optimized to obtain a structured improved report including: Based on the structured report, collecting user feedback information; Forming a feedback analysis report based on the feedback information; Determine algorithm optimization requirements for report parsing and report generation through the feedback analysis report, and form an algorithm optimization requirements list; Design an algorithm optimization solution based on the algorithm optimization requirements list; Using the algorithm optimization solution, an optimization algorithm for report parsing and report generation is obtained; Through the optimization algorithm, a structured improvement report is obtained.

7. A system for generating a rational drug use report in combination with artificial intelligence, the system using a method for generating a rational drug use report in combination with artificial intelligence as claimed in any one of claims 1 to 6, characterized in that: The system comprises a report selection module, a report parsing module, a report generation module, a self-learning module, a rich text editing module, a reset and rejection module, an export module and a template management module; the report parsing module is connected to the report selection module, the self-learning module and the report generation module; the report generation module is connected to the report parsing module, the self-learning module, the rich text editing module, the reset and rejection module, the export module and the template management module; The report selection module is used to select the report category for which a report needs to be generated; The report parsing module is used to receive the report category information transmitted by the report selection module, and transmit the parsed report data to the report generation module and the self-learning module; The report generation module is used to receive the parsed data and generate a final report in combination with the rich text content, template information, and reset instructions edited by the user; The self-learning module is used to collect user feedback and operation data to improve the parsing and generation algorithms; The rich text editing module is used to pass the rich text content edited by the user to the report generating module; The reset and elimination module is used to delete specific content in the report according to user instructions; The export module is used to export the generated report into a PDF or Word document, and the PDF document can be added with a custom watermark; The template management module is used to provide rich text information pre-edited and saved by the user.

8. The system for generating a rational drug use report combining artificial intelligence according to claim 7 is characterized in that: The report parsing module includes a data preprocessing unit, a feature extraction unit and a semantic understanding unit; The data preprocessing unit is used to clean and format report data; The feature extraction unit uses deep learning technology to extract key information from the report; The semantic understanding unit uses natural language processing technology to analyze the logical relationship and contextual meaning between data.

9. The system for generating a rational drug use report combining artificial intelligence according to claim 7, characterized in that: The self-learning module uses a reinforcement learning algorithm to achieve optimization through the following steps: Monitor and record user feedback during use, including editorial changes, content deletions, and report regeneration requests; Analyze the feedback behavior, identify key factors affecting report quality and user experience, and obtain analysis results; According to the analysis results, the deep learning model parameters and natural language processing strategies are adjusted.

10. The system for generating rational drug use report combining artificial intelligence according to claim 7, characterized in that: The template management module includes a version control unit and a rights management unit; The version control unit allows users to track and compare different versions of templates; The authority management unit sets access and editing authority to private templates and shared templates according to user roles.

Images