Knowledge mining method and system for trauma nursing medical field

Abstract

The invention relates to the technical field of medical care data mining, in particular to a knowledge mining method and system for the field of trauma nursing medicine. The method comprises the following steps: acquiring original medical data of a wound; performing data preprocessing on the trauma original medical data to obtain trauma medical data; standardizing the trauma medical data to obtain standard trauma medical data; acquiring nursing monitoring data of the trauma patient; and performing monitoring type division on the nursing monitoring data of the traumatic patient to obtain nursing monitoring type data. According to the invention, a data processing technology, a mode identification technology and a deep learning technology are adopted; trauma nursing medical data fusion is carried out on trauma medical text information, image information and nursing monitoring information, and relevance knowledge hierarchical division is carried out on trauma nursing medical fusion data; therefore, the knowledge system in the field of trauma nursing medicine is presented more comprehensively.

Description

Technical Field

[0001] The present invention relates to the technical field of healthcare data mining, and particularly to a knowledge mining method and system for the field of trauma care medicine. Background Art

[0002] Initial knowledge mining mainly relied on traditional statistical methods and expert systems to extract useful knowledge by analyzing historical data and expert experience. With the application of big data technology in the field of trauma care medicine, researchers began to use massive medical data for knowledge mining, specifically using big data technology to process complex and multi-dimensional trauma care medicine data sets. The current trauma care knowledge mining methods highly rely on data, especially large-scale and high-quality medical data sets, which include patients' medical record information, physiological monitoring data, imaging materials, etc. By analyzing these data, representative feature information in the field of trauma care medicine is extracted. However, the existing knowledge mining technologies are difficult to fuse trauma care medical data from trauma medical text information, image information, and nursing monitoring information, and it is difficult to divide the associated knowledge levels of the fused trauma care medical data, resulting in a low presentation degree of the knowledge system in the field of trauma care medicine. Summary of the Invention

[0003] Based on this, it is necessary to provide a knowledge mining method and system for the field of trauma care medicine to solve at least one of the above technical problems.

[0004] To achieve the above object, a knowledge mining method for the field of trauma care medicine, the method includes the following steps: Step S1: Obtain original trauma medical data; perform data preprocessing on the original trauma medical data to obtain trauma medical data; perform standardization processing on the trauma medical data to obtain standard trauma medical data; Step S2: Obtain trauma patient care monitoring data; perform monitoring type classification on the trauma patient care monitoring data to obtain care monitoring type data; fuse the standard trauma medical data and the care monitoring type data to generate trauma medicine - care monitoring data; Step S3: Determine trauma type characteristics for the trauma medicine - care monitoring data to generate trauma type characteristic data; evaluate the degree of trauma recovery for the trauma medicine - care monitoring data to generate trauma recovery degree data; perform trauma care relevance analysis on the trauma medicine - care monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma care association data; perform trauma care knowledge level division on the trauma care association data to generate trauma care knowledge level data; Step S4: Hierarchically cluster the trauma care knowledge level data to obtain hierarchically clustered data; construct a trauma care knowledge system from the hierarchically clustered data to generate a trauma care knowledge system.

[0005] By obtaining the original trauma medical data and performing data preprocessing and standardization, the present invention can ensure the accuracy and consistency of trauma medical data. This lays a solid foundation for subsequent data analysis and applications, improving the usability and reliability of the data. By obtaining the trauma patient care monitoring data and classifying the monitoring types, the care monitoring data can be effectively classified and organized. Fusing the standard trauma medical data with the care monitoring type data to generate trauma medical-care monitoring data realizes the organic combination of medical data and care monitoring data. Determining the trauma type characteristics and evaluating the trauma recovery degree for the trauma medical-care monitoring data can accurately identify the trauma type and evaluate the trauma recovery situation. Conducting a trauma care relevance analysis on the trauma medical-care monitoring data based on the trauma type characteristic data and the trauma recovery degree data reveals the internal connection between trauma care and patient recovery. Conducting a trauma care knowledge level division on the trauma care association data to generate trauma care knowledge level data provides a basis for constructing a systematic trauma care knowledge system. Hierarchically clustering the trauma care knowledge level data can classify and integrate similar care knowledge to form a clearer and more systematic knowledge structure. Constructing a trauma care knowledge system from the hierarchically clustered data to generate a trauma care knowledge system provides systematic and structured knowledge support for the practice and research of trauma care, improving the quality and efficiency of trauma care. Therefore, the present invention uses data processing technology, pattern recognition technology, and deep learning technology; realizes the fusion of trauma care medical data from trauma medical text information, image information, and care monitoring information, and realizes the division of relevant knowledge levels of the fused trauma care medical data; thus making the knowledge system in the field of trauma care medicine more comprehensively presented.

[0006] Preferably, step S1 includes the following steps: Step S11: Obtain the original trauma medical data, where the original trauma medical data includes trauma patient imaging examination data, trauma patient biochemical test data, and trauma patient electronic medical record data; Step S12: Perform image enhancement on the trauma patient imaging examination data to obtain enhanced trauma imaging data; perform data calibration on the trauma patient biochemical test data to obtain calibrated biochemical test data; perform text cleaning on the trauma patient electronic medical record data to obtain standardized electronic medical record data; Step S13: Integrate the enhanced trauma imaging data, calibrated biochemical test data, and standardized electronic medical record data to obtain trauma medical data; Step S14: Standardize the trauma medical data in the corresponding format to generate medically formatted standardized data; perform standardized coding on the trauma medical data to obtain standard trauma medical data.

[0007] The present invention comprehensively obtains the original medical data of trauma patients, including imaging examination data, biochemical test data, and electronic medical record data. The comprehensive acquisition of these data provides a rich information basis for subsequent analysis and processing, ensuring the integrity and diversity of the data. By performing image enhancement on the imaging examination data of trauma patients, the clarity and readability of the images can be improved, making the detailed features in the images more obvious. Data calibration of the biochemical test data of trauma patients can eliminate measurement errors and ensure the accuracy and reliability of the biochemical test data. Text cleaning of the electronic medical record data of trauma patients can remove irrelevant information and incorrect data to obtain standardized electronic medical record data. Integrating the trauma image enhancement data, biochemical test calibration data, and electronic medical record specification data enables the organic combination of data from different sources and types, forming a comprehensive data set. Performing format standardization and coding standardization on the trauma medical data generates medically formatted standardized data and standard trauma medical data. Format standardization ensures the compatibility and consistency of the data between different systems and platforms, facilitating data sharing and exchange. Coding standardization gives the data a unified identifier and specification, facilitating data management and analysis, and improving the operability and traceability of the data.

[0008] Preferably, step S2 includes the following steps: Step S21: Monitor the wound healing status of the patient; obtain the patient's drug use situation; record the activity of the patient's traumatized limb; Step S22: Identify the wound healing characteristics of the patient's wound healing status to generate wound healing characteristic types; determine the wound medication situation for the patient's drug use situation to generate wound medication situation types; classify the activity of the patient's traumatized limb to generate traumatized limb activity types; Step S23: Divide the nursing monitoring data of trauma patients according to the wound healing characteristic types, wound medication situation types, and traumatized limb activity types to obtain nursing monitoring type data; Step S24: Perform data fusion on the standard trauma medical data and the nursing monitoring type data to generate trauma medical-nursing monitoring data.

[0009] By monitoring the wound healing status of patients, obtaining the drug usage of patients, and recording the activities of the traumatized limbs of patients, the present invention can comprehensively understand various key indicators and behavioral characteristics of patients during the trauma recovery process, ensuring a comprehensive grasp of the patient's recovery status; by identifying the characteristics of the wound healing status of patients, determining the medication situation of drug usage, and classifying the types of activities of the traumatized limbs, the complex information monitored can be effectively simplified and classified, facilitating targeted analysis and evaluation of different aspects of the patient's recovery process; by classifying the nursing monitoring data of trauma patients according to the types of wound healing characteristics, wound medication situations, and types of traumatized limb activities, the nursing monitoring data can be classified more meticulously and systematically. The obtained nursing monitoring type data can better understand the performance and needs of patients in different nursing monitoring dimensions; by fusing the standard trauma medical data with the nursing monitoring type data, the medical data and the nursing monitoring data are organically combined to form a more comprehensive and integrated data set. This provides richer information support for analyzing the recovery process of trauma patients.

[0010] Preferably, step S24 includes the following steps: Step S241: Identify the wound appearance characteristics of the wound healing characteristic type to obtain wound appearance characteristic data; judge the wound healing process based on the wound appearance characteristic data to generate wound healing process data; Step S242: Correlate the trauma image enhancement data with the wound appearance characteristic data and the wound healing process data to obtain image-healing correspondence data; Step S243: Determine the types of wound medications for the wound medication situation type to obtain wound medication type data; record the wound disinfection frequency based on the wound medication type data to obtain the wound disinfection frequency; extract the medication dosage for the drug usage of trauma patients according to the wound disinfection frequency to generate the wound medication dosage; Step S244: Match and fuse the biochemical test calibration data with the wound medication type data, the wound disinfection frequency, and the wound medication dosage to obtain test-medication fusion data; Step S245: Mark the limb joints for the type of traumatized limb activities to generate traumatized limb joint data; determine the range of joint activities based on the traumatized limb joint data to obtain the range of joint activities data; Step S246: Extract the medical order requirements for the traumatized limb activities from the electronic medical record specification data to obtain the medical order data for the traumatized limb activities; correlate the medical order data for the traumatized limb activities with the range of joint activities data to obtain medical record-limb activity data; Step S247: Perform data fusion on the image-healing correspondence data, the detection-medication fusion data, and the medical record-limb movement data to generate trauma medicine-nursing monitoring data.

[0011] The present invention identifies the appearance feature types of wound healing characteristics, and can accurately obtain the appearance feature data of the wound. Further, the wound healing process is judged based on the wound appearance feature data to generate wound healing process data. This provides quantitative and qualitative bases for evaluating the healing status of the wound, making the monitoring of wound healing more accurate and systematic; the trauma image enhancement data is corresponded to the wound appearance feature data and the wound healing process data, so that the image data is closely combined with the wound healing status information, providing more intuitive and accurate data support for monitoring wound healing by imaging means and better understanding the dynamic changes of wound healing; the types of wound medications are determined, and the disinfection frequency of the wound medication type data is recorded; the medication dosage is extracted according to the wound disinfection frequency for the drug use situation, making the management of the wound medication situation more refined and providing detailed data records and bases for rational drug use; the biochemical test calibration data is matched and fused with the wound medication type data, the wound disinfection frequency, and the wound medication dosage, so that the biochemical test results are closely combined with the wound medication situation, providing more comprehensive and integrated data support for evaluating the drug use effect and monitoring the changes in the patient's biochemical indicators and better analyzing the impact of the drug on the patient's biochemical indicators; the limb joints of the trauma limb movement type are marked to generate trauma limb joint data, and the range of joint movement of the trauma limb joint data is determined, providing accurate joint movement information for monitoring the movement of the trauma limb, making the evaluation of the range of limb movement more accurate and quantitative and clarifying the recovery status of the patient's limb function; the requirements of the trauma limb movement doctor's order are extracted from the standardized electronic medical record data to obtain trauma limb movement doctor's order data, and the trauma limb movement doctor's order data is corresponded to the range of joint movement data to obtain medical record-limb movement data. This corresponding relationship makes the doctor's order information in the medical record closely combined with the actual limb movement situation, providing data support for evaluating whether the patient performs limb movement according to the doctor's order; the image-healing correspondence data, the detection-medication fusion data, and the medical record-limb movement data are fused, enabling the organic combination of data from different dimensions and sources to form a comprehensive data set.

[0012] Preferably, step S3 includes the following steps: Step S31: Identify the trauma characteristics of the trauma medicine-nursing monitoring data to obtain trauma characteristic data; extract the manifestation characteristics of the trauma characteristic data to generate trauma manifestation characteristic data; Step S32: Perform characterization division on the trauma manifestation feature data to obtain trauma characterization division data; perform trauma type feature pairing based on the trauma characterization division data to generate trauma type feature data; Step S33: Evaluate the trauma recovery degree of the trauma medicine - nursing monitoring data to generate trauma recovery degree data; Step S34: Perform trauma nursing relevance analysis on the trauma medicine - nursing monitoring data based on the trauma type feature data and the trauma recovery degree data to obtain trauma nursing correlation data; Step S35: Perform trauma nursing knowledge level division on the trauma nursing correlation data to generate trauma nursing knowledge level data.

[0013] The present invention performs trauma feature recognition on the trauma medicine - nursing monitoring data, and can accurately extract trauma feature data. Further, perform manifestation feature extraction on the trauma feature data to generate trauma manifestation feature data. This provides detailed basic data for in - depth analysis of trauma features, making the understanding of trauma features more comprehensive and in - depth; perform characterization division on the trauma manifestation feature data to obtain trauma characterization division data. Perform trauma type feature pairing based on the trauma characterization division data to generate trauma type feature data. This process makes the classification of trauma features clearer and more systematic, better identifying and distinguishing different types of trauma. Evaluate the trauma recovery degree of the trauma medicine - nursing monitoring data to generate trauma recovery degree data. This kind of evaluation can quantify the trauma recovery situation, providing objective data support for monitoring the trauma recovery process, making the monitoring of trauma recovery more accurate and quantifiable, and better understanding the patient's recovery status. Perform trauma nursing relevance analysis on the trauma medicine - nursing monitoring data based on the trauma type feature data and the trauma recovery degree data to obtain trauma nursing correlation data. This kind of analysis can reveal the internal connection between trauma features and nursing measures, providing a scientific basis for optimizing the nursing plan, improving the effectiveness and pertinence of nursing, and ensuring that nursing measures can better meet the patient's recovery needs. Perform trauma nursing knowledge level division on the trauma nursing correlation data, making the organization of trauma nursing knowledge more systematic and hierarchical, providing a basis for constructing a comprehensive trauma nursing knowledge system, and better managing and applying trauma nursing knowledge.

[0014] Preferably, step S33 includes the following steps: Step S331: Determine the tissue structure of the trauma healing area for the image - healing corresponding data to obtain the tissue structure of the healing area; perform tissue healing integrity recognition on the tissue structure of the healing area to generate a structural healing integrity index; Step S332: Determine the medication cycle for the detection-medication fusion data to obtain the trauma medication cycle data; map the promotion effect of trauma healing drugs on the tissue structure of the healing area according to the trauma medication cycle data to generate a drug promotion effect index; Step S333: Judge the movable range of the limb for the medical record-limb movement data to obtain the movable range data of the limb; determine the degree of limb function recovery for the movable range of the limb to generate a limb function recovery index; Step S334: Calculate the trauma recovery degree index based on the structural healing integrity index, the drug promotion effect index, and the limb function recovery index to obtain the trauma recovery degree data.

[0015] The present invention determines the tissue structure of the trauma healing area for the image-healing corresponding data, and can accurately identify the tissue structure characteristics of the healing area. Further, identify the healing integrity of the tissue structure of the healing area to generate a structural healing integrity index. This provides a quantitative index for evaluating the tissue integrity of trauma healing, making the monitoring of the tissue structure recovery of the healing area more accurate and objective; determine the medication cycle for the detection-medication fusion data to obtain the trauma medication cycle data. Map the promotion effect of trauma healing drugs on the tissue structure of the healing area according to the trauma medication cycle data to generate a drug promotion effect index. This process can quantify the promotion effect of drugs on trauma healing, provide specific data support for evaluating the role of drugs in the trauma recovery process, and better understand the impact of drugs on the healing process. Judge the movable range of the limb for the medical record-limb movement data, and further determine the degree of limb function recovery for the movable range of the limb to generate a limb function recovery index. This provides a quantitative index for evaluating the functional recovery of the traumatized limb, making the monitoring of limb movement ability and functional recovery more accurate and comprehensive, and better understanding the functional recovery status of the patient's limb. Calculate the trauma recovery degree index based on the structural healing integrity index, the drug promotion effect index, and the limb function recovery index to obtain the trauma recovery degree data. The comprehensive calculation makes the evaluation of the trauma recovery degree more comprehensive and comprehensive, and can comprehensively consider the impacts of tissue healing, drug action, and limb function recovery and other aspects.

[0016] Preferably, step S34 includes the following steps: Step S341: Identify the healing feature nursing mode for the image-healing corresponding data based on the trauma type feature data to obtain the healing feature nursing mode; Step S342: Associate the drug usage situation for the detection-medication fusion data according to the healing feature nursing mode to obtain the drug usage association data; Step S343: Extract the contribution degree of the trauma nursing mode recovery for the medical record-limb movement data based on the trauma recovery degree data to obtain the nursing mode contribution degree; Step S344: Sort the contribution degrees of the nursing models to generate a recovery contribution degree sorting value; correlate the trauma medicine-nursing monitoring data according to the recovery contribution degree sorting value to obtain trauma nursing effect data; Step S345: Combine the healing feature nursing model, the drug use correlation data, and the trauma nursing effect data through trauma nursing correlation to obtain trauma nursing correlation data.

[0017] Based on the trauma type characteristic data, the present invention identifies the healing feature nursing model for the imaging-healing corresponding data, and can accurately identify the nursing model that matches the trauma healing characteristics to obtain the healing feature nursing model, enabling the nursing measures to better adapt to the actual situation of trauma healing, and improving the pertinence and effectiveness of nursing. Correlate the detection-medication fusion data according to the healing feature nursing model to obtain drug use correlation data. It can combine the drug use situation with the nursing model to better understand the specific application of drugs in trauma nursing. Extract the contribution degree of the trauma nursing model recovery from the medical record-limb movement data based on the trauma recovery degree data, which can quantify the contribution degree of different nursing models to trauma recovery, provide specific data indicators for evaluating the effectiveness of the nursing model, and identify the nursing model with a greater contribution to trauma recovery. Sort the contribution degrees of the nursing models, and correlate the trauma medicine-nursing monitoring data according to the recovery contribution degree sorting value, making the effect evaluation of the nursing model clearer and more systematic, and enabling an intuitive understanding of the influence degree of different nursing models on trauma recovery. Combine the healing feature nursing model, the drug use correlation data, and the trauma nursing effect data through trauma nursing correlation to obtain trauma nursing correlation data. This combination organically combines nursing information from different dimensions to form a comprehensive nursing correlation dataset, providing rich data support for comprehensively analyzing all aspects of trauma nursing, and helping to better understand and optimize the entire process of trauma nursing.

[0018] Preferably, step S35 includes the following steps: Step S351: Identify the healing feature stage for the trauma nursing correlation data to obtain the healing feature stage; divide the trauma nursing model according to the healing feature stage for the healing feature nursing model to generate a trauma nursing model hierarchy; Step S352: Identify the nursing drug types for the trauma nursing correlation data to obtain the nursing drug types; extract the nursing drug doses for the trauma nursing correlation data to obtain the nursing drug doses; determine the nursing medication cycle for the trauma nursing correlation data to obtain the nursing medication cycle; Step S353: Divide the trauma nursing drugs according to the nursing drug types, nursing drug doses, and nursing medication cycle to generate a trauma nursing drug hierarchy; Step S354: Perform a hierarchical division of the trauma care - related data for the trauma care effect to generate the trauma care effect hierarchy; Step S355: Perform a hierarchical division of the trauma care knowledge for the trauma care mode hierarchy, the trauma care drug hierarchy, and the trauma care effect hierarchy to obtain the trauma care knowledge hierarchy data.

[0019] The present invention identifies the healing characteristic stages for the trauma care - related data, can accurately divide different stages of trauma healing to obtain the healing characteristic stages. Based on the healing characteristic stages, a hierarchical division of the healing characteristic care mode is performed to generate the trauma care mode hierarchy. This provides a basis for formulating a hierarchical care mode according to different stages of trauma healing, enabling nursing measures to better adapt to the dynamic changes of trauma healing, and improving the pertinence and systematicness of nursing. By identifying the types of nursing drugs, extracting the nursing drug doses, and determining the nursing medication cycles for the trauma care - related data, comprehensive drug information involved in trauma care can be obtained, including the types of nursing drugs, the nursing drug doses, and the nursing medication cycles. The acquisition of this information provides detailed data support for subsequent drug hierarchical division and nursing optimization. According to the types of nursing drugs, the nursing drug doses, and the nursing medication cycles, a hierarchical division of the trauma care drugs is performed to generate the trauma care drug hierarchy. This division can systematically organize and classify drug information, improving the rationality and effectiveness of drug use. Performing a hierarchical division of the trauma care - related data for the trauma care effect to generate the trauma care effect hierarchy. This makes the evaluation of the nursing effect more systematic and hierarchical, enabling the analysis of the effects of nursing measures from different dimensions and levels, providing a more comprehensive and detailed data basis for the monitoring and improvement of the nursing effect, and helping to identify more effective nursing strategies. A comprehensive division of the trauma care mode hierarchy, the trauma care drug hierarchy, and the trauma care effect hierarchy is performed to obtain the trauma care knowledge hierarchy data. This comprehensive division makes the organization of trauma care knowledge more systematic and hierarchical, provides a basis for constructing a comprehensive trauma care knowledge system, helps to better manage and apply trauma care knowledge, and improves the efficiency and quality of nursing work.

[0020] Preferably, step S4 includes the following steps: Step S41: Measure the inter - layer similarity of the trauma care knowledge hierarchy data to obtain the inter - layer similarity measurement value; extract the inter - layer association strength from the inter - layer similarity measurement value to obtain the inter - layer association strength; Step S42: Determine the inter - layer similarity coefficient for the trauma care knowledge hierarchy data according to the inter - layer association strength to generate the hierarchical similarity coefficient; perform a similar layer marking on the trauma care knowledge hierarchy data based on the hierarchical similarity coefficient to obtain the similar knowledge layer; cluster - associate the similar knowledge layers to form the hierarchical clustering data; Step S43: Extract knowledge nodes from the hierarchical clustering data to obtain trauma care knowledge nodes; determine the hierarchical structure of the trauma care knowledge nodes to obtain node hierarchical structure data; determine the logical relationships of the trauma care knowledge nodes to obtain node logical relationships. Step S44: Construct association relationships for the hierarchical clustering data based on the node hierarchical structure data and the node logical relationships to obtain a trauma care knowledge system.

[0021] The present invention measures the inter-layer similarity of trauma care knowledge hierarchical data, which can quantify the similarity degree between different layers and obtain an inter-layer similarity measurement value. Further, extract the association strength of the inter-layer similarity measurement value to obtain the inter-layer association strength. This provides a quantitative basis for identifying the internal connections between knowledge layers, reveals the mutual relationships and mutual influences between different nursing knowledge layers, and lays a foundation for subsequent knowledge integration and optimization. Determine the inter-layer similarity coefficient for the trauma care knowledge hierarchical data according to the inter-layer association strength to generate a hierarchical similarity coefficient. Based on the hierarchical similarity coefficient, mark the similar layers of the trauma care knowledge hierarchical data to obtain similar knowledge layers, and cluster and associate the similar knowledge layers to form hierarchical clustering data. This enables similar knowledge layers to be effectively identified and classified, provides clearer and more orderly data support for constructing a systematic knowledge structure, and improves the logic and systematicness of knowledge organization. Extract knowledge nodes from the hierarchical clustering data to obtain trauma care knowledge nodes. Further, determine the hierarchical structure of the trauma care knowledge nodes to obtain node hierarchical structure data, and determine the logical relationships of the trauma care knowledge nodes to obtain node logical relationships. This provides specific data support for clarifying the hierarchical relationships and logical relationships between knowledge nodes, makes the organization of knowledge nodes more orderly and reasonable, and lays a foundation for constructing a complete knowledge system. Construct association relationships for the hierarchical clustering data based on the node hierarchical structure data and the node logical relationships to obtain a trauma care knowledge system. This construction makes the organization of trauma care knowledge more systematic and structured, forming a complete knowledge system.

[0022] In this specification, a knowledge mining system for the field of trauma care medicine is provided, which is used to execute the above-mentioned knowledge mining method for the field of trauma care medicine. The knowledge mining system for the field of trauma care medicine includes: A trauma medical data acquisition module, which is used to obtain original trauma medical data; perform data preprocessing on the original trauma medical data to obtain trauma medical data; perform standardization processing on the trauma medical data to obtain standard trauma medical data. The trauma medicine - nursing monitoring integration module is used to obtain the nursing monitoring data of trauma patients; classify the monitoring types of the nursing monitoring data of trauma patients to obtain nursing monitoring type data; fuse the standard trauma medicine data with the nursing monitoring type data to generate trauma medicine - nursing monitoring data; The trauma nursing knowledge level classification module is used to determine the trauma type characteristics of the trauma medicine - nursing monitoring data to generate trauma type characteristic data; evaluate the trauma recovery degree of the trauma medicine - nursing monitoring data to generate trauma recovery degree data; conduct a trauma nursing relevance analysis on the trauma medicine - nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma nursing correlation data; classify the trauma nursing correlation data into trauma nursing knowledge levels to generate trauma nursing knowledge level data; The trauma nursing knowledge system construction module is used to perform hierarchical clustering on the trauma nursing knowledge level data to obtain hierarchical clustering data; construct a trauma nursing knowledge system from the hierarchical clustering data to generate a trauma nursing knowledge system.

[0023] Through the trauma medical data acquisition module, the original trauma medical data is obtained and subjected to data preprocessing and standardization processing, which can ensure the accuracy and consistency of the trauma medical data. This lays a solid foundation for subsequent data analysis and applications, and improves the usability and reliability of the data. Through the trauma medicine-nursing monitoring integration module, the nursing monitoring data of trauma patients is obtained and the monitoring types are classified, which can effectively classify and organize the nursing monitoring data. The standard trauma medical data and the nursing monitoring type data are integrated to generate trauma medicine-nursing monitoring data, realizing the organic combination of medical data and nursing monitoring data. Through the trauma nursing knowledge level division module, the trauma type characteristics are determined and the trauma recovery degree is evaluated for the trauma medicine-nursing monitoring data, which can accurately identify the trauma type and evaluate the trauma recovery situation. Based on the trauma type characteristic data and the trauma recovery degree data, the trauma nursing relevance analysis is carried out on the trauma medicine-nursing monitoring data to reveal the internal relationship between trauma nursing and patient recovery. The trauma nursing relevance data is divided into trauma nursing knowledge levels to generate trauma nursing knowledge level data, providing a basis for constructing a systematic trauma nursing knowledge system. Through the trauma nursing knowledge system construction module, hierarchical clustering is carried out on the trauma nursing knowledge level data, which can classify and integrate similar nursing knowledge to form a clearer and more systematic knowledge structure. The hierarchical clustering data is used to construct a trauma nursing knowledge system to generate a trauma nursing knowledge system, providing systematic and structured knowledge support for the practice and research of trauma nursing, and improving the quality and efficiency of trauma nursing. Therefore, the present invention realizes the fusion of trauma nursing medical data for trauma medical text information, image information, and nursing monitoring information through data processing technology, pattern recognition technology, and deep learning technology; and realizes the division of relevant knowledge levels of the fused trauma nursing medical data; thereby making the knowledge system in the field of trauma nursing medicine more comprehensively presented. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic flow chart of the steps of a knowledge mining method for the field of trauma nursing medicine; Figure 2 is Figure 1 a detailed implementation step flow chart of step S3 in Figure 3 is Figure 2 a detailed implementation step flow chart of step S33 in The realization, functional characteristics, and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] The technical method of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0026] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0027] It should be understood that although the terms "first", "second", etc. may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit is called the second unit, and similarly the second unit is called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed related items.

[0028] To achieve the above object, please refer to Figures 1 to 3 , a knowledge mining method for the field of trauma care medicine, the method comprising the following steps: Step S1: Obtain original trauma medical data; perform data preprocessing on the original trauma medical data to obtain trauma medical data; perform standardization processing on the trauma medical data to obtain standard trauma medical data; Step S2: Obtain trauma patient care monitoring data; perform monitoring type classification on the trauma patient care monitoring data to obtain care monitoring type data; perform data fusion on the standard trauma medical data and the care monitoring type data to generate trauma medical-care monitoring data; Step S3: Determine trauma type characteristics for the trauma medical-care monitoring data to generate trauma type characteristic data; evaluate the degree of trauma recovery for the trauma medical-care monitoring data to generate trauma recovery degree data; perform trauma care relevance analysis on the trauma medical-care monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma care association data; perform trauma care knowledge level classification on the trauma care association data to generate trauma care knowledge level data; Step S4: Perform hierarchical clustering on the trauma care knowledge level data to obtain hierarchical clustering data; construct a trauma care knowledge system from the hierarchical clustering data to generate a trauma care knowledge system.

[0029] By obtaining the original trauma medical data and performing data preprocessing and standardization, the present invention can ensure the accuracy and consistency of trauma medical data. This lays a solid foundation for subsequent data analysis and applications, improving the usability and reliability of the data. By obtaining the trauma patient care monitoring data and classifying the monitoring types, the care monitoring data can be effectively classified and organized. Fusing the standard trauma medical data with the care monitoring type data to generate trauma medical-care monitoring data realizes the organic combination of medical data and care monitoring data. Determining the trauma type characteristics and evaluating the trauma recovery degree of the trauma medical-care monitoring data can accurately identify the trauma type and evaluate the trauma recovery situation. Conducting a trauma care relevance analysis on the trauma medical-care monitoring data based on the trauma type characteristic data and the trauma recovery degree data reveals the internal connection between trauma care and patient recovery. Conducting a trauma care knowledge level division on the trauma care correlation data to generate trauma care knowledge level data provides a basis for constructing a systematic trauma care knowledge system. Performing hierarchical clustering on the trauma care knowledge level data can classify and integrate similar care knowledge to form a clearer and more systematic knowledge structure. Constructing a trauma care knowledge system from the hierarchical clustering data to generate a trauma care knowledge system provides systematic and structured knowledge support for the practice and research of trauma care, improving the quality and efficiency of trauma care. Therefore, the present invention uses data processing technology, pattern recognition technology, and deep learning technology; realizes the fusion of trauma care medical data for trauma medical text information, image information, and care monitoring information, and realizes the division of relevant knowledge levels for the fused trauma care medical data; thus making the knowledge system in the field of trauma care medicine more comprehensively presented.

[0030] In the embodiment of the present invention, referring to Figure 1 as shown, in this example, the knowledge mining method for the field of trauma care medicine includes the following steps: Step S1: Obtain the original trauma medical data; perform data preprocessing on the original trauma medical data to obtain trauma medical data; perform standardization processing on the trauma medical data to obtain standard trauma medical data; In the embodiments of the present invention, the original trauma medical data is obtained from the hospital's information systems, such as the Electronic Health Record (EHR) system and the Hospital Information System (HIS). For example, through the interfaces in the HIS system, the electronic medical records of patients, imaging data (such as CT, MRI scans), and laboratory test results are extracted. Data preprocessing includes steps such as data cleaning, format conversion, and missing value processing. For imaging data, image processing techniques are used for denoising and enhancement. For example, median filtering is used to remove noise in the image, and then histogram equalization technology is used to enhance the contrast of the image. For structured data, such as the text data in electronic medical records, natural language processing (NLP) techniques are used for text cleaning and standardization. For example, irrelevant symbols and stop words are removed. The purpose of the standardization process is to make the data comparable and consistent among different systems and institutions. For text data, medical terms are mapped to a unified standard term system, such as the ICD-10 coding system. For numerical data, normalization is performed to unify its value range to between [0,1]. For example, the min-max normalization method is used. In addition, a unified data model is established to define data formats, field lengths, value ranges, etc., to ensure the standardization and normalization of the data.

[0031] Step S2: Obtain the nursing monitoring data of trauma patients; classify the nursing monitoring data of trauma patients to obtain the nursing monitoring type data; fuse the standard trauma medical data and the nursing monitoring type data to generate trauma medical-nursing monitoring data; In the embodiments of the present invention, the nursing monitoring data of trauma patients is obtained through the hospital's Electronic Health Record (EHR) system and the Nursing Information System (NIS). For example, through the interfaces in the EHR system, the nursing records of patients are extracted, including vital sign monitoring data (such as heart rate, blood pressure, body temperature, etc.), nursing operation records (such as wound dressing change, infusion records, etc.), and the psychological state assessment of patients. The classification of monitoring types is achieved through data classification techniques. First, different monitoring types are defined, such as vital sign monitoring, wound care monitoring, psychological state monitoring, etc. Then, classification algorithms in data mining, such as decision trees or Support Vector Machines (SVM), are used to classify the nursing monitoring data. For example, for vital sign data, features such as heart rate and blood pressure are extracted, and the decision tree algorithm is used to classify them into categories such as normal and abnormal. Data fusion is achieved through data integration techniques. First, the standard trauma medical data and the nursing monitoring type data are aligned to ensure the consistency of the data in terms of timestamp and patient identification. Then, data fusion algorithms, such as weighted average or Bayesian fusion methods, are used to fuse the two types of data. For example, the imaging examination results of patients (such as CT scan results) are fused with the vital sign monitoring data to generate a comprehensive dataset that contains the imaging features and vital sign information of the patient.

[0032] Step S3: Determine the trauma type characteristics of the trauma medicine - nursing monitoring data to generate trauma type characteristic data; evaluate the trauma recovery degree of the trauma medicine - nursing monitoring data to generate trauma recovery degree data; perform trauma nursing correlation analysis on the trauma medicine - nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma nursing correlation data; perform hierarchical division of trauma nursing knowledge on the trauma nursing correlation data to generate trauma nursing knowledge hierarchy data; In the embodiment of the present invention, the determination of trauma type characteristics is achieved through feature extraction and selection techniques in machine learning. First, extract features related to the trauma type from the trauma medicine - nursing monitoring data, such as the trauma location, trauma severity (such as the RISS score), etc. Then use feature selection algorithms, such as mutual information or recursive feature elimination (RFE), to select the most representative feature combination. For example, for the trauma location feature, extract the injury conditions of parts such as the head, chest, and abdomen, and combine with the trauma severity score to determine the trauma type characteristic data set. The evaluation of trauma recovery degree is achieved through data analysis and modeling techniques. Define evaluation indicators for the trauma recovery degree, such as the stability of vital signs, wound healing conditions, etc. Then, use regression analysis or time series analysis methods to model and predict the patient's recovery situation. For example, by analyzing the changing trends of the patient's vital sign data (such as heart rate, blood pressure) over time, use a linear regression model to evaluate the recovery degree. The trauma nursing correlation analysis is achieved through association rule mining techniques. Associate the trauma type characteristic data and the trauma recovery degree data with the nursing monitoring data to construct an association data set. Use the Apriori algorithm or the FP - Growth algorithm to mine the association rules between the trauma type and nursing measures. For example, it is found through analysis that for patients with head trauma, there is a significant association between timely neurosurgical nursing measures and a higher recovery degree. The hierarchical division of trauma nursing knowledge is achieved through hierarchical clustering analysis techniques. First, construct a feature matrix according to the features in the trauma nursing correlation data, such as trauma type, nursing measures, recovery degree, etc. Then, use hierarchical clustering algorithms, such as the AGNES algorithm, to perform hierarchical division on the data to form different knowledge hierarchies. For example, divide trauma nursing knowledge into levels such as basic nursing, specialized nursing, and rehabilitation nursing, and each level contains corresponding nursing knowledge and measures.

[0033] Step S4: Perform hierarchical clustering on the trauma nursing knowledge hierarchy data to obtain hierarchical clustering data; construct a trauma nursing knowledge system from the hierarchical clustering data to generate a trauma nursing knowledge system.

[0034] In the embodiments of the present invention, hierarchical clustering is implemented through the AGNES algorithm. First, each knowledge point in the hierarchical data of trauma care knowledge is regarded as an initial cluster. Then, the distances between each cluster are calculated, and common distance metrics include Euclidean distance or cosine similarity. Next, according to the selected merging strategy (such as average linkage or Ward's method), the nearest clusters are gradually merged. For example, using Ward's method, the clusters to be merged are selected by minimizing the variance within the clusters, thereby forming a hierarchical clustering structure. Finally, a dendrogram is generated to show the merging process and hierarchical structure of the clusters. The construction of the trauma care knowledge system is implemented through knowledge graph technology. The knowledge points in the hierarchical clustering data are used as nodes, and the relationships between the nodes are determined according to the clustering results. Specifically, basic care, specialized care, and rehabilitation care are used as the main nodes, and hierarchical relationships between these nodes are established according to the results of hierarchical clustering. Then, graph database technology (such as Neo4j) is used to store and manage the knowledge graph, and each node and its relationship are stored as a graph structure. In this way, a complete trauma care knowledge system is constructed to support the query and reasoning of nursing knowledge.

[0035] Preferably, step S1 includes the following steps: Step S11: Obtain the original trauma medical data, where the original trauma medical data includes trauma patient imaging examination data, trauma patient biochemical test data, and trauma patient electronic medical record data; Step S12: Perform image enhancement on the trauma patient imaging examination data to obtain enhanced trauma imaging data; perform data calibration on the trauma patient biochemical test data to obtain calibrated biochemical test data; perform text cleaning on the trauma patient electronic medical record data to obtain standardized electronic medical record data; Step S13: Integrate the enhanced trauma imaging data, calibrated biochemical test data, and standardized electronic medical record data to obtain trauma medical data; Step S14: Standardize the corresponding format of the trauma medical data to generate standardized medical format data; perform standardized coding on the trauma medical data to obtain standardized trauma medical data.

[0036] In the embodiments of the present invention, the original trauma medical data includes trauma patient imaging examination data, trauma patient biochemical test data, and trauma patient electronic medical record data. The imaging examination data is obtained through the hospital's Picture Archiving and Communication System (PACS), such as CT and MRI scan image files; the biochemical test data is obtained from the hospital's Laboratory Information System (LIS), such as numerical data of test results of blood, urine, etc.; the electronic medical record data is obtained through the hospital's Electronic Health Record (EHR) system, including text data such as the patient's medical history records and doctor's order information. Image enhancement processing is performed using image processing software such as SimpleITK or OpenCV. For example, contrast enhancement is performed on CT images, and the histogram equalization method is used. By adjusting the histogram distribution of the image, the contrast of the image becomes more obvious, thus obtaining enhanced trauma image data. Data calibration is performed using statistical analysis software such as R or the NumPy library of Python. For example, for the blood glucose value in blood tests, if a systematic error is found, the relationship between the blood glucose value and the standard value is fitted through a linear regression model, and then all blood glucose data is calibrated according to the fitting result to obtain calibrated biochemical test data. Text cleaning is performed using natural language processing (NLP) tools such as NLTK or Spacy. For example, the medical history records in the electronic medical record are cleaned, removing irrelevant symbols and stop words (such as "of", "is", etc.), and performing stemming to obtain standardized electronic medical record data. The enhanced trauma image data, calibrated biochemical test data, and standardized electronic medical record data are integrated using data fusion technology. A unified identifier is defined for each type of data, such as patient ID and timestamp, and data fusion algorithms such as weighted average or Bayesian fusion method are used to fuse data from different sources. For example, the imaging feature data, biochemical test numerical data, and text feature data in the electronic medical record of the patient are fused. Format standardization of trauma medical data is performed using data conversion tools such as ETL (Extract, Transform, Load) software. For example, the image data is converted to DICOM format, the biochemical test data is converted to HL7 format, and the electronic medical record data is converted to XML format to generate standardized medical format data. Standardized coding of trauma medical data is performed using medical terminology standards such as ICD-10 or SNOMED CT. The names in the electronic medical record are mapped to ICD-10 codes, and the terms in the imaging report are mapped to SNOMED CT codes to obtain standard trauma medical data.

[0037] Preferably, step S2 includes the following steps: Step S21: Monitor the wound healing status of the patient; obtain the patient's drug usage; record the movement of the traumatized limb of the patient; Step S22: Identify the wound healing characteristics of the patient's wound healing status to generate the wound healing characteristic type; determine the wound medication situation of the patient's drug use to generate the wound medication situation type; classify the activities of the traumatized limb of the patient to generate the traumatized limb activity type; Step S23: Classify the nursing monitoring data of the trauma patient according to the wound healing characteristic type, the wound medication situation type, and the traumatized limb activity type to obtain the nursing monitoring type data; Step S24: Perform data fusion on the standard trauma medical data and the nursing monitoring type data to generate trauma medicine - nursing monitoring data.

[0038] In the embodiment of the present invention, optical coherence tomography (OCT) technology is used to non-invasively monitor the changes in blood vessels and structural characteristics during wound healing over time. For example, through OCT imaging, features such as vasodilation around and immediately below the wound, and the migration of newly formed blood vessels into the wound are observed. The patient's drug use records, including information such as drug name, dosage, and frequency of use, are extracted from the hospital's electronic health record (EHR) system. Wearable devices such as accelerometers and gyroscopes are used to real-time monitor the patient's limb movement data, such as range of motion and frequency of movement. Image recognition technology is used to process the wound pictures, and the proportion of each tissue type in the wound image is identified based on the wound color. For example, by analyzing the wound pictures, the wound healing characteristic type is obtained. The drug use records are analyzed to extract drug information related to the wound, thereby generating the wound medication situation type. Machine learning algorithms are used to classify the limb movement data. For example, the support vector machine (SVM) algorithm is adopted, and according to the movement data recorded by the accelerometer and gyroscope, the patient's limb movements are classified into types such as rest, slight activity, moderate activity, and vigorous activity. According to the wound healing characteristic type, the wound medication situation type, and the traumatized limb activity type, the nursing monitoring data of the trauma patient is classified. Data fusion is performed on the standard trauma medical data and the nursing monitoring type data, and data fusion techniques such as weighted average or Bayesian fusion method are used. For example, the patient's imaging feature data (such as CT scan results), biochemical test numerical data (such as blood sugar, white blood cell count, etc.), and text feature data (such as medical record) in the electronic medical record are fused with the nursing monitoring type data to generate a trauma medicine - nursing monitoring dataset.

[0039] Preferably, step S24 includes the following steps: Step S241: Identify the wound appearance characteristics of the wound healing characteristic type to obtain the wound appearance characteristic data; judge the wound healing process of the wound appearance characteristic data to generate the wound healing process data; Step S242: Correlate the enhanced trauma image data with the wound appearance feature data and the wound healing process data to obtain image-healing correspondence data; Step S243: Determine the types of wound medications to obtain wound medication type data; record the wound disinfection frequency for the wound medication type data to obtain the wound disinfection frequency; extract the medication dosage for the trauma patient's drug use based on the wound disinfection frequency to generate the wound medication dosage; Step S244: Match and fuse the biochemical test calibration data with the wound medication type data, the wound disinfection frequency, and the wound medication dosage to obtain test-medication fusion data; Step S245: Mark the limb joints for the types of trauma limb activities to generate trauma limb joint data; determine the range of joint activities for the trauma limb joint data to obtain the joint activity range data; Step S246: Extract the trauma limb activity medical order requirements from the electronic medical record specification data to obtain the trauma limb activity medical order data; correlate the trauma limb activity medical order data with the joint activity range data to obtain medical record-limb activity data; Step S247: Fuse the image-healing correspondence data, the test-medication fusion data, and the medical record-limb activity data to generate trauma medicine-nursing monitoring data.

[0040] In the embodiments of the present invention, image recognition technology is used to process wound pictures. By using a deep convolutional neural network model (such as ResNet), features of the image in the wound area are extracted to identify appearance features such as the contour and color distribution of the wound. Through the convolutional layer and pooling layer of the model, pixel features in the wound area are extracted to obtain wound appearance feature data. Based on the wound appearance feature data and in combination with a preset wound healing process standard, a judgment is made. Specifically, the evaluation indexes for the wound healing process are set as the reduction ratio of the wound area, the proportion of granulation tissue, etc. The wound area in the wound appearance feature data is compared with the initial area to calculate the reduction ratio; at the same time, the proportion of granulation tissue in the wound is counted, and wound healing process data is generated by integrating these indexes. The trauma image enhancement data is corresponded to the wound appearance feature data and the wound healing process data. For example, the enhanced CT scan image is spatially registered with the wound contour in the wound appearance feature data to ensure their consistency in anatomical position. Then, the healing indexes in the wound healing process data are associated with the corresponding areas in the image data. For example, the reduction ratio of the wound area is corresponded to the size change of the wound area in the image to obtain image-healing corresponding data. The types of wound medications are analyzed to extract drug ingredient information. For example, ingredients such as antibiotics and painkillers are identified from the drug names to obtain wound medication type data. The number of times of wound disinfection of the patient is counted, and the time point of each disinfection is recorded. For example, by analyzing the disinfection operation records in the nursing records, the frequency of the patient receiving wound disinfection within a specific time period is extracted to obtain the wound disinfection frequency; according to the wound disinfection frequency, the drug dosage used during each disinfection is extracted from the drug use records. For example, if the patient used 10 ml of povidone iodine solution during a certain disinfection, this dosage information is associated with the corresponding disinfection time point to generate the wound medication dosage. The biochemical test calibration data is matched and fused with the wound medication type data, the wound disinfection frequency, and the wound medication dosage. For example, biochemical test indexes such as the patient's blood glucose and white blood cell count are associated with the wound medication type data to analyze the impact of drug use on the biochemical indexes; at the same time, in combination with the wound disinfection frequency and the medication dosage, the rationality and effect of drug use are evaluated to obtain test-medication fusion data. The types of activities of the traumatized limb are analyzed to identify the involved joint parts. For example, by analyzing the video or image of the patient's limb movement, computer vision technology is used to mark key parts such as the shoulder joint and the knee joint to generate traumatized limb joint data; the range of motion angles of the joints is measured. For example, a wearable device such as a gyroscope is used to record the rotation angles of the joints during movement, and the maximum and minimum ranges of motion of the joints are calculated to obtain joint range of motion data; the requirements of the doctor's orders for the activities of the traumatized limb are extracted from the standardized electronic medical record data to obtain traumatized limb activity order data. For example, the doctor's orders for the patient's limb movement, such as "perform shoulder joint activities daily, and the range of motion is 0°-120°", are extracted from the medical record text.Then, the trauma limb movement doctor's order data is corresponded with the joint range of motion data to compare whether the patient's actual joint range of motion meets the doctor's order requirements, and the medical record - limb movement data is obtained. The imaging - healing correspondence data, the test - medication fusion data, and the medical record - limb movement data are fused. The Bayesian fusion method is used to calculate the fusion weights according to the credibility and relevance of each data source, and the data is weighted and fused. For example, the wound healing process in the imaging - healing correspondence data is associated with the drug use effect in the test - medication fusion data, combined with the movement situation in the medical record - limb movement data, and the trauma recovery status of the patient is comprehensively evaluated to generate a comprehensive trauma medicine - nursing monitoring data set.

[0041] As an example of the present invention, refer to Figure 2 shown, in this example, the step S3 includes: Step S31: Identify the trauma characteristics of the trauma medicine - nursing monitoring data to obtain trauma characteristic data; extract the manifestation characteristics of the trauma characteristic data to generate trauma manifestation characteristic data; Step S32: Divide the trauma manifestation characteristic data to obtain trauma manifestation division data; pair the trauma type characteristics based on the trauma manifestation division data to generate trauma type characteristic data; Step S33: Evaluate the degree of trauma recovery of the trauma medicine - nursing monitoring data to generate trauma recovery degree data; Step S34: Analyze the trauma nursing relevance of the trauma medicine - nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma nursing association data; Step S35: Divide the trauma nursing association data into trauma nursing knowledge levels to generate trauma nursing knowledge level data.

[0042] In the embodiments of the present invention, a machine learning algorithm is used to perform feature recognition on trauma medicine - nursing monitoring data. The support vector machine (SVM) algorithm is adopted, and the image features (such as wound area, depth), biochemical detection indexes (such as white blood cell count, blood glucose level), and nursing monitoring data (such as wound dressing change frequency, patient pain score) in the data set are used as input features to train the model to identify the feature patterns related to trauma, and trauma feature data is obtained. The trauma feature data is further processed to extract the performance features. The color distribution feature of the wound is extracted from the image features, and the proportion of red and yellow pixels in the wound area is calculated using image processing technology; the features related to the inflammatory response are extracted from the biochemical detection indexes, such as the change trend of white blood cell count; the pain relief situation after wound dressing change is extracted from the nursing monitoring data, etc., to generate trauma performance feature data. Cluster analysis is performed on the trauma performance feature data to achieve characterization division. For example, the K - means clustering algorithm is used to divide the trauma performance feature data into different characterization categories according to the color distribution feature of the wound, inflammatory response index, pain relief situation, etc. The degree of trauma recovery of the trauma medicine - nursing monitoring data is evaluated using the Bayesian fusion method. First, the evaluation indexes of the degree of trauma recovery are defined, such as the wound healing area, the decline range of pain score, etc. Then, according to the credibility and relevance of each data source, the fusion weights are calculated, and the data is weighted and fused. For example, the wound healing area in the image data is fused with the decline range of pain score in the nursing monitoring data to comprehensively evaluate the degree of trauma recovery of the patient, and trauma recovery degree data is generated. Based on the trauma type feature data and the trauma recovery degree data, the correlation analysis of the trauma medicine - nursing monitoring data is carried out. For example, the association rule mining algorithm, such as the Apriori algorithm, is used to analyze the association rules between the trauma type features and nursing measures, find out the association rules between the two, and obtain trauma nursing association data. Hierarchical cluster analysis is performed on the trauma nursing association data to achieve the hierarchical division of trauma nursing knowledge. For example, the AGNES hierarchical clustering algorithm is used to construct a feature matrix according to the features in the trauma nursing association data, such as trauma type, nursing measures, recovery degree, etc., and then hierarchical clustering is carried out to form different knowledge levels. The trauma nursing knowledge is divided into levels such as basic nursing, specialized nursing, and rehabilitation nursing, and each level contains corresponding nursing knowledge and measures, generating trauma nursing knowledge level data.

[0043] As an example of the present invention, refer to Figure 3 shown, in this example, step S33 includes: Step S331: Determine the tissue structure of the trauma healing area for the image - healing corresponding data to obtain the tissue structure of the healing area; identify the integrity of tissue healing for the tissue structure of the healing area, and generate a structural healing integrity index; Step S332: Determine the medication cycle for the detection-medication fusion data to obtain the trauma medication cycle data; map the promoting effect of trauma healing drugs on the tissue structure of the healing area according to the trauma medication cycle data to generate a drug promoting effect index; Step S333: Judge the movable range of the limb for the medical record-limb movement data to obtain the movable range data of the limb; determine the degree of limb function recovery for the movable range of the limb to generate a limb function recovery index; Step S334: Calculate the trauma recovery degree index based on the structural healing integrity index, the drug promoting effect index, and the limb function recovery index to obtain the trauma recovery degree data.

[0044] In the embodiments of the present invention, image processing technology is used to analyze the image-healing corresponding data. For example, a deep convolutional neural network model (such as ResNet) constructed using a deep learning framework is used to extract features and perform semantic segmentation on the image data of the wound area, and different tissue structures within the wound healing area are identified, such as the epidermis, dermis, granulation tissue, etc. Through the convolutional layer and pooling layer of the model, the pixel features of the tissue structure are extracted to obtain the tissue structure of the healing area. Further analysis is performed on the tissue structure of the healing area to identify the integrity of tissue healing. For example, image analysis algorithms are used to calculate the continuity and integrity indicators of the tissue structure in the healing area, such as the smoothness of the tissue edge and the size of the tissue gap. Through these indicators, a structural healing integrity indicator is generated to evaluate the quality of tissue healing. Time series analysis is performed on the detection-medication fusion data to determine the trauma medication cycle. For example, a time series clustering algorithm, such as K-means clustering, is used to cluster the time points in the drug use records, and adjacent drug use time points are divided into a medication cycle. By analyzing the frequency and duration of drug use, trauma medication cycle data is obtained; based on the trauma medication cycle data, a drug promotion effect mapping is performed on the tissue structure of the healing area. For example, within the determined medication cycle, the association between drug use and tissue structure changes is analyzed, and a regression analysis model is used to evaluate the promotion effect of the drug on tissue healing. The time points of drug use are corresponded to the improvement of the tissue structure to generate a drug promotion effect indicator; the medical record-limb activity data is analyzed to determine the range of limb movement. For example, non-contact measurement methods, such as a camera device and a protractor, are used to obtain the image data of the patient's limb movement. Through image processing algorithms, the range of movement angles of the limb joints is calculated to obtain the range of limb movement data; further analysis is performed on the range of limb movement to determine the degree of limb function recovery. For example, the actual range of limb movement is compared with the normal range of movement, and the recovery ratio is calculated. Combining indicators such as the patient's pain score and activity quality, a limb function recovery indicator is generated to evaluate the recovery of limb function; based on the structural healing integrity indicator, the drug promotion effect indicator, and the limb function recovery indicator, a trauma recovery degree indicator is calculated. For example, a weighted average method is used to perform a weighted sum of the structural healing integrity indicator, the drug promotion effect indicator, and the limb function recovery indicator according to the importance weights of each indicator. Through the calculated comprehensive indicator, trauma recovery degree data is obtained to comprehensively evaluate the trauma recovery of the patient.

[0045] Preferably, step S34 includes the following steps: Step S341: Based on the trauma type feature data, perform healing feature nursing mode recognition on the image-healing corresponding data to obtain a healing feature nursing mode; Step S342: Correlate the drug usage of the detection-medication fusion data according to the healing feature care model to obtain drug usage correlation data; Step S343: Extract the recovery contribution degree of the trauma care model from the medical record-limb movement data based on the trauma recovery degree data to obtain the care model contribution degree; Step S344: Sort the contribution degrees of the care model to generate a recovery contribution degree sorting value; Correlate the trauma medicine-nursing monitoring data according to the recovery contribution degree sorting value to obtain trauma care effect data; Step S345: Combine the healing feature care model, the drug usage correlation data, and the trauma care effect data through trauma care correlation to obtain trauma care correlation data.

[0046] In the embodiments of the present invention, based on the trauma type characteristic data, the image-healing corresponding data is used to identify the healing characteristic nursing mode. Using the convolutional neural network (CNN) model in deep learning, the trauma type characteristic data is used as the input of the model. Combining information such as the tissue structure of the wound healing area and the structural healing integrity index in the image-healing corresponding data, the model is trained to identify the healing characteristic nursing modes under different trauma types. Through the convolutional layer and pooling layer of the model, the nursing mode features related to the healing characteristics are extracted to obtain the healing characteristic nursing mode; according to the healing characteristic nursing mode, the drug usage situation in the detection-medication fusion data is associated. The Apriori algorithm, such as the Apriori algorithm, is used to perform an association analysis between the key features in the healing characteristic nursing mode and the drug usage records in the detection-medication fusion data. The drug usage patterns related to the wound healing effect under a specific healing characteristic nursing mode are found, such as the association between the usage frequency of antibiotics and the growth rate of granulation tissue, to obtain the drug usage association data; based on the trauma recovery degree data, the contribution of the trauma nursing mode to the recovery in the medical record-limb movement data is extracted. Using the regression analysis method, the trauma recovery degree data is used as the dependent variable, and the nursing modes (such as joint movement training, muscle strength training, etc.) in the medical record-limb movement data are used as the independent variables to establish a regression model. Through model analysis, the contribution of different nursing modes to the trauma recovery degree is calculated to obtain the nursing mode contribution degree; the nursing mode contribution degrees are sorted using a sorting algorithm, such as the quicksort or merge sort algorithm. According to the numerical values of the contribution degrees, the nursing modes are sorted to generate the recovery contribution degree sorting values; according to the recovery contribution degree sorting values, the nursing effect in the trauma medicine-nursing monitoring data is associated. The nursing modes ranked higher are associated with the relevant indicators (such as wound healing time, pain score, etc.) in the trauma medicine-nursing monitoring data to find out the specific impact of the nursing mode on the nursing effect, to obtain the trauma nursing effect data; the healing characteristic nursing mode, the drug usage association data, and the trauma nursing effect data are merged using data fusion techniques, such as weighted average or Bayesian fusion methods. According to the credibility and relevance of each data source, the fusion weights are calculated, and the nursing mode features in the healing characteristic nursing mode, the drug usage patterns in the drug usage association data, and the nursing effect indicators in the trauma nursing effect data are weighted and fused to generate the trauma nursing association data.

[0047] Preferably, step S35 includes the following steps: Step S351: Identify the healing characteristic stage for the trauma nursing association data to obtain the healing characteristic stage; divide the trauma nursing mode hierarchy for the healing characteristic nursing mode according to the healing characteristic stage to generate the trauma nursing mode hierarchy; Step S352: Identify the types of nursing drugs for the trauma care - related data to obtain the types of nursing drugs; extract the dosages of nursing drugs from the trauma care - related data to obtain the dosages of nursing drugs; determine the nursing medication cycles for the trauma care - related data to obtain the nursing medication cycles. Step S353: Classify the trauma care drugs hierarchically based on the types of nursing drugs, the dosages of nursing drugs, and the nursing medication cycles to generate the trauma care drug hierarchy. Step S354: Classify the trauma care effects hierarchically for the trauma care - related data to generate the trauma care effect hierarchy. Step S355: Classify the trauma care model hierarchy, the trauma care drug hierarchy, and the trauma care effect hierarchy to obtain the trauma care knowledge hierarchy data.

[0048] In the embodiments of the present invention, time series analysis technology is used to analyze the image-healing corresponding data in the trauma care-related data. For example, the image data during the wound healing process is arranged in chronological order, and different stages of wound healing are identified through the dynamic time warping (DTW) algorithm; according to the identified healing characteristic stages, hierarchical division of the healing characteristic nursing modes is carried out. For example, the nursing mode during the inflammatory stage is divided into the first-level nursing mode, which mainly focuses on wound cleaning and anti-infection measures; the nursing mode during the proliferation stage is divided into the second-level nursing mode, with the focus on promoting the growth of granulation tissue and wound contraction. Text mining is performed on the trauma care-related data to extract the type information of nursing drugs. For example, the named entity recognition (NER) algorithm in natural language processing (NLP) technology is used to identify drug names from the nursing record text. The drug dosage information is extracted from the nursing drug usage records. For example, by parsing the dosage description in the drug usage record, such as "100 mg each time", the dosage information is associated with the drug type to obtain the nursing drug dosage; the time stamp of the drug usage record is analyzed to determine the nursing medication cycle. The time series clustering algorithm is used to divide the time period of continuous use of a certain drug into a medication cycle; according to the nursing drug type, nursing drug dosage, and nursing medication cycle, hierarchical division of trauma care drugs is carried out. For example, antibiotic drugs are divided into first-level drugs (such as low dose, short cycle) and second-level drugs (such as high dose, long cycle) according to different dosages and medication cycles, forming a trauma care drug hierarchy. The trauma care-related data is analyzed to extract trauma care effect indicators, such as wound healing time, pain score, etc., and hierarchical division is carried out according to the numerical range of the indicators. For example, the wound healing time is divided into three levels: short, medium, and long, and the pain score is divided into three levels: mild, medium, and severe, generating a trauma care effect hierarchy. The trauma care mode hierarchy, trauma care drug hierarchy, and trauma care effect hierarchy are integrated to form trauma care knowledge hierarchy data. For example, the first-level nursing mode is associated with the corresponding drug hierarchy and nursing effect hierarchy to construct a comprehensive knowledge hierarchy structure including nursing mode, drug use, and effect evaluation.

[0049] Preferably, step S4 includes the following steps: Step S41: Perform inter-layer similarity measurement on the trauma care knowledge hierarchy data to obtain an inter-layer similarity measurement value; extract the inter-layer association strength from the inter-layer similarity measurement value to obtain the inter-layer association strength; Step S42: Determine the inter-layer similarity coefficient for the trauma care knowledge hierarchy data according to the inter-layer association strength to generate a hierarchical similarity coefficient; perform similar layer marking on the trauma care knowledge hierarchy data based on the hierarchical similarity coefficient to obtain similar knowledge layers; cluster and associate the similar knowledge layers to form hierarchical clustering data; Step S43: Extract knowledge nodes from the hierarchical clustering data to obtain trauma care knowledge nodes; determine the hierarchical structure of the trauma care knowledge nodes to obtain node hierarchical structure data; determine the logical relationship of the trauma care knowledge nodes to obtain node logical relationships; Step S44: Based on the node hierarchical structure data and node logical relationships, construct an association relationship for the hierarchical clustering data to obtain a trauma care knowledge system.

[0050] In an embodiment of the present invention, a network similarity comparison method based on high-order information is used to measure the similarity of different levels in the trauma care knowledge hierarchical data. Calculate the node high-order clustering coefficient distribution and node distance distribution between each level, and then according to the similarity calculation formula, obtain the inter-layer similarity measurement value; extract the association strength from the inter-layer similarity measurement value. For example, set a strength threshold, such as the strength threshold is 10; when the similarity measurement value is higher than this threshold, it is considered that there is a strong association strength between the two levels, so as to obtain the inter-layer association strength. According to the inter-layer association strength, calculate the hierarchical similarity coefficient. For example, use methods such as cosine similarity or Pearson correlation coefficient to convert the inter-layer association strength into a similarity coefficient to generate a hierarchical similarity coefficient; based on the hierarchical similarity coefficient, perform similar layer marking on the trauma care knowledge hierarchical data. Mark the levels with higher similarity coefficients as similar knowledge layers, and then use a hierarchical clustering algorithm (such as the AGNES algorithm) to cluster and associate them according to the similarity coefficients between the similar layers to form hierarchical clustering data. Analyze the hierarchical clustering data to extract trauma care knowledge nodes. For example, take the key knowledge points in each cluster as knowledge nodes, such as "wound cleaning", "antibiotic use", etc.; according to the positional relationship of the knowledge nodes in the hierarchical clustering data, determine the node hierarchical structure data. For example, take the "wound cleaning" node as a node at the basic care level and the "antibiotic use" node as a node at the drug care level; analyze the logical relationship between the knowledge nodes. For example, use a directed graph to represent the causal relationship or sequence relationship between the nodes, such as the "wound cleaning" node pointing to the "antibiotic use" node, indicating that wound cleaning is a prerequisite for antibiotic use, to obtain node logical relationships; based on the node hierarchical structure data and node logical relationships, construct an association relationship for the hierarchical clustering data. For example, associate the nodes with the same hierarchical structure to form a hierarchical knowledge structure; at the same time, according to the node logical relationships, connect the nodes with causal relationships or sequence relationships to construct a complete trauma care knowledge system.

[0051] In this specification, a knowledge mining system for the field of trauma care medicine is provided, which is used to execute the above-mentioned knowledge mining method for the field of trauma care medicine. The knowledge mining system for the field of trauma care medicine includes: The trauma medicine data acquisition module is used to obtain the original trauma medical data; perform data preprocessing on the original trauma medical data to obtain the trauma medical data; perform standardization processing on the trauma medical data to obtain the standard trauma medical data; The trauma medicine-nursing monitoring integration module is used to obtain the nursing monitoring data of trauma patients; classify the monitoring types of the nursing monitoring data of trauma patients to obtain the nursing monitoring type data; fuse the standard trauma medical data and the nursing monitoring type data to generate the trauma medicine-nursing monitoring data; The trauma nursing knowledge level classification module is used to determine the trauma type characteristics of the trauma medicine-nursing monitoring data to generate the trauma type characteristic data; evaluate the trauma recovery degree of the trauma medicine-nursing monitoring data to generate the trauma recovery degree data; perform trauma nursing relevance analysis on the trauma medicine-nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain the trauma nursing association data; classify the trauma nursing association data into trauma nursing knowledge levels to generate the trauma nursing knowledge level data; The trauma nursing knowledge system construction module is used to perform hierarchical clustering on the trauma nursing knowledge level data to obtain the hierarchical clustering data; construct the trauma nursing knowledge system from the hierarchical clustering data to generate the trauma nursing knowledge system.

[0052] Through the trauma medical data acquisition module, the original trauma medical data is obtained and subjected to data preprocessing and standardization processing, which can ensure the accuracy and consistency of trauma medical data. This lays a solid foundation for subsequent data analysis and applications, and improves the usability and reliability of the data. Through the trauma medicine-nursing monitoring integration module, the nursing monitoring data of trauma patients is obtained and the monitoring types are classified, which can effectively classify and organize the nursing monitoring data. The standard trauma medical data and the nursing monitoring type data are fused to generate trauma medicine-nursing monitoring data, realizing the organic combination of medical data and nursing monitoring data. Through the trauma nursing knowledge level division module, the trauma type characteristics are determined and the trauma recovery degree is evaluated for the trauma medicine-nursing monitoring data, which can accurately identify the trauma type and evaluate the trauma recovery situation. Based on the trauma type characteristic data and the trauma recovery degree data, the trauma nursing relevance analysis is carried out on the trauma medicine-nursing monitoring data, revealing the internal connection between trauma nursing and patient recovery. The trauma nursing knowledge level division is carried out on the trauma nursing correlation data to generate trauma nursing knowledge level data, providing a basis for constructing a systematic trauma nursing knowledge system. Through the trauma nursing knowledge system construction module, hierarchical clustering is carried out on the trauma nursing knowledge level data, which can classify and integrate similar nursing knowledge to form a clearer and more systematic knowledge structure. The hierarchical clustering data is used to construct a trauma nursing knowledge system, generating a trauma nursing knowledge system, providing systematic and structured knowledge support for the practice and research of trauma nursing, and improving the quality and efficiency of trauma nursing. Therefore, through data processing technology, pattern recognition technology and deep learning technology, the present invention realizes the fusion of trauma nursing medical data for trauma medical text information, image information and nursing monitoring information, and realizes the division of relevant knowledge levels of the fused trauma nursing medical data; thus making the knowledge system in the field of trauma nursing medicine more comprehensively presented.

[0053] Therefore, in any aspect, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to cover all changes falling within the meaning and scope of the equivalent elements of the application documents within the present invention.

[0054] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A knowledge mining method for the field of trauma care medicine, characterized in that: The following steps are involved: Step S1: obtaining original trauma medical data; performing data preprocessing on the original trauma medical data to obtain trauma medical data; Standardize trauma medical data to obtain standard trauma medical data; Step S2: Obtaining nursing monitoring data of trauma patients; classifying the nursing monitoring data of trauma patients into monitoring types to obtain nursing monitoring type data; Fusing standard trauma medicine data with nursing monitoring type data to generate trauma medicine-nursing monitoring data; Step S3: determining trauma type characteristics of the trauma medicine-nursing monitoring data to generate trauma type characteristic data; evaluating the degree of trauma recovery of the trauma medicine-nursing monitoring data to generate trauma recovery degree data; performing trauma nursing correlation analysis on the trauma medicine-nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma nursing correlation data; dividing the trauma nursing knowledge level of the trauma nursing correlation data to generate trauma nursing knowledge level data; Step S4: performing hierarchical clustering on the trauma nursing knowledge hierarchical data to obtain hierarchical clustering data; The hierarchical clustering data is used to construct a trauma nursing knowledge system to generate a trauma nursing knowledge system.

2. The knowledge mining method for the field of trauma care medicine according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire original medical data of trauma, wherein the original medical data of trauma includes imaging examination data of trauma patients, biochemical test data of trauma patients and electronic medical record data of trauma patients; Step S12: performing image enhancement on the trauma patient imaging data to obtain trauma imaging enhancement data; performing data calibration on the trauma patient biochemical detection data to obtain biochemical detection calibration data; performing text cleaning on the trauma patient electronic medical record data to obtain electronic medical record standard data; Step S13: integrating trauma image enhancement data, biochemical detection calibration data and electronic medical record standard data to obtain trauma medical data; Step S14: Standardize the trauma medical data in a corresponding format to generate medical format standardized data; standardize and encode the trauma medical data to obtain standard trauma medical data.

3. The knowledge mining method for the field of trauma care medicine according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: monitor the patient's wound healing status; obtain the patient's medication usage; and record the patient's injured limb activity; Step S22: performing wound healing feature identification on the patient's wound healing status to generate a wound healing feature type; performing wound medication status determination on the patient's medication usage to generate a wound medication status type; performing trauma limb activity type classification on the patient's trauma limb activity status to generate a trauma limb activity type; Step S23: classifying the nursing monitoring data of trauma patients according to the wound healing characteristic type, the wound medication type and the trauma limb activity type to obtain nursing monitoring type data; Step S24: Fusing the standard trauma medicine data with the nursing monitoring type data to generate trauma medicine-nursing monitoring data.

4. The knowledge mining method for the field of trauma care medicine according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: performing wound appearance feature recognition on the wound healing feature type to obtain wound appearance feature data; performing wound healing progress judgment on the wound appearance feature data to generate wound healing progress data; Step S242: Corresponding the wound image enhancement data with the wound appearance feature data and the wound healing process data to the wound healing state, and obtaining image-healing correspondence data; Step S243: determining the wound medication type for the wound medication situation type to obtain wound medication type data; recording the wound disinfection frequency for the wound medication type data to obtain the wound disinfection frequency; extracting the medication dosage for the trauma patient's medication usage according to the wound disinfection frequency to generate the wound medication dosage; Step S244: Matching and fusing the biochemical test calibration data with the wound medication type data, wound disinfection frequency, and wound medication dosage to obtain test-medication fusion data; Step S245: marking the limb joints of the injured limb activity type to generate injured limb joint data; determining the joint activity range of the injured limb joint data to obtain joint activity range data; Step S246: extracting the medical instructions for traumatic limb movement from the electronic medical record standard data to obtain the medical instructions for traumatic limb movement data; matching the medical instructions for traumatic limb movement data with the joint range of motion data to obtain the medical record-limb movement data; Step S247: Fuse the image-healing correspondence data, the detection-drug fusion data and the medical record-limb movement data to generate trauma medicine-nursing monitoring data.

5. The knowledge mining method for the field of trauma care medicine according to claim 4, characterized in that: Step S3 includes the following steps: Step S31: performing trauma feature recognition on the trauma medicine-nursing monitoring data to obtain trauma feature data; performing performance feature extraction on the trauma feature data to generate trauma performance feature data; Step S32: characterizing and dividing the trauma manifestation characteristic data to obtain trauma characterization and division data; performing trauma type characteristic pairing based on the trauma characterization and division data to generate trauma type characteristic data; Step S33: performing trauma recovery degree assessment on the trauma medicine-nursing monitoring data to generate trauma recovery degree data; Step S34: performing trauma nursing correlation analysis on the trauma medicine-nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma nursing correlation data; Step S35: dividing the trauma care related data into trauma care knowledge levels to generate trauma care knowledge level data.

6. The knowledge mining method for the field of trauma care medicine according to claim 5, characterized in that: Step S33 includes the following steps: Step S331: determining the tissue structure of the wound healing region based on the image-healing correspondence data to obtain the tissue structure of the healing region; identifying the tissue healing integrity of the healing region tissue structure to generate a structural healing integrity index; Step S332: determining the medication cycle of the detection-medication fusion data to obtain wound medication cycle data; mapping the wound healing drug promotion effect on the healing area tissue structure according to the wound medication cycle data to generate a drug promotion effect index; Step S333: judging the range of limb movement based on the medical record-limb movement data to obtain limb movement range data; determining the degree of limb function recovery based on the limb movement range to generate a limb function recovery index; Step S334: Calculate the trauma recovery degree index based on the structural healing integrity index, the drug promotion effect index and the limb function recovery index to obtain the trauma recovery degree data.

7. The knowledge mining method for the field of trauma care medicine according to claim 5, characterized in that: Step S34 includes the following steps: Step S341: performing healing characteristic nursing mode recognition on the image-healing corresponding data based on the wound type characteristic data to obtain the healing characteristic nursing mode; Step S342: Correlate the detection-medication fusion data with the drug usage according to the healing characteristic nursing mode to obtain drug usage correlation data; Step S343: extracting the trauma nursing mode recovery contribution from the medical record-limb activity data based on the trauma recovery degree data to obtain the nursing mode contribution; Step S344: sorting the nursing mode contributions to generate a recovery contribution ranking value; correlating the trauma medicine-nursing monitoring data with the nursing effect according to the recovery contribution ranking value to obtain trauma nursing effect data; Step S345: The healing characteristic nursing pattern, the drug use associated data and the wound nursing effect data are combined for wound nursing association to obtain wound nursing associated data.

8. The knowledge mining method for the field of trauma care medicine according to claim 7, characterized in that: Step S35 includes the following steps: Step S351: performing healing characteristic stage identification on the wound care associated data to obtain the healing characteristic stage; dividing the healing characteristic nursing mode into wound care mode hierarchies according to the healing characteristic stage to generate a wound care mode hierarchy; Step S352: identifying the type of nursing medicine for the wound nursing related data to obtain the type of nursing medicine; extracting the dosage of nursing medicine for the wound nursing related data to obtain the dosage of nursing medicine; determining the nursing medication cycle for the wound nursing related data to obtain the nursing medication cycle; Step S353: dividing the wound nursing medicine hierarchy according to the nursing medicine type, nursing medicine dosage and nursing medication cycle to generate the wound nursing medicine hierarchy; Step S354: dividing the wound care related data into wound care effect levels to generate a wound care effect level; Step S355: dividing the trauma care model level, the trauma care drug level and the trauma care effect level into trauma care knowledge levels to obtain trauma care knowledge level data.

9. The knowledge mining method for the field of trauma care medicine according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: performing inter-layer similarity measurement on the trauma nursing knowledge hierarchy data to obtain an inter-layer similarity measurement value; performing inter-layer association strength extraction on the inter-layer similarity measurement value to obtain an inter-layer association strength; Step S42: determining the inter-layer similarity coefficient of the trauma nursing knowledge hierarchy data according to the inter-layer association strength to generate the hierarchical similarity coefficient; marking the trauma nursing knowledge hierarchy data with similar layers based on the hierarchical similarity coefficient to obtain similar knowledge layers; clustering and associating the similar knowledge layers to form hierarchical clustering data; Step S43: extracting knowledge nodes from the hierarchical clustering data to obtain trauma care knowledge nodes; determining the hierarchical structure of the trauma care knowledge nodes to obtain node hierarchical structure data; determining the logical relationship of the trauma care knowledge nodes to obtain node logical relationship; Step S44: constructing association relationships for the hierarchical clustering data based on the node hierarchy data and the node logical relationships to obtain a trauma care knowledge system.

10. A knowledge mining system for trauma care medicine, characterized in that: Used to execute the knowledge mining method for the field of wound care medicine as claimed in claim 1, the knowledge mining system for the field of wound care medicine comprises: The trauma medical data acquisition module is used to obtain the original trauma medical data; perform data preprocessing on the original trauma medical data to obtain trauma medical data; perform standardization on the trauma medical data to obtain standard trauma medical data; The trauma medicine-nursing monitoring fusion module is used to obtain nursing monitoring data of trauma patients; classify the nursing monitoring data of trauma patients into monitoring types to obtain nursing monitoring type data; fuse the standard trauma medicine data with the nursing monitoring type data to generate trauma medicine-nursing monitoring data; The trauma nursing knowledge level division module is used to determine the trauma type characteristics of the trauma medicine-nursing monitoring data and generate trauma type characteristic data; to evaluate the degree of trauma recovery of the trauma medicine-nursing monitoring data and generate trauma recovery degree data; to perform trauma nursing correlation analysis on the trauma medicine-nursing monitoring data based on the trauma type characteristic data and the trauma recovery degree data to obtain trauma nursing correlation data; to divide the trauma nursing knowledge level of the trauma nursing correlation data and generate trauma nursing knowledge level data; The trauma nursing knowledge system construction module is used to perform hierarchical clustering on the trauma nursing knowledge hierarchy data to obtain hierarchical clustering data; and construct the trauma nursing knowledge system using the hierarchical clustering data to generate the trauma nursing knowledge system.

Images