Medical data traceability management system and method based on big data analysis

Abstract

The invention relates to the related technical field of data traceability management, in particular to a medicine data traceability management system and method based on big data analysis, and the method comprises the steps: an information recognition module extracts the basic medicine information of Chinese medicine classics, and a deep mining module constructs a medicine information atlas; the traceability analysis module establishes a traditional Chinese medicine information tree through multi-node traceability; the data analysis module is combined with the modern medical theory to construct a complementary knowledge network; the traceability management module optimizes the network, forms a medical knowledge framework, realizes structural traceability management of medical data, solves the technical problem that most of traditional Chinese medicine knowledge is summarized by experience and lacks systematic management, realizes multi-node traceability analysis of the traditional Chinese medicine knowledge, and combines modern medical theories to perform mutual fusion and complementation, so that the traceability management of the traditional Chinese medicine knowledge is realized. The scientificity and applicability of the medicine data are enhanced, a medicine knowledge framework is formed, and the technical effect of integral structured traceability management of the medicine data is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field related to data traceability management, and in particular to a medical data traceability management system and method based on big data analysis. Background Art

[0002] As people's awareness of health management and disease prevention increases, the combination of traditional medicine and modern medicine has gradually become an important direction of medical research. As an important part of Chinese traditional culture, Chinese medicine has accumulated rich theoretical and practical experience. However, traditional Chinese medicine classics are obscure, and Chinese medicine knowledge is mostly a summary of experience, lacking systematic management.

[0003] In summary, although traditional Chinese medicine has unique advantages in theory and application, its knowledge system still has problems such as fragmentation and incomplete information, making it difficult to fully play its role. Most existing medical information systems focus on modern medicine and lack effective integration and utilization of traditional Chinese medicine knowledge. In addition, medical education and research models often separate the two, creating obstacles to information exchange.

[0004] In summary, the existing technology has the technical problem that most of the knowledge of traditional Chinese medicine is summarized from experience and lacks systematic management. Summary of the invention

[0005] This application provides a medical data traceability management system and method based on big data analysis, aiming to solve the technical problem that traditional Chinese medicine knowledge in the prior art is mostly empirical summary and lacks systematic management.

[0006] In view of the above problems, the technical solution to implement this application is: On the one hand, the present application provides a medical data traceability management system based on big data analysis, wherein the medical data traceability management system is connected to a TCM classics digitization module, and the TCM classics digitization module uses text recognition and natural language processing technology to recognize and convert text information in TCM classics into an editable digital format. The medical data traceability management system includes: The information identification module is used to identify the basic medical information in the TCM classics through the TCM classics digitization module, and the basic medical information includes the drug name, nature, taste, meridian, and efficacy classification; the deep mining module is used to use the basic medical information as the root node, deeply mine and construct a medical information map, and the medical information map includes a disease case node, a prescription composition node, and a usage and dosage node; the traceability analysis module is used to use the disease case node of the medical information map as the first search stem, the prescription composition node of the medical information map as the second search stem, and the usage and dosage node of the medical information map as the third search stem to perform traceability analysis and establish a TCM information tree , wherein the traceability orientation includes the interaction relationship between drugs, the trend of drug efficacy changes, and potential action clues. The first search stem bract, the second search stem bract, and the third search stem bract all belong to the starting nodes in the medical information map, and the first search stem bract and the first search branch, the second search stem bract and the second search branch, and the third search stem bract and the third search branch have a unidirectional association; a data analysis module is used to use a data analysis engine to combine modern medical theories to establish a complementary knowledge network; a traceability management module is used to optimize the complementary knowledge network through the traditional Chinese medicine information tree to form a medical knowledge framework and perform overall structured traceability management of medical data.

[0007] On the other hand, the present application provides a medical data traceability management method based on big data analysis, wherein the medical data traceability management system is connected to a TCM classics digitization module, and the TCM classics digitization module uses text recognition and natural language processing technology to recognize and convert text information in TCM classics into an editable digital format, and the method includes: Through the TCM classics digitization module, the basic medical information in TCM classics is identified, and the basic medical information includes drug name, nature, flavor, meridian, and efficacy classification; the basic medical information is used as the root node to deeply mine and construct a medical information map, and the medical information map includes disease case nodes, prescription composition nodes, and usage and dosage nodes; the disease case nodes of the medical information map are used as the first search stem, the prescription composition nodes of the medical information map are used as the second search stem, and the usage and dosage nodes of the medical information map are used as the third search stem to conduct traceability analysis and establish a TCM information tree, wherein, The traceability orientation includes the interaction relationship between drugs, the trend of drug efficacy changes, and potential action clues. The first search stem bale, the second search stem bale, and the third search stem bale all belong to the starting nodes in the medical information map, and the first search stem bale and the first search branch, the second search stem bale and the second search branch, and the third search stem bale and the third search branch have a unidirectional association; using the data analysis engine, combined with modern medical theory, a complementary knowledge network is established; based on the complementary knowledge network, optimization is performed through the traditional Chinese medicine information tree to form a medical knowledge framework, and overall structured traceability management of medical data is performed.

[0008] In summary, one or more technical solutions provided in this application solve the technical problem that most of TCM knowledge is a summary of experience and lacks systematic management, realizes multi-node traceability analysis of TCM knowledge, and combines and complements modern medical theories, enhances the scientificity and applicability of medical data, constitutes a medical knowledge framework, and achieves the technical effect of holistic structured traceability management of medical data. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A structural diagram of a medical data traceability management system based on big data analysis is provided for this application.

[0010] Figure 2 A flow chart of a medical data traceability management method based on big data analysis is provided for this application.

[0011] Explanation of the reference numerals: information identification module M100, deep mining module M200, traceability analysis module M300, data analysis module M400, traceability management module M500. DETAILED DESCRIPTION

[0012] Embodiment 1 The present application is described in detail below in conjunction with the accompanying drawings. Figure 1As shown, the present application provides a medical data traceability management system based on big data analysis, wherein the medical data traceability management system is connected to a TCM classics digitization module, and the TCM classics digitization module uses text recognition and natural language processing technology to recognize and convert text information in TCM classics into an editable digital format. The medical data traceability management system includes: The information identification module M100 is used to identify the basic medical information in the TCM classics through the TCM classics digitization module. The basic medical information includes the drug name, nature and taste, meridians, and efficacy classification; the deep mining module M200 is used to use the basic medical information as the root node to deeply mine and construct a medical information map. The medical information map includes disease case nodes, prescription composition nodes, and usage and dosage nodes.

[0013] Specifically, the TCM classics digitization module is a system component that uses text recognition and natural language processing technologies to convert information in traditional Chinese medicine books into digital format for further analysis and utilization; basic medical information refers to basic data extracted from TCM classics, including drug names (such as "ginseng"), nature and flavors, meridians (such as "sweet, warm, spleen and lung meridians"), and efficacy classification (such as "tonifying qi, strengthening the spleen"); the medical information map is used to display the relationship between different medical information, including disease case nodes, prescription composition nodes, and usage and dosage nodes.

[0014] Execution steps: Through the TCM classics digitization module, apply OCR (optical character recognition) and NLP (natural language processing) technologies to scan and analyze traditional TCM classics. For example, identify "ginseng" in the text and extract its nature, flavor, meridians and efficacy classification information, which involves algorithm analysis of the structure and semantics of the text in the book; the extracted basic medical information will be used as the root node, for example, the identified "ginseng" will be used as a root node, and its nature, flavor and efficacy information will be further attached, and organized into structured data for subsequent processing and query.

[0015] A medical information map is constructed using the extracted basic medical information. The nodes of the medical information map are not limited to basic information, but also include: disease case nodes (such as "fatigue caused by spleen deficiency"), prescription composition nodes (such as "prescriptions for the combination of ginseng and astragalus"), and usage and dosage nodes (such as "3 grams per day, decocted in water"). Through a relational database or a graph database, these nodes are linked through different relationships, for example, linking "ginseng" with its related disease cases and prescription compositions.

[0016] For example, the basic medical information identified is "Huangqi": the root node is "Huangqi", its nature and flavor are "sweet, warm, spleen and lung meridians", and its efficacy is classified as "replenishing qi and enhancing immunity". In the constructed medical information map, the following nodes can be created: disease case node: associated with "cold" or "fatigue"; prescription composition node: such as "Shenling Baizhu San"; usage and dosage node: such as "10 grams per day, decocted in water". It not only helps users quickly find relevant information, but also supports complex queries, such as looking for "all qi-replenishing drugs" or "prescriptions related to Huangqi". Through the above steps, it is possible to effectively construct and manage information related to traditional Chinese medicine, providing a basis for subsequent in-depth analysis and application.

[0017] The traceability analysis module M300 is used to perform traceability analysis on the disease case node of the medical information map as the first search stem, the prescription composition node of the medical information map as the second search stem, and the usage and dosage node of the medical information map as the third search stem, so as to establish a traditional Chinese medicine information tree, wherein the traceability orientation includes the interaction relationship between drugs, the trend of drug efficacy changes, and potential action clues, and the first search stem, the second search stem, and the third search stem all belong to the starting node in the medical information map, and the first search stem and the first search branch, the second search stem and the second search branch, and the third search stem and the third search branch have a unidirectional association; the data analysis module M400 is used to use the data analysis engine and combine modern medical theories to establish a complementary knowledge network; the traceability management module M500 is used to optimize the traditional Chinese medicine information tree based on the complementary knowledge network to form a medical knowledge framework and perform overall structured traceability management of medical data.

[0018] Specifically, the first search stem bud, the second search stem bud, and the third search stem bud are the starting nodes in the medical information map, and are the main information sources for traceability analysis; traceability analysis refers to tracing the relationship between drugs, symptoms, and treatment plans, and analyzing the causes and impacts behind them; the complementary knowledge network combines traditional medical knowledge with modern medical theories to form a rich information network for data analysis and knowledge acquisition; the medical knowledge framework is a comprehensive medical data traceability model that covers both traditional Chinese medicine and modern medicine, and is used to achieve comprehensive and structured management of medical knowledge.

[0019] Execution steps: Set the disease case node (such as "spleen deficiency") in the medical information map as the first search stem; set the prescription composition node (such as "ginseng, astragalus") as the second search stem; set the usage and dosage node (such as "3 grams per day, decocted in water") as the third search stem. Starting from the first search stem, analyze its relationship with the second and third search stems, explore drug interactions (for example, the synergistic effect of ginseng and astragalus); evaluate the trend of drug efficacy changes, observe the improvement of symptoms after using ginseng (such as the physical recovery of patients with spleen deficiency); find potential clues, such as the adaptability or side effects of drugs to specific diseases. Based on the traceability analysis results of the search stem, establish a traditional Chinese medicine information tree, associate different nodes (diseases, prescriptions, usage), for example, connect "spleen deficiency" to "ginseng, astragalus" and their corresponding usage and clinical cases, and form a hierarchical information structure.

[0020] Data analysis engines (such as machine learning algorithms) are used in combination with modern medical theories (such as physiology and pathology) to process data. Algorithms are used to mine potential patterns in the data, for example, to analyze the differences in the effects of different drugs in treating spleen deficiency.

[0021] Combine knowledge extracted from traditional Chinese medicine and modern medicine to form a complementary knowledge network. For example, link "spleen deficiency" with modern medicine's understanding of its causes (such as diet and lifestyle influences); optimize the TCM information tree based on the complementary knowledge network and construct a medical knowledge framework; apply the medical knowledge framework to medical data management, and promote the modernization and digitization of TCM knowledge by combining it with modern medical theories, and realize holistic structured traceability management.

[0022] Furthermore, the disease case node of the medical information map is used as the first search stem for source tracing analysis, and the source tracing analysis module M300 is used to perform the following method: Based on the symptom case node as a starting point, an association analysis index is obtained, and the association analysis index includes the cause of the symptom, the progression of the symptom, and the symptom manifestation; through the cause of the symptom, the progression of the symptom, and the symptom manifestation in the association analysis index, a first tracing space is established; in the first tracing space, the symptom case node is used as a first search stem bud for tracing analysis to generate a first search branch.

[0023] Specifically, association analysis indicators are used to describe and analyze the relationship between symptoms and their causes, development processes and manifestations, mainly including the causes of symptoms, the progression of symptoms and the manifestations of symptoms; the first traceability space refers to the analysis framework established based on specific nodes (such as symptom case nodes) in data analysis, which is used to deeply explore the various correlations of the node; the first search branch is a series of related information and data branches generated from the first search stem in the traceability analysis, which helps to track and understand the various characteristics of the disease.

[0024] Execution steps: Set the disease case node, for example, select a disease (such as "hypertension") as the first search stem; then, through the medical information map, identify the basic information related to "hypertension", such as the cause of the disease (such as genetics, improper diet), the course of the disease (such as stage changes), and the symptoms of the disease (such as headache, dizziness); further, perform data mining on "hypertension" and extract association analysis indicators. For example, through historical case analysis, it is found that a patient's blood pressure rises due to a high-salt diet, which can be used as an example of the cause of the disease; use the extracted association analysis indicators to create a first traceability space. In the above example, the first traceability space defines the development stage of "hypertension" (such as mild, moderate, severe) and its manifestations (such as physical examination data, patient subjective feelings).

[0025] In the first traceability space, use "hypertension" as the first search stem to conduct source analysis; analyze its relationship with factors such as diet, exercise, and genetics through association indicators to generate the first search branch. For example, a direct association between "high-salt diet" and "hypertension" is found. Further, starting from the first search stem, multiple search branches related to "hypertension" are established, each branch represents a symptom or cause, for example, branch A (high-salt diet), branch B (genetic factors), and branch C (lack of exercise) are generated; conduct in-depth analysis of each search branch, and record the corresponding patient cases and research data. For example, branch A may record multiple patient data on hypertension caused by a high-salt diet for subsequent research. Through the above steps, the system can better understand the causes, development, and manifestations of the disease.

[0026] Furthermore, in the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch. The source tracing analysis module M300 is also used to execute the following method: In the first tracing space, fitness evaluation is performed through the drug interaction relationship in the tracing guide, and a first enclosed area is established, and the pharmacokinetic parameters of the first enclosed area match the pharmacokinetic parameters of the disease case node; in the first tracing space, fitness evaluation is performed through the drug efficacy change trend in the tracing guide, and a second enclosed area is established, and the pharmacodynamic parameters of the second enclosed area match the pharmacodynamic parameters of the disease case node; in the first tracing space, fitness evaluation is performed through the potential action clues in the tracing guide, and a third enclosed area is established, and the potential action targets of the third enclosed area match the potential action targets of the disease case node; the disease case node is used as the first search stem, and an iterative search is performed through the first enclosed area, the second enclosed area, and the third enclosed area to generate a first search branch.

[0027] Specifically, fitness evaluation is used to judge the effectiveness or suitability of a variable under specific conditions; the first enclosing area, the second enclosing area, and the third enclosing area refer to the range determined by fitness evaluation in data analysis, which are used to define possible valid data or decision-making areas; iterative search: the process of repeatedly searching to find a better solution by continuously adjusting and updating parameters.

[0028] Execution steps: set the disease case node, for example, select a specific disease (such as "diabetes") as the first search stem, extract relevant information from the medical information map; determine the key factors and characteristics of "diabetes" to form a first traceability space for in-depth analysis, and further, evaluate the fitness of drugs (such as insulin and oral hypoglycemic drugs) used by "diabetes" patients by analyzing their interactions, for example, determine whether the combined use of drug A (insulin) and drug B (oral hypoglycemic drugs) can effectively lower blood sugar levels; in the first traceability space, perform fitness evaluation based on the drug interaction relationship in the traceability guide to delineate the first enclosed area, which includes all effective drug combinations and their interactions.

[0029] The same execution steps as the first enclosed area, further, in the first traceability space, fitness evaluation is performed through the trend of drug efficacy changes in the traceability guide, and a second enclosed area is established. The pharmacodynamic parameters of the second enclosed area match the pharmacodynamic parameters of the disease case node. Exemplarily, the changes in the efficacy of drugs on "diabetes" patients in different time periods are analyzed, for example, the fluctuation trend of blood sugar in patients after using a certain drug is monitored, and its long-term effect is evaluated; according to the change in drug efficacy, a second enclosed area is established, and the second enclosed area includes the change pattern of drug efficacy over time; other factors that may affect "diabetes" (such as living habits, diet, and psychological state) are analyzed, and their potential effects are evaluated, and a third enclosed area is established based on this, which may include new treatments or interventions.

[0030] The disease case node (such as "diabetes") is used as the first search stem, and an iterative search is performed through three enclosing areas. In the first enclosing area, the fitness of different drug combinations is evaluated and the best combination is screened out; in the second enclosing area, the drug effects are monitored and the best time to use the drugs is found; in the third enclosing area, potential action clues are combined; based on the results of the iterative search, the first search branch is generated to record all valid combinations and related information, for example, determining the best plan for the combined use of insulin and oral hypoglycemic drugs, and recording the patient's responses and effects at different stages of treatment.

[0031] Furthermore, the disease case node is used as the first search stem, and an iterative search is performed through the first enclosed area, the second enclosed area, and the third enclosed area to generate a first search branch. The traceability analysis module M300 is also used to execute the following method: In the first enclosed area and multiple fitness neighborhoods, point to the previous low fitness neighborhood and the next high fitness neighborhood, bidirectionally traverse to continuously update the fitness, and establish a fitness value ranking; select the points that are in the top 3 in the fitness value ranking and are not on the same straight line as the leading solution, and perform iterative search; use the disease case node as the first search stem, compare it with the real-time updated leading solution, and fit the first search branch population.

[0032] Specifically, fitness value sorting is to sort the effectiveness of each search point according to the fitness value in order to find the best solution.

[0033] The specific execution steps are as follows: setting the disease case node, for example, selecting a specific disease (such as "hypertension") as the first search stem, extracting relevant data for subsequent analysis; in the first enclosing area and multiple fitness neighborhoods, using relevant data (such as treatment effect, drug interaction, etc.), pointing to the previous low fitness neighborhood and the next high fitness neighborhood, bidirectionally traversing and continuously updating the fitness of each search point, for example, by comparing the efficacy of different drugs (such as the antihypertensive effect of drug A and the effect of drug B), to update their respective fitness values; sorting the fitness values ​​of all candidate solutions, for example, if the fitness of drugs A, B, and C are 0.8, 0.9, and 0.85, respectively, the generated ranking is B (0.9), C (0.85), and A (0.8).

[0034] Select the top 3 points from the fitness value ranking as the leading solutions, and ensure that these points are not on the same straight line (i.e. not in the same direction), which means selecting solutions with different characteristics or mechanisms of action, such as selecting drugs A, B, and C, to ensure that the leading solutions have different mechanisms of action; use the disease case node (such as "hypertension") as the first search stem, and use the leading solution updated in real time for iterative search; by comparing the leading solution, adjust and optimize the search branch, for example, if the effect of drug A is better than other drugs in a specific situation, the weight of the drug in the treatment plan can be increased. Through the above steps, the system can effectively use data-driven methods to continuously optimize and adjust the treatment plan.

[0035] Furthermore, the points that are in the top 3 of the fitness value ranking and are not on the same straight line are selected as leading solutions for iterative search. The traceability analysis module M300 is also used to execute the following method: If the points in the top 3 positions in the fitness value ranking are on the same straight line, it is defined as the first fitness line, and the point in the 4th position in the fitness value ranking is determined; it is judged whether the point in the 4th position in the fitness value ranking is on the first fitness line, if not, the points in the top 4 positions in the fitness value ranking and not on the same straight line are selected as leading solutions, and an iterative search is performed; at the same time, a second search branch group is fitted in the second enclosing area, and a third search branch group is fitted in the third enclosing area; the disease case node is used as the first search stem, and is incorporated into the first search branch group, the second search branch group, and the third search branch group to generate the first search branch.

[0036] Specifically, when the selected fitness solutions are on the same straight line, the straight line is defined as the first fitness straight line, indicating that these solutions may be redundant in the feature space.

[0037] Execution steps: Select the top 3 solutions (such as drugs A, B, C) from the fitness value ranking as the leading solutions, and ensure that the leading solutions are not on the same line in the feature space to avoid selecting redundant solutions; but inevitably, if the selected drugs A, B, C are indeed on the same line, mark them as the "first fitness line" and check the 4th drug (such as drug D); determine whether drug D is above the first fitness line; if not, select the top 4 drugs (A, B, C, D) as the new leading solutions for iterative search; if so, select the top 5 drugs (A, B, C, D) as the new leading solutions for iterative search, and so on.

[0038] In the second enclosed area, drug D and its features are used for search to generate the second search branch; in the third enclosed area, the third search branch is generated through the features of drug C; the disease case node (such as "hypertension") is used as the first search stem, and then merged into the first search branch group, the second search branch group, and the third search branch group to form a complete first search branch. Through the above steps, the fitness value is efficiently used to select and optimize the solution, providing a basis for personalized treatment plans. The iterative search process ensures that the model is continuously updated and improved, providing support for the precise traceability management of medical data.

[0039] Furthermore, in the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch. The source tracing analysis module M300 is also used to execute the following method: In the second tracing space, the prescription composition node is used as the second search stem to perform traceability analysis to generate a second search branch; in the third tracing space, the usage and dosage node is used as the third search stem to perform traceability analysis to generate a third search branch; the traditional Chinese medicine information tree is established through the first search stem and the first search branch, the second search stem and the second search branch, and the third search stem and the third search branch.

[0040] Specifically, the disease case node refers to the instance related to a specific disease (such as hypertension, diabetes, etc.), which is the basic data node for traceability analysis; the first traceability space is used to analyze the space for specific disease cases, which contains relevant data and information to support the traceability analysis; the prescription composition node refers to the drugs contained in the traditional Chinese medicine prescription and their combination information, which serves as the second data node for traceability analysis; the second traceability space is used to analyze the prescription composition space, which contains relevant drug and formula information; the usage and dosage node is used to describe the usage and recommended dosage of the prescription; the third traceability space is used to analyze the usage and dosage space; the traditional Chinese medicine information tree is a structured information system formed by organizing different nodes through association relationships.

[0041] Execution steps: In the first traceability space, select the disease case node (such as "hypertension") as the first search stem; conduct traceability analysis, generate the first search branch based on the identified association indicators (such as cause, clinical manifestations), and further explore drugs, therapies and cases related to "hypertension". In the second traceability space, use a specific prescription component node (such as "Tian Ma Gou Teng Yin") as the second search stem; conduct traceability analysis on the components of the prescription to find out its efficacy, indications, usage, etc. for the disease, and generate the second search branch, which may include similar prescriptions and their indications. In the third traceability space, also use the prescription component node (such as "Liuwei Dihuang Wan") as the third search stem; analyze the relationship between the components of the prescription and the corresponding disease, and generate the third search branch, which may contain information such as its therapeutic effect and interaction.

[0042] The first searched stem bud ("hypertension") is associated with the first search branch (drugs and therapies related to hypertension, etc.); similarly, the second searched stem bud ("Gastrodia Uncaria Decoction") is associated with the second search branch (the efficacy and indications of its constituent drugs) and the third searched stem bud is associated with its search branch. Through these associations, a complete TCM information tree is constructed to form a structured information system to facilitate subsequent query and analysis. Through the above steps, a dynamically updated TCM information tree is formed to integrate various medical information and provide data support for subsequent research and clinical applications. This process not only optimizes the efficiency of information retrieval, but also provides new perspectives and methods for the research of TCM.

[0043] Furthermore, based on the complementary knowledge network, the Chinese medicine information tree is optimized to form a medicine knowledge framework, and the traceability management module M500 is used to execute the following method: Based on the modern medical theory, a modern medical information tree is established in comparison with the traditional Chinese medicine information tree; the traditional Chinese medicine information tree and the modern medical information tree are used as the two sides of the medical knowledge framework respectively, and the complementary knowledge network is used as the connecting layer. The machine learning model is used to perform deep learning optimization to establish the medical knowledge framework.

[0044] Specifically, a complementary knowledge network refers to a knowledge system that integrates medical knowledge from different sources, especially the mutual complementation and collaboration between traditional Chinese medicine and modern medicine; the traditional Chinese medicine information tree is a structured traditional Chinese medicine knowledge system, which contains nodes of various symptoms, prescriptions and their relationships; the modern medicine information tree is a medical knowledge system constructed based on modern medical theory, which usually covers information such as disease mechanisms, treatment plans and drug effects; the medical knowledge framework is a comprehensive medical knowledge system formed by integrating the traditional Chinese medicine information tree and the modern medicine information tree; the machine learning model is a computational model that uses algorithms to learn from data and make predictions or classifications, and is used to optimize the medical knowledge framework in the embodiments of the present application.

[0045] Execution steps: According to the existing modern medical theories (such as biomedicine, pharmacology, etc.), extract relevant data from the traditional Chinese medicine information tree one by one, such as the physiological mechanism of the disease, treatment methods, etc., to form a modern medicine information tree; for example, extract relevant biomarkers, modern treatment methods (such as ACE inhibitors) and the latest research results from the "hypertension" node.

[0046] With the TCM information tree as one side of the framework and the modern medicine information tree as the other side, the structures and information of the two trees complement each other, providing medical knowledge from different perspectives. By constructing a graphical representation, the connection between the two is clearly shown, such as the different mechanisms of action of the traditional Chinese medicine "Gastrodia elata" on hypertension in traditional and modern medicine.

[0047] A complementary knowledge network is established between the TCM information tree and the modern medicine information tree. The complementary knowledge network connects the TCM information tree and the modern medicine information tree to facilitate the flow and integration of information. For example, by analyzing the drug interactions and efficacy comparisons between the two, a more comprehensive knowledge network is formed. Further, a machine learning algorithm (such as a decision tree, random forest or neural network) is used to perform deep learning optimization on the data of the connection layer. When training the model, the data of the TCM information tree and the modern medicine information tree are input to optimize the structure and content of the medical knowledge framework. For example, the knowledge points related to "hypertension" are optimized through historical cases and treatment results. After training and optimization of the machine learning model, a medical knowledge framework is obtained, which integrates the knowledge of TCM and modern medicine, and facilitates the combination of TCM treatment and modern drugs. Through the above steps, a more comprehensive and accurate medical knowledge framework is established to achieve effective integration and management of TCM knowledge and improve the utilization efficiency of medical data.

[0048] In summary, the beneficial effects of the embodiments of the present application are: 1. Through big data analysis, traditional Chinese medicine information is combined with modern medical theories to promote the modernization and digitization of Chinese medicine knowledge and improve the utilization efficiency of medical data.

[0049] 2. Construct a medical information map to achieve effective integration and management of traditional Chinese medicine knowledge, thus greatly improving the visualization of medical information.

[0050] 3. Based on modern medical theory, a modern medical information tree is established in comparison with the TCM information tree; the TCM information tree and the modern medical information tree are used as the two sides of the medical knowledge framework, and the complementary knowledge network is used as the connection layer. The machine learning model is used for deep learning optimization to establish the medical knowledge framework. Through the above steps, a more comprehensive and accurate medical knowledge framework is established, the effective integration and management of TCM knowledge is achieved, and the utilization efficiency of medical data is improved.

[0051] Embodiment 2 Based on the same inventive concept as the medical data traceability management system based on big data analysis in the aforementioned embodiment, Figure 2 As shown, the embodiment of the present application provides a medical data traceability management method based on big data analysis, wherein the medical data traceability management system is connected to a TCM classics digitization module, and the TCM classics digitization module uses text recognition and natural language processing technology to recognize and convert text information in TCM classics into an editable digital format. The method includes: S1: Identify basic medical information in TCM classics through the TCM classics digitization module, the basic medical information includes drug name, nature, flavor, meridian, and efficacy classification; S2: Taking the basic medical information as the root node, deeply mining and constructing a medical information map, wherein the medical information map includes a disease case node, a prescription composition node, and a usage and dosage node; S3: Use the disease case node of the medical information map as the first search stem, the prescription composition node of the medical information map as the second search stem, and the usage and dosage node of the medical information map as the third search stem to perform traceability analysis, and establish a traditional Chinese medicine information tree, wherein the traceability orientation includes the interaction relationship between drugs, the trend of drug efficacy changes, and potential action clues. The first search stem, the second search stem, and the third search stem all belong to the starting node in the medical information map, and the first search stem has a one-way association with the first search branch, the second search stem has a one-way association with the second search branch, and the third search stem has a one-way association with the third search branch; S4: Use data analysis engines and combine them with modern medical theories to establish a complementary knowledge network; S5: Based on the complementary knowledge network, the TCM information tree is optimized to form a medical knowledge framework, and the overall structured traceability management of medical data is performed.

[0052] Furthermore, the disease case node of the medical information map is used as the first search stem for source tracing analysis. The method of the present application includes: Based on the symptom case node as a starting point, obtaining association analysis indicators, the association analysis indicators include symptom cause, symptom progression, and symptom manifestation; Establishing a first tracing space through the cause of the disease, the course of the disease, and the manifestation of the disease in the association analysis indicators; In the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch.

[0053] Furthermore, in the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch. The method of the present application includes: In the first tracing space, fitness evaluation is performed through the drug-drug interaction relationship in the traceability guide to establish a first enclosed area, and the pharmacokinetic parameters of the first enclosed area match the pharmacokinetic parameters of the disease case node; In the first tracing space, fitness evaluation is performed through the drug efficacy change trend in the traceability guide, and a second enclosed area is established, and the pharmacodynamic parameters of the second enclosed area match the pharmacodynamic parameters of the disease case node; In the first tracing space, fitness evaluation is performed through potential action clues in the traceability guidance to establish a third enclosed area, and the potential action targets of the third enclosed area are matched with the potential action targets of the disease case node; The disease case node is used as the first search stem, and an iterative search is performed through the first enclosed area, the second enclosed area, and the third enclosed area to generate a first search branch.

[0054] Further, the disease case node is used as the first search stem, and an iterative search is performed through the first enclosed area, the second enclosed area, and the third enclosed area to generate a first search branch. The method of the present application includes: In the first enclosed area and multiple fitness neighborhoods, point to the previous low fitness neighborhood and the next high fitness neighborhood, traverse bidirectionally and continuously update the fitness, and establish a fitness value ranking; Selecting the points that are in the top three of the fitness value ranking and are not on the same straight line as leading solutions, and performing iterative search; The disease case node is used as the first search stem, and the first search branch population is fitted by comparing it with the leading solution updated in real time.

[0055] Furthermore, the points that are in the top three of the fitness value ranking and are not on the same straight line are selected as leading solutions, and an iterative search is performed. The method of the present application also includes: If the first three points in the fitness value ranking are on the same straight line, it is defined as the first fitness straight line, and the point in the fourth place in the fitness value ranking is determined; Determine whether the point ranked 4th in the fitness value ranking is on the first fitness line. If not, select the points ranked in the top 4 in the fitness value ranking and not on the same line as the leading solution and perform iterative search. At the same time, fitting a second search branch group in the second enclosed area and fitting a third search branch group in the third enclosed area; The disease case node is used as the first search stem, and is incorporated into the first search branch group, the second search branch group, and the third search branch group to generate the first search branch.

[0056] Furthermore, in the first tracing space, the disease case node is used as a first search stem to perform source tracing analysis to generate a first search branch. The method of the present application also includes: In the second tracing space, the prescription composition node is used as the second search stem to perform source tracing analysis to generate a second search branch; In the third tracing space, the usage and amount node is used as the third search stem to perform source tracing analysis to generate a third search branch; The traditional Chinese medicine information tree is established through the first search stem bud and the first search branch, the second search stem bud and the second search branch, and the third search stem bud and the third search branch.

[0057] Furthermore, based on the complementary knowledge network, the TCM information tree is optimized to form a medical knowledge framework. The method of the present application includes: Based on the modern medical theory and in comparison with the traditional Chinese medicine information tree, a modern medicine information tree is established; The traditional Chinese medicine information tree and the modern medicine information tree are used as the two sides of the medical knowledge framework respectively, and the complementary knowledge network is used as the connection layer. The machine learning model is used to perform deep learning optimization to establish the medical knowledge framework.

[0058] In summary, any step can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor, and no unnecessary restrictions are made here.

[0059] Furthermore, the above-mentioned technical scheme only reflects the preferred technical scheme of the technical scheme of the embodiment of the present application. Some changes that may be made to some parts thereof by technicians in this technical field all reflect the novel principles of the embodiment of the present application. Obviously, technicians in this field can make various changes and modifications to the present application without departing from the scope of the present application.

Claims

1. A medical data traceability management system based on big data analysis, characterized in that: The medical data traceability management system is connected to a TCM classics digitization module, which uses text recognition and natural language processing technology to identify and convert text information in TCM classics into an editable digital format. The medical data traceability management system includes: An information identification module is used to identify basic medical information in traditional Chinese medicine classics through a traditional Chinese medicine classics digitization module, wherein the basic medical information includes drug name, nature, flavor, meridian, and efficacy classification; A deep mining module, used to use the basic medical information as a root node, deeply mine and construct a medical information map, wherein the medical information map includes a disease case node, a prescription composition node, and a usage and dosage node; A traceability analysis module is used to perform traceability analysis using the disease case node of the medical information map as the first search stem, the prescription composition node of the medical information map as the second search stem, and the usage and dosage node of the medical information map as the third search stem, to establish a traditional Chinese medicine information tree, wherein the traceability orientation includes the interaction relationship between drugs, the trend of drug efficacy changes, and potential action clues, and the first search stem, the second search stem, and the third search stem are all starting nodes in the medical information map, and the first search stem and the first search branch, the second search stem and the second search branch, and the third search stem and the third search branch have a one-way association; Data analysis module, which is used to build a complementary knowledge network by using data analysis engine and combining modern medical theory; The traceability management module is used to optimize the TCM information tree based on the complementary knowledge network, form a medical knowledge framework, and perform overall structured traceability management of medical data.

2. The medical data traceability management system based on big data analysis according to claim 1, characterized in that: The disease case node of the medical information map is used as the first search stem to perform source tracing analysis, including: Based on the symptom case node as a starting point, obtaining association analysis indicators, the association analysis indicators include symptom cause, symptom progression, and symptom manifestation; Establishing a first tracing space through the cause of the disease, the course of the disease, and the manifestation of the disease in the association analysis indicators; In the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch.

3. The medical data traceability management system based on big data analysis as claimed in claim 2, characterized in that: In the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch, including: In the first tracing space, fitness evaluation is performed through the drug-drug interaction relationship in the traceability guide to establish a first enclosed area, and the pharmacokinetic parameters of the first enclosed area match the pharmacokinetic parameters of the disease case node; In the first tracing space, fitness evaluation is performed through the drug efficacy change trend in the traceability guide, and a second enclosed area is established, and the pharmacodynamic parameters of the second enclosed area match the pharmacodynamic parameters of the disease case node; In the first tracing space, fitness evaluation is performed through potential action clues in the traceability guidance to establish a third enclosed area, and the potential action targets of the third enclosed area are matched with the potential action targets of the disease case node; The disease case node is used as the first search stem, and an iterative search is performed through the first enclosed area, the second enclosed area, and the third enclosed area to generate a first search branch.

4. The medical data traceability management system based on big data analysis as claimed in claim 3, characterized in that: The disease case node is used as the first search stem, and an iterative search is performed through the first enclosed area, the second enclosed area, and the third enclosed area to generate a first search branch, including: In the first enclosed area and multiple fitness neighborhoods, point to the previous low fitness neighborhood and the next high fitness neighborhood, traverse bidirectionally and continuously update the fitness, and establish a fitness value ranking; Selecting the points that are in the top three of the fitness value ranking and are not on the same straight line as leading solutions, and performing iterative search; The disease case node is used as the first search stem, and the first search branch population is fitted by comparing it with the leading solution updated in real time.

5. The medical data traceability management system based on big data analysis as claimed in claim 4, characterized in that: Selecting the points that are in the top three of the fitness value ranking and are not on the same straight line as the leading solution, and performing iterative search, further comprising: If the first three points in the fitness value ranking are on the same straight line, it is defined as the first fitness straight line, and the point in the fourth place in the fitness value ranking is determined; Determine whether the point ranked 4th in the fitness value ranking is on the first fitness line. If not, select the points ranked in the top 4 in the fitness value ranking and not on the same line as the leading solution and perform iterative search. At the same time, fitting a second search branch group in the second enclosed area and fitting a third search branch group in the third enclosed area; The disease case node is used as the first search stem, and is incorporated into the first search branch group, the second search branch group, and the third search branch group to generate the first search branch.

6. The medical data traceability management system based on big data analysis according to claim 5, characterized in that: In the first tracing space, the disease case node is used as the first search stem to perform source tracing analysis to generate a first search branch, and further includes: In the second tracing space, the prescription composition node is used as the second search stem to perform source tracing analysis to generate a second search branch; In the third tracing space, the usage and amount node is used as the third search stem to perform source tracing analysis to generate a third search branch; The traditional Chinese medicine information tree is established through the first search stem bud and the first search branch, the second search stem bud and the second search branch, and the third search stem bud and the third search branch.

7. The medical data traceability management system based on big data analysis according to claim 1, characterized in that: Based on the complementary knowledge network, the Chinese medicine information tree is optimized to form a medical knowledge framework, including: Based on the modern medical theory and in comparison with the traditional Chinese medicine information tree, a modern medicine information tree is established; The traditional Chinese medicine information tree and the modern medicine information tree are used as the two sides of the medical knowledge framework respectively, and the complementary knowledge network is used as the connection layer. The machine learning model is used to perform deep learning optimization to establish the medical knowledge framework.

8. A medical data traceability management method based on big data analysis, characterized in that: A method for implementing the medical data traceability management system based on big data analysis as described in any one of claims 1 to 7, wherein the medical data traceability management system is connected to a TCM classics digitization module, and the TCM classics digitization module uses text recognition and natural language processing technology to recognize and convert text information in TCM classics into an editable digital format, and the method comprises: Through the TCM classics digitization module, basic medical information in TCM classics is identified, and the basic medical information includes drug names, properties, flavors, meridians, and efficacy classifications; Taking the basic medical information as the root node, deeply mining and constructing a medical information map, the medical information map includes a disease case node, a prescription composition node, and a usage and dosage node; The disease case node of the medical information map is used as the first search stem, the prescription composition node of the medical information map is used as the second search stem, and the usage and dosage node of the medical information map is used as the third search stem to perform traceability analysis, and establish a traditional Chinese medicine information tree, wherein the traceability orientation includes the interaction relationship between drugs, the trend of drug efficacy changes, and potential action clues. The first search stem, the second search stem, and the third search stem are all starting nodes in the medical information map, and the first search stem and the first search branch, the second search stem and the second search branch, and the third search stem and the third search branch have a one-way association; Using data analysis engines and combining modern medical theories to build a complementary knowledge network; Based on the complementary knowledge network, the TCM information tree is optimized to form a medical knowledge framework, and the overall structured traceability management of medical data is carried out.

Images