Electronic management method and system for sick and wounded information

Abstract

The invention provides a sick and wounded information electronic management method and system. The method comprises the steps of receiving basic information and detailed medical records of the sick and wounded, and generating a comprehensive medical data set; based on the comprehensive medical data set, performing multilevel information analysis and entity association processing by applying a hypergraph neural network algorithm, dynamically adjusting importance weight of each information node by adopting a self-adaptive attention mechanism technology through hyperedge structure modeling high-order interaction, and generating a precise medical information model; based on the precise medical information model, accelerating optimization of a personalized treatment scheme selection process of the sick and wounded, performing efficient classification and regression analysis by adopting a quantum random forest method, and generating personalized treatment scheme suggestions; and based on the personalized treatment scheme suggestions, developing data backup and recovery functions into an integrated health file management system, and sending the integrated health file management system to a third-party system. According to the technical scheme provided by the invention, the medical information management efficiency and precision are improved, and the medical service level is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of medical information technology, and in particular to a method and system for electronic management of patient information. Background Art

[0002] With the rapid development of medical informatization, patient information management plays an increasingly important role in improving the quality and efficiency of medical services. In order to better support clinical decision-making and personalized treatment, an efficient, accurate and secure electronic management system for patient information is needed. In addition, the medical information system must be able to handle complex medical information interactions and provide personalized treatment plan recommendations for each patient. Finally, the system also needs to have powerful data backup and recovery functions to ensure data integrity and availability, and meet the needs of long-term storage and instant access.

[0003] At present, most medical information management systems mainly rely on traditional database technology and statistical analysis methods to process and analyze information about the injured and sick. These systems usually include the following steps: first, receiving data from different sources, performing preliminary data cleaning and standardization; then, using simple machine learning algorithms to classify and predict medical data; and finally, generating treatment plan recommendations manually or semi-automatically. Although these methods have improved the level of medical information management to a certain extent, they still have many limitations.

[0004] There are many defects in the existing medical information management system. Traditional methods are difficult to effectively capture and analyze the complex high-order interactions between medical information, resulting in low efficiency and accuracy of medical information management, which affects the selection of subsequent treatment plans. In addition, the existing personalized treatment plan selection process is often time-consuming and cannot fully utilize advanced computing resources to accelerate optimization. More importantly, due to the lack of an effective data backup and recovery mechanism, once data is lost or damaged, it will seriously affect the continuity and reliability of medical services. Therefore, there is an urgent need for a more intelligent, efficient and secure electronic management method for the information of the wounded and sick to comprehensively improve the effectiveness of medical information management and personalized treatment. Summary of the invention

[0005] The embodiments of the present application provide a method and system for electronic management of patient information, which is used to solve the problem of low efficiency and accuracy of medical information management in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a method for electronic management of information of the injured and sick, including:

[0007] Receive basic information and detailed medical records of the injured and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set;

[0008] Based on the comprehensive medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and a hyperedge structure is used to model the high-order interaction between the medical information of the injured and the sick. The adaptive attention mechanism technology is used to dynamically adjust the importance weight of each information node to generate a precise medical information model.

[0009] Based on the precision medical information model, the quantum enhanced learning algorithm is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick through the efficient parallel processing capabilities of quantum computing, and the quantum random forest method is used for efficient classification and regression analysis to generate personalized treatment plan recommendations;

[0010] Based on the personalized treatment plan recommendation, develop data backup and recovery functions to ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendation to a third-party system.

[0011] Optionally, based on the comprehensive medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, a high-order interaction between medical information of the injured and sick is modeled through a hyperedge structure, and an adaptive attention mechanism technology is used to dynamically adjust the importance weight of each information node to generate a precise medical information model, including:

[0012] Based on the comprehensive medical data set, data cleaning and standardization are performed to remove irrelevant information and noise, ensure that the data has a high degree of consistency, and generate a clean medical data set;

[0013] Based on the clean medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and a preliminary medical information model is generated by modeling high-order interactions between medical information of the injured and sick through a hyperedge structure;

[0014] Based on the preliminary medical information model, an adaptive attention mechanism technology is used to evaluate the importance, dynamically adjust the importance weight of each information node, enhance the attention to key medical information, and generate an optimized medical information model;

[0015] Based on the optimized medical information model, the model parameters are further adjusted to ensure that the model accurately captures the internal connections and patterns of the medical information of the injured and sick, and generates a precise medical information model.

[0016] Optionally, based on the clean medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and a high-order interaction between the medical information of the injured and the sick is modeled through a hyperedge structure to generate a preliminary medical information model, including:

[0017] Based on the clean medical data set, preprocess the text and structured data, extract key features, construct initial representations of nodes and edges, and generate node feature vectors;

[0018] Based on the node feature vectors, a hypergraph neural network algorithm is used to perform multi-layer information transmission on the medical information of the injured and sick, and the complex dependency relationship between different medical entities is captured through the hyperedge structure to generate an intermediate layer feature representation;

[0019] Based on the intermediate layer feature representation, a hypergraph convolution operation is used to further strengthen the interaction between nodes, context information is introduced to enhance the model's understanding ability, and an enhanced feature representation is generated;

[0020] Based on the enhanced feature representation, all levels of information are aggregated to form a complete medical information map of the injured and sick, and generate a preliminary medical information model.

[0021] Optionally, based on the preliminary medical information model, an adaptive attention mechanism technology is used to perform importance assessment, dynamically adjust the importance weight of each information node, enhance the attention to key medical information, and generate an optimized medical information model, specifically including:

[0022] Based on the preliminary medical information model, feature extraction is performed on each entity and its relationship in the medical information of the injured and sick to generate an entity feature vector;

[0023] Based on the entity feature vector, the importance of each entity is evaluated using the adaptive attention mechanism technology, a relative attention score is obtained, and an attention weight vector is generated;

[0024] Based on the attention weight vector, dynamically adjust the importance weight of each information node, weaken the influence of irrelevant information, and generate a weighted medical information model;

[0025] Based on the weighted medical information model, the model parameters are further corrected to ensure accurate reflection of the key features and internal connections of the medical information of the injured and sick, thereby generating an optimized medical information model.

[0026] Optionally, based on the precision medical information model, a quantum enhanced learning algorithm is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick through the efficient parallel processing capabilities of quantum computing, and a quantum random forest method is used for efficient classification and regression analysis to generate personalized treatment plan recommendations, including:

[0027] Based on the precision medical information model, deeply analyze the patient's condition and historical treatment data, extract key medical indicators and treatment response characteristics, and generate detailed medical feature vectors;

[0028] Based on the detailed medical feature vector, a quantum enhanced learning algorithm is used to construct a treatment plan search space, and the optimal treatment path is quickly explored through the efficient parallel processing capability of quantum computing to generate a preliminary treatment plan set;

[0029] Based on the preliminary treatment plan set, the quantum random forest method is used to classify and regress the effects of different treatment plans, evaluate potential effects and risks, and generate a treatment plan evaluation report;

[0030] Based on the treatment plan evaluation report, a personalized treatment plan recommendation is generated by comprehensively considering the specific conditions and individual preferences of the injured and sick.

[0031] Optionally, based on the detailed medical feature vector, a quantum enhanced learning algorithm is used to construct a treatment plan search space, and the optimal treatment path is quickly explored through the efficient parallel processing capability of quantum computing to generate a preliminary treatment plan set, including:

[0032] Based on the detailed medical feature vector, quantitatively analyze the historical treatment response, determine the key influencing factors of the treatment effect, and generate a treatment-related feature matrix;

[0033] Based on the treatment-related feature matrix, a quantum enhanced learning algorithm is used to initialize the treatment plan search space, define potential treatment options and parameter ranges, consider the possibility of combining multiple treatment plans through the principle of quantum state superposition, and generate an initial treatment plan candidate pool;

[0034] Based on the initial treatment plan candidate pool, using the efficient parallel processing capability of quantum computing, multiple rounds of iterative optimization processes are performed to perform step-by-step screening and generate an optimized treatment plan list;

[0035] Based on the optimized treatment plan list, the specific implementation steps and expected results are further refined to ensure the feasibility and effectiveness of the plan, and generate a preliminary treatment plan set.

[0036] Optionally, based on the personalized treatment plan recommendation, a data backup and recovery function is developed to ensure data integrity and availability, and an integrated health record management system is generated, including:

[0037] Based on the personalized treatment plan recommendations, the medical information and treatment plans of the injured and sick are structured to ensure the orderliness and accessibility of the information and generate a standardized medical information database;

[0038] Based on the standardized medical information database, develop an efficient data backup mechanism, automatically perform incremental backup operations regularly, use encryption technology to protect the security of backup data, and generate a secure backup data set;

[0039] Based on the secure backup data set, a fast data recovery process is established to support rapid recovery to the most recent state in case of system failure and generate a data recovery strategy;

[0040] Based on the data recovery strategy, the data management and user authority control functional modules are integrated to generate an integrated health record management system.

[0041] In a second aspect, the present application provides an electronic management system for the information of the injured and sick, including:

[0042] The receiving module is used to receive basic information and detailed medical records of the injured and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set;

[0043] A processing module, which is used to perform multi-level information analysis and entity association processing based on the comprehensive medical data set using a hypergraph neural network algorithm, model the high-order interaction between the medical information of the injured and the sick through a hyperedge structure, and dynamically adjust the importance weight of each information node using an adaptive attention mechanism technology to generate a precise medical information model;

[0044] An analysis module is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick based on the precision medical information model, using a quantum enhanced learning algorithm and the efficient parallel processing capabilities of quantum computing, and to use a quantum random forest method for efficient classification and regression analysis to generate personalized treatment plan recommendations;

[0045] A generation module is used to develop data backup and recovery functions based on the personalized treatment plan recommendation, ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendation to a third-party system.

[0046] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for electronic management of information of the injured and sick as described in the first aspect.

[0047] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a method for electronic management of information of the injured and sick as described in the first aspect.

[0048] In an embodiment of the present application, basic information and detailed medical records of the injured and sick are received from different sources to ensure the comprehensiveness and accuracy of the information and generate a comprehensive medical data set; based on the comprehensive medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and high-order interactions between the medical information of the injured and sick are modeled through hyperedge structure, and the importance weight of each information node is dynamically adjusted using adaptive attention mechanism technology to generate a precise medical information model; based on the precise medical information model, a quantum enhanced learning algorithm is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick through the efficient parallel processing capabilities of quantum computing, and a quantum random forest method is used for efficient classification and regression analysis to generate personalized treatment plan recommendations; based on the personalized treatment plan recommendations, a data backup and recovery function is developed to ensure data integrity and availability, an integrated health record management system is generated, and the personalized treatment plan recommendations are sent to a third-party system. By integrating data from different sources, the comprehensiveness and accuracy of the information is ensured, and a comprehensive medical data set is generated, which not only improves the quality and reliability of the data, but also lays a solid foundation for subsequent advanced analysis; by using advanced technologies such as hypergraph neural network algorithms and quantum enhanced learning algorithms, in-depth analysis of medical information and optimization of personalized treatment options are achieved; finally, by developing data backup and recovery functions, the integrity and availability of the data are guaranteed, and an integrated health record management system is generated.

[0049] Furthermore, through data cleaning and standardization, a high degree of data consistency is ensured, the interference of irrelevant information and noise is eliminated, and the data quality and the reliability of analysis results are greatly improved; the adaptive attention mechanism technology is used to dynamically adjust the importance weight of each information node, thereby enhancing the model's attention to key medical information, thereby greatly improving the accuracy and explanatory power of the model, providing strong support for personalized medicine.

[0050] Furthermore, by deeply analyzing the conditions of the injured and patients and historical treatment data, key medical indicators and treatment response characteristics were extracted, and detailed medical feature vectors were generated, ensuring the scientificity and rationality of personalized treatment plans; the efficient parallel processing capabilities of quantum computing were used to quickly explore the optimal treatment path, significantly shortening the time for selecting personalized treatment plans; the effects and risks of different treatment plans were evaluated through the quantum random forest method, and a treatment plan evaluation report was generated, further improving the effectiveness and safety of the treatment plan; finally, taking into account the specific conditions and personal preferences of the injured and patients, the generated personalized treatment plan recommendations are more in line with actual needs, greatly improving the treatment effect and patient satisfaction.

[0051] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 A flowchart of a method for electronic management of sick and injured information provided in an embodiment of the present application;

[0054] Figure 2 A schematic diagram of the structure of an electronic management system for the information of the injured and sick provided in an embodiment of the present application;

[0055] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0057] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0058] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0059] Figure 1 A flowchart of a method for electronic management of patient information is provided for the present application embodiment, such as Figure 1 As shown, the method includes:

[0060] 101. Receive basic information and detailed medical records of the injured and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set;

[0061] In this step, the comprehensive medical data set refers to the integration of basic information and detailed medical records of the injured and sick from multiple medical institutions, aiming to provide a comprehensive and accurate data foundation.

[0062] The basic information of the injured and sick includes but is not limited to name, age, gender, contact information, address, medical history, allergy history, etc. This information is used to identify the injured and sick and understand their basic health status.

[0063] Detailed medical records include diagnosis results, treatment process, medication records, examination reports, surgical records, nursing records, etc., providing detailed medical history and current treatment status. These records are crucial for understanding the specific condition of the injured and sick and its development and changes.

[0064] In the embodiment of the present application, it is assumed that efficient information management of the injured and sick needs to be achieved in a multi-source data environment; first, the system collects basic information and detailed medical records of the injured and sick from multiple hospitals and clinics (i.e., multiple different sources); second, these scattered data sources are merged into a unified data set through data integration tools; third, data cleaning algorithms are applied to remove duplicate or erroneous data entries to ensure that the information of each injured and sick is accurate; finally, standardization technology is used to unify the format and verify the consistency of the data to generate a high-quality comprehensive medical data set, providing a solid foundation for subsequent analysis.

[0065] 102. Based on the comprehensive medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, a hyperedge structure is used to model the high-order interaction between the medical information of the injured and the sick, and an adaptive attention mechanism technology is used to dynamically adjust the importance weight of each information node to generate a precise medical information model;

[0066] In this step, the hypergraph neural network algorithm is an advanced machine learning method that is particularly suitable for processing complex relationships and high-order interactions. It can parse multi-level information and perform entity association processing, capturing many-to-many relationships that traditional graph neural networks cannot express.

[0067] Multi-level information analysis refers to the extraction and analysis of medical information from different levels (such as symptoms, diseases, treatments, etc.). This analysis not only considers the information at a single level, but also focuses on the mutual influence between different levels, thereby providing a more comprehensive understanding.

[0068] Entity association processing refers to identifying and associating the relationships between different medical entities (such as drugs, diseases, symptoms, etc.). In this way, we can better understand how various factors work together to affect the health status of the injured and sick.

[0069] Hyperedge structure is a connecting element in a hypergraph, which is used to model the interaction between multiple nodes. Unlike the one-to-one or one-to-many relationship in traditional graphs, hyperedge can represent many-to-many relationship, which is suitable for describing complex medical information interaction.

[0070] Higher-order interactions refer to the complex interplay between multiple medical entities, such as the synergistic effects or conflicts between multiple drugs and the causal relationships between different diseases. These interactions are crucial to understanding and predicting the course of disease in patients.

[0071] Adaptive attention mechanism technology is used to dynamically adjust the importance weight of each information node to enhance the focus on key medical information. Through this mechanism, the system can automatically assign different weights to different information nodes according to the specific situation to highlight the most important information.

[0072] The precision medical information model is the result obtained by modeling the data processed in the above steps. The model can accurately reflect the medical information of the injured and sick and their internal connections, and provide a scientific basis for personalized treatment plans.

[0073] In the embodiment of the present application, it is assumed that the depth and accuracy of medical data analysis need to be improved; first, based on a comprehensive medical data set, a hypergraph neural network algorithm is used to parse the multi-level structure in the medical information of the wounded and sick; secondly, a hyperedge structure is used to model the high-order interactions between different medical entities; thirdly, an adaptive attention mechanism technology is used to evaluate and adjust the importance weight of each information node to highlight key information; finally, an accurate medical information model is generated to provide a scientific basis for personalized treatment plans.

[0074] Optionally, the method in step 102 is based on the comprehensive medical data set, uses a hypergraph neural network algorithm to perform multi-level information analysis and entity association processing, models high-order interactions between medical information of the wounded and the sick through a hyperedge structure, and uses an adaptive attention mechanism technology to dynamically adjust the importance weight of each information node to generate a precise medical information model, including: based on the comprehensive medical data set, performing data cleaning and standardization processing to remove irrelevant information and noise, ensure that the data has a high degree of consistency, and generate a clean medical data set; based on the clean medical data set, uses a hypergraph neural network algorithm to perform multi-level information analysis and entity association processing, and uses a hyperedge structure to model high-order interactions between medical information of the wounded and the sick to generate a preliminary medical information model; based on the preliminary medical information model, uses an adaptive attention mechanism technology to perform importance assessment, dynamically adjusts the importance weight of each information node, enhances attention to key medical information, and generates an optimized medical information model; based on the optimized medical information model, further adjusts model parameters to ensure that the model accurately captures the intrinsic connections and patterns of the medical information of the wounded and the sick, and generates a precise medical information model.

[0075] In this step, clean medical data sets refer to high-quality data sets that have been cleaned and standardized to ensure data consistency and reliability. Data cleaning involves removing duplicate, erroneous or incomplete records, while standardization includes unifying data formats, units and coding to eliminate differences between data from different sources.

[0076] Hyperedge structure modeling is to represent many-to-many relationships through hyperedges in a hypergraph. It is suitable for describing complex medical information interactions. Different from the one-to-one or one-to-many relationships in traditional graphs, hyperedges can capture complex interactions between multiple nodes, such as synergies or conflicts between multiple drugs.

[0077] The preliminary medical information model is an initial model generated based on the clean medical data set, which captures the basic pattern of medical information of the injured and sick. Although the model at this stage is not fully optimized, it can already reflect the main characteristics and internal connections of medical information.

[0078] The optimized medical information model further enhances the focus on key medical information. By applying adaptive attention mechanism technology to evaluate and adjust the importance weight of each information node, the model can focus more on factors that have a significant impact on the health status of the injured and sick.

[0079] In the embodiment of the present application, first, data cleaning and standardization are performed based on the comprehensive medical data set to remove irrelevant information and noise, ensure that the data has a high degree of consistency, and generate a clean medical data set; secondly, using the clean medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and the high-order interaction between the medical information of the wounded and the sick is modeled through the hyperedge structure to generate a preliminary medical information model; thirdly, based on the preliminary medical information model, the adaptive attention mechanism technology is used to evaluate the importance of each information node, and the weight is dynamically adjusted to enhance the attention to key medical information and generate an optimized medical information model; finally, the model parameters are further adjusted to ensure that the model can accurately capture the intrinsic connections and patterns of the medical information of the wounded and the sick, and generate a precise medical information model.

[0080] Optionally, based on the clean medical data set, a hypergraph neural network algorithm is used to perform multi-level information parsing and entity association processing, and a high-order interaction between the medical information of the injured and the sick is modeled through a hyperedge structure to generate a preliminary medical information model, including: based on the clean medical data set, text and structured data are preprocessed, key features are extracted, initial representations of nodes and edges are constructed, and node feature vectors are generated; based on the node feature vectors, a hypergraph neural network algorithm is used to perform multi-layer information transmission on the medical information of the injured and the sick, and complex dependencies between different medical entities are captured through a hyperedge structure to generate an intermediate layer feature representation; based on the intermediate layer feature representation, a hypergraph convolution operation is used to further strengthen the interaction between nodes, contextual information is introduced to enhance the model's comprehension ability, and an enhanced feature representation is generated; based on the enhanced feature representation, all levels of information are aggregated to form a complete medical information map for the injured and the sick, and a preliminary medical information model is generated.

[0081] The method, based on the preliminary medical information model, uses an adaptive attention mechanism technology to evaluate the importance, dynamically adjusts the importance weight of each information node, enhances the attention to key medical information, and generates an optimized medical information model, specifically including: based on the preliminary medical information model, extracting features of each entity and its relationship in the medical information of the wounded and sick, and generating an entity feature vector; based on the entity feature vector, using the adaptive attention mechanism technology to evaluate the importance of each entity, obtain a relative attention score, and generate an attention weight vector; based on the attention weight vector, dynamically adjusts the importance weight of each information node, weakens the influence of irrelevant information, and generates a weighted medical information model; based on the weighted medical information model, further corrects the model parameters to ensure accurate reflection of the key features and internal connections of the medical information of the wounded and sick, and generates an optimized medical information model.

[0082] In this step, text and structured data preprocessing refers to the preliminary processing of unstructured text information (such as medical record descriptions) and structured data (such as laboratory test results) from the injured and sick, extracting key features and constructing initial representations of nodes and edges to generate node feature vectors.

[0083] Node feature vectors are generated through the preprocessing step to represent the feature vectors of each medical entity (such as disease, drug, symptom, etc.), which capture the main attributes of each entity and provide a basis for subsequent analysis.

[0084] Multi-layer information transmission refers to the transmission and updating of node features layer by layer through a multi-layer network structure in a hypergraph neural network to capture deeper information. During each layer of transmission, nodes interact through a hyperedge structure to enhance the model's understanding of complex dependencies.

[0085] The intermediate layer feature representation is generated after multiple layers of information transmission, which contains complex dependencies and high-order interactions between nodes. These features provide rich contextual information for subsequent convolution operations.

[0086] The hypergraph convolution operation is a special convolution method suitable for hypergraph structures. It further strengthens the interaction between nodes and introduces contextual information, enabling the model to better understand the complex relationships between medical entities.

[0087] The enhanced feature representation is generated through hypergraph convolution operations and contains richer and more detailed medical information. These features not only reflect the characteristics of the node itself, but also take into account its interaction with other nodes, thus enhancing the expressiveness of the model.

[0088] A complete medical information map of the wounded and sick refers to a comprehensive medical information representation formed by integrating all levels of information. This map not only covers the characteristics of individual medical entities, but also includes the complex interactions between them, providing a scientific basis for personalized treatment.

[0089] Entity feature vectors are vectors generated after feature extraction of various entities and their relationships in the medical information of the injured and sick. These vectors capture the main attributes of each entity and provide a basis for evaluating its importance.

[0090] The attention weight vector is calculated through the adaptive attention mechanism technology to indicate the importance of each information node. These weights can be adjusted dynamically to reduce the impact of irrelevant information and ensure that the model focuses on the most critical data.

[0091] The weighted medical information model refers to a medical information model adjusted according to the attention weight vector. In this way, the model can more accurately reflect the key characteristics and internal connections of the medical information of the injured and sick, and improve the accuracy of prediction and decision-making.

[0092] In the embodiment of the present application, firstly, based on the clean medical data set, the text and structured data are preprocessed, the key features are extracted, the initial representation of nodes and edges is constructed, and the hypergraph neural network algorithm is used to perform multi-layer information transmission on the medical information of the wounded and sick. The complex dependency relationship between different medical entities is captured through the hyperedge structure to generate the intermediate layer feature representation; secondly, the hypergraph convolution operation is used to further strengthen the interaction between nodes, introduce contextual information to enhance the model's understanding ability, aggregate all levels of information, form a complete medical information map of the wounded and sick, and generate a preliminary medical information model; thirdly, based on the preliminary medical information model, the features of each entity and its relationship in the medical information of the wounded and sick are extracted, and the adaptive attention mechanism technology is used to evaluate the importance of each entity, obtain the relative attention score, and generate the attention weight vector; finally, based on the attention weight vector, the importance weight of each information node is dynamically adjusted to reduce the influence of irrelevant information, and the model parameters are further corrected to ensure accurate reflection of the wounded and sick.

[0093] Assume that a large amount of patient information needs to be efficiently managed in a large hospital network environment. First, the system collects text medical records and structured examination reports of patients from multiple departments, performs preprocessing, extracts key features, constructs initial representations of nodes and edges, and uses a hypergraph neural network algorithm to perform multi-layer information transmission on the patient medical information. The complex dependencies between different medical entities are captured through the hyperedge structure to generate intermediate layer feature representations. Secondly, the hypergraph convolution operation is used to further strengthen the interaction between nodes, introduce contextual information to enhance the model's understanding ability, aggregate all levels of information, form a complete patient medical information map, and generate a preliminary medical information model. Thirdly, based on the preliminary medical information model, feature extraction is performed on each entity and its relationship in the patient medical information, and the adaptive attention mechanism technology is used to evaluate the importance of each entity, obtain the relative attention score, and generate the attention weight vector. Finally, the importance weight of each information node is dynamically adjusted to reduce the influence of irrelevant information, and the model parameters are further corrected to ensure accurate reflection of the key features and internal connections of the patient medical information, generate an optimized medical information model, and support doctors to make more accurate diagnoses and personalized treatment plans.

[0094] This application takes into account the problems of low efficiency in medical information processing and inability to effectively capture complex dependencies in the prior art, so the invention embodiment proposes this optional solution to solve the above technical problems. By introducing the hypergraph neural network algorithm and attention mechanism, it is possible to more efficiently analyze the multi-level medical information of the injured and sick, and generate an accurate medical information model, thereby improving the speed and accuracy of selecting personalized treatment plans.

[0095] Optionally, based on the node feature vector, a hypergraph neural network algorithm is used to perform multi-layer information transmission on the medical information of the injured and sick, and the complex dependency relationship between different medical entities is captured through the hyperedge structure to generate an intermediate layer feature representation, including:

[0096] Based on the node feature vectors, a normalization technique is used to ensure that all feature vectors are on the same scale;

[0097] Through the graph embedding method, the node and its neighborhood structure information are encoded into a fixed-length vector representation to capture the local and global relationships between nodes to generate a set of node feature vectors;

[0098] The node feature vector set is calculated using the following formula:

[0099]

[0100] Among them, H (l+1) is the node feature vector set of the l+1th layer; H (l) is the node feature vector set of the lth layer; ε represents the set of hyperedges connected to node v; ε v is the set of all hyperedges related to node v; W (l) is the weight matrix of the lth layer; X ε is the average value of the initial eigenvectors of all nodes in the hyperedge ε; A i is the attention coefficient vector of node i in the hyperedge ε; ⊙ represents the element-by-element multiplication operation; σ is the activation function ReLU or tanh, which is used to introduce nonlinearity; For each node v, traverse all hyperedges ε connected to it and sum them; ∑ i∈ε For each hyperedge ε, traverse all nodes i in the hyperedge and sum them; λ is a learnable parameter that controls the strength of the residual connection; D (l) is the weight matrix of the residual connection;

[0101] Based on the node feature vector set, an attention mechanism is used to evaluate the importance of different nodes, node features are weighted and aggregated, residual connections are introduced to retain original feature information, and different hyperedge features are combined in a nonlinear manner to generate an enhanced feature vector set;

[0102] The enhanced feature vector set is calculated using the following formula:

[0103]

[0104] Among them, Q (l+1) is the set of enhanced feature vectors of the l+1th layer; H (l+1) is the node feature vector set of the l+1th layer; V (l) is an additional weight matrix used to enhance the model's expressiveness; β and α are learnable parameters that control the influence of different items respectively; B (l) Ensure information flow for the residual connection weight matrix; γ i is the adaptive weight coefficient of each node i in the hyperedge ε; represents some complex fusion operation, which is a multi-layer perceptron; ρ is another activation function, which is LeakyReLU or other nonlinear transformation; For each node v, traverse all hyperedges ε connected to it and sum them; ∑ i∈ε For each hyperedge ε, we traverse all nodes i within each hyperedge and sum them; μ is a learnable parameter that controls the strength of differential connections; F (l) is the weight matrix of differential connection; H (l) is the node feature vector set of the lth layer; ε represents the set of hyperedges connected to node v; ε v is the set of all hyperedges related to node v; X ε is the average value of the initial eigenvectors of all nodes in the hyperedge ε; A i is the attention coefficient vector of node i within the hyperedge ε;

[0105] Based on the enhanced feature vector set, batch normalization and principal component analysis methods are applied to perform feature normalization and dimensionality reduction processing, and nonlinearity is increased through activation functions to generate intermediate layer feature representations.

[0106] This method aims to use the hypergraph neural network algorithm to perform multi-layer information transmission on the medical information of the wounded and sick, capture the complex dependencies between different medical entities through the hyperedge structure, and generate intermediate layer feature representations. This method can not only process high-dimensional data, but also enhance the focus on key medical information, ensure that the model accurately reflects the internal connections and patterns of the medical information of the wounded and sick, and provide a scientific basis for personalized treatment.

[0107] In the node feature vector set, the hyperedge interaction term This part is used to calculate the interaction between each node and its neighbors (i.e., all related hyperedges). Traverse all relevant hyperedges, Averaging is performed to reduce the impact of the number of nodes, W (l) Learn the interaction weights between nodes, X Erepresents the average value of the initial eigenvectors of all nodes in the hyperedge, ∑ i∈E Traverse all nodes in the hyperedge, A i Evaluate node importance, ⊙ element-wise multiplication operation to fuse features; residual connection term λ·D (l) ·H (l) : This part introduces residual connection to retain the original feature information, λ controls the strength of residual connection, D (l) is the residual connection weight matrix, H (l) is the set of node feature vectors of the previous layer;

[0108] Among them, H (l+1) ,H (l) Obtained through the output of the previous layer; E v ,E is extracted from the medical entity association relationship; W (l) Optimized through training process; X E Calculate the average value of all relevant node features; A i Dynamically adjusted through adaptive attention mechanism;λ,D (l) Automatic learning during training;

[0109] In the enhanced feature vector set, the current layer feature item H (l+1) : Directly input the node feature vector set of the current layer; This part is used to enhance the feature expression ability, and β controls the influence of the additional weight matrix. Traverse all relevant hyperedges, Square averaging is used to reduce the impact of the number of nodes, V (l) Additional weight matrix, X E represents the average value of the initial eigenvectors of all nodes in the hyperedge, Complex fusion operations, usually multi-layer perceptrons, ∑ i∈E Traverse all nodes in the hyperedge, γ i Adaptive weight coefficient evaluates node importance, A i Evaluate the importance of nodes, ⊙ element-by-element multiplication operation to fuse features; residual retention term α·B (l) ·H (l) : This part introduces residual connection to retain the original feature information, α controls the influence of residual connection, B (l) is the residual connection weight matrix, H (l) is the node feature vector set of the previous layer; the feature change term μF (l) ·(H (l+1) -H (l) ): This part introduces differential connection to maintain feature changes, μ controls the strength of differential connection, F (l) is the differential connection weight matrix, (H (l+1) -H (l)) indicates feature changes;

[0110] Among them, Q (l+1) ,H (l+1) ,H (l) Obtained through the output of the previous layer; E v ,E is extracted from the medical entity association relationship; V (l) ,B (l) ,F (l) Optimized through training process; X E Calculate the average value of all relevant node features; A i ,γ i Dynamically adjusted through adaptive attention mechanism; β, α, μ are automatically learned during training;

[0111] Assume that the feature vector H of the 0th layer node (0) =[0.5,0.7,0.8],W (0) =[0.2,0.3,0.5],X E =[0.6,0.4,0.9],A i =[0.3,0.4,0.3],λ=

[0112] 0.1,D (0) =[0.1,0.2,0.7];

[0113]

[0114] Assume V (0) =[0.4,0.6,0.8],β=0.2,X E =[0.6,0.4,0.9],γ i =[0.5,0.3,0.2],A i =[0.3,0.4,0.3],α=

[0115] 0.3,B (0) =[0.1,0.2,0.7],μ=0.4,F (0) =[0.2,0.3,0.5];

[0116]

[0117] Assuming the threshold is 0.7, since the calculated result Q (1)= Each element of [0.72, 0.81, 0.93] is greater than the set threshold, which shows that the information characteristics of the injured and sick are effectively enhanced, which not only verifies the effectiveness of the model, but also ensures the accuracy of subsequent diagnosis and treatment recommendations. Through the above steps, the system can more accurately capture the inherent connections and patterns of the medical information of the injured and sick, thereby generating more scientific and personalized treatment plans, significantly improving the quality and efficiency of medical services.

[0118] 103. Based on the precision medical information model, the quantum enhanced learning algorithm is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick through the efficient parallel processing capabilities of quantum computing, and the quantum random forest method is used for efficient classification and regression analysis to generate personalized treatment plan recommendations;

[0119] In this step, the quantum enhanced learning algorithm combines the efficient parallel processing capabilities of quantum computing with the advantages of traditional machine learning. By utilizing the superposition and entanglement characteristics of quantum states, it can process a large amount of data in a short time and significantly improve computing efficiency.

[0120] Quantum computing is a computing model based on the principles of quantum mechanics. It has efficient parallel processing capabilities and can process multiple states at the same time, greatly speeding up the solution of complex problems.

[0121] Efficient parallel processing capabilities enable the algorithm to process large amounts of data in a short period of time, which is particularly important for medical scenarios that require rapid response and can provide personalized treatment recommendations to the injured and sick in a timely manner.

[0122] The personalized treatment plan selection process refers to choosing the most suitable treatment plan based on the specific situation of the injured or sick person. This process requires comprehensive consideration of factors such as the injured or sick person's condition, historical record, individual differences, etc. to ensure that the treatment plan is both effective and safe.

[0123] The quantum random forest method is a classification and regression analysis technology based on quantum computing. It can quickly evaluate the effects and risks of different treatment options by constructing multiple decision trees and executing them in parallel on quantum computers.

[0124] Classification and regression analysis refers to the classification of different treatment options (such as effective / ineffective) and prediction of effects (such as cure rate). This method can help doctors better understand the potential outcomes of each option and make more informed decisions.

[0125] Personalized treatment plan recommendations are the optimal treatment paths generated based on the specific circumstances of the injured and patients and the medical information model. These recommendations are designed to maximize treatment effectiveness while minimizing side effects and risks, ensuring that the treatment plans are both scientific and meet individual needs.

[0126] In the embodiments of the present application, it is assumed that an efficient personalized treatment plan needs to be quickly generated in a resource-limited environment; first, based on the precision medical information model, a quantum-enhanced learning algorithm is used to explore the optimal treatment path; second, the selection of treatment plans is accelerated and optimized through the efficient parallel processing capabilities of quantum computing; third, the quantum random forest method is used to classify and regress the effects and risks of different treatment plans; finally, a personalized treatment plan recommendation is generated to ensure that the treatment plan is both scientific and meets individual needs.

[0127] Optionally, the method in step 103 is based on the precision medical information model, uses a quantum enhanced learning algorithm, and uses the efficient parallel processing capability of quantum computing to accelerate and optimize the personalized treatment plan selection process for the injured and sick, and uses a quantum random forest method to perform efficient classification and regression analysis to generate personalized treatment plan recommendations, including: based on the precision medical information model, deeply analyze the condition and historical treatment data of the injured and sick, extract key medical indicators and treatment response characteristics, and generate a detailed medical feature vector; based on the detailed medical feature vector, use a quantum enhanced learning algorithm to construct a treatment plan search space, and use the efficient parallel processing capability of quantum computing to quickly explore the optimal treatment path and generate a preliminary treatment plan set; based on the preliminary treatment plan set, use a quantum random forest method to classify and regress the effects of different treatment plans, evaluate potential effects and risks, and generate a treatment plan evaluation report; based on the treatment plan evaluation report, comprehensively consider the specific conditions and personalized preferences of the injured and sick, and generate personalized treatment plan recommendations.

[0128] Among them, based on the detailed medical feature vector, a quantum enhanced learning algorithm is used to construct a treatment plan search space, and the optimal treatment path is quickly explored through the efficient parallel processing capability of quantum computing to generate a preliminary treatment plan set, including: based on the detailed medical feature vector, a quantitative analysis of historical treatment responses is performed to determine the key influencing factors of the treatment effect and generate a treatment-related feature matrix; based on the treatment-related feature matrix, a quantum enhanced learning algorithm is used to initialize the treatment plan search space, define potential treatment options and parameter ranges, and consider the possibility of combining multiple treatment plans through the principle of quantum state superposition to generate an initial treatment plan candidate pool; based on the initial treatment plan candidate pool, multiple rounds of iterative optimization processes are performed using the efficient parallel processing capability of quantum computing to perform step-by-step screening and generate an optimized treatment plan list; based on the optimized treatment plan list, specific implementation steps and expected effects are further refined to ensure the feasibility and effectiveness of the plan and generate a preliminary treatment plan set.

[0129] In this step, detailed medical feature vectors are key medical indicators and treatment response characteristics extracted through in-depth analysis of the patient's condition and historical treatment data. These feature vectors capture the patient's specific health status and treatment response, providing a basis for subsequent analysis.

[0130] The treatment plan search space refers to the set of all possible treatment options and their parameter ranges. By initializing the space, the possibility of combining multiple treatment plans is considered to ensure that the optimal treatment path is found.

[0131] The preliminary treatment plan set is a set of candidate treatment plans generated by exploring the treatment plan search space. These plans preliminarily evaluate the effects and risks of different treatment options based on the specific conditions of the injured or sick.

[0132] The treatment plan evaluation report is the result of classification and regression analysis of the effects of different treatment plans. This report evaluates the potential effects and risks of each plan in detail to provide decision-making support for doctors.

[0133] The treatment-related characteristic matrix refers to a matrix generated after quantitative analysis of historical treatment responses. It is used to determine the key influencing factors of treatment effectiveness. The matrix captures the response patterns of the injured and patients to different treatment plans and provides a basis for optimizing treatment plans.

[0134] The initial treatment candidate pool is a set of candidate options generated by considering the possibility of combining multiple treatment options through the principle of quantum state superposition. These options cover a variety of potential treatment paths and provide a basis for further screening.

[0135] The multi-round iterative optimization process refers to the process of gradually screening and optimizing treatment plans through the efficient parallel processing capabilities of quantum computing. Each round of iteration is improved based on the results of the previous round, and finally an optimized list of treatment plans is generated.

[0136] In the embodiment of the present application, firstly, based on the precision medical information model, the patient's condition and historical treatment data are deeply analyzed, key medical indicators and treatment response characteristics are extracted, and a treatment plan search space is constructed by using a quantum enhanced learning algorithm. Through the efficient parallel processing capability of quantum computing, the optimal treatment path is quickly explored to generate a preliminary treatment plan set; secondly, based on the preliminary treatment plan set, the quantum random forest method is used to classify and regress the effects of different treatment plans, evaluate the potential effects and risks, and comprehensively consider the specific conditions and personalized preferences of the patient to generate personalized treatment plan recommendations; thirdly, based on the detailed medical feature vector, the historical treatment response is quantitatively analyzed to determine the key influencing factors of the treatment effect, and the quantum enhanced learning algorithm is used to initialize the treatment plan search space, define potential treatment options and parameter ranges, and consider the possibility of combining multiple treatment plans through the principle of quantum state superposition to generate an initial treatment plan candidate pool; finally, based on the initial treatment plan candidate pool, the efficient parallel processing capability of quantum computing is used to perform multiple rounds of iterative optimization processes, perform step-by-step screening, further refine the specific implementation steps and expected effects, ensure the feasibility and effectiveness of the plan, and generate a preliminary treatment plan set.

[0137] Suppose that in a medical assistance project in a remote area, it is necessary to use limited medical resources to quickly generate efficient personalized treatment plans; first, based on the precision medical information model, deeply analyze the condition and historical treatment data of the injured and sick, extract key medical indicators and treatment response characteristics, use quantum enhanced learning algorithms to construct a treatment plan search space, and use the efficient parallel processing capabilities of quantum computing to quickly explore the optimal treatment path and generate a preliminary treatment plan set; secondly, based on the preliminary treatment plan set, use the quantum random forest method to classify and regress the effects of different treatment plans, evaluate potential effects and risks, and comprehensively consider the specific conditions and personalized preferences of the injured and sick to generate personalized treatment plans. Then, based on the detailed medical feature vector, the historical treatment responses are quantitatively analyzed to determine the key influencing factors of the treatment effect. The quantum enhanced learning algorithm is used to initialize the treatment plan search space, define potential treatment options and parameter ranges, and consider the possibility of combining multiple treatment plans through the principle of quantum state superposition to generate an initial treatment plan candidate pool. Finally, based on the initial treatment plan candidate pool, the efficient parallel processing capabilities of quantum computing are used to perform multiple rounds of iterative optimization processes, conduct step-by-step screening, further refine the specific implementation steps and expected results, ensure the feasibility and effectiveness of the plan, generate a preliminary treatment plan set, and support doctors in making more accurate diagnoses and personalized treatment plans.

[0138] This application takes into account that existing technologies are often limited to traditional machine learning models, which make it difficult to fully capture and process high-dimensional and complex medical data. Therefore, this embodiment introduces a quantum-enhanced learning algorithm to overcome the limitations of traditional methods. By using the principle of quantum state superposition to initialize the treatment plan search space, the possibility of combining multiple treatment plans can be considered at the same time, thereby more comprehensively exploring potential effective treatment paths. In addition, this method can efficiently process large amounts of complex data, ensuring that the generated treatment plan is not only based on the specific condition and historical treatment data of the injured and sick, but also comprehensively considers personal preferences and environmental factors, providing more personalized and scientific treatment recommendations, and providing strong technical support for the realization of precision medicine.

[0139] Optionally, based on the treatment-related feature matrix, a quantum enhanced learning algorithm is used to initialize the treatment plan search space, define potential treatment options and parameter ranges, consider the possibility of combining multiple treatment plans through the principle of quantum state superposition, and generate an initial treatment plan candidate pool, including:

[0140] Based on the treatment-related feature matrix, Z-score standardization preprocessing is performed;

[0141] By using the mutual information-based weight distribution method to assign importance weights, the key features that significantly affect the selection of treatment options are highlighted, and feature selection technology is used to remove irrelevant features to generate quantum state representations:

[0142] The quantum state representation is calculated using the following formula:

[0143]

[0144] Where S is the quantum state representation of the treatment plan search space; α i is the amplitude coefficient of the ith potential treatment option, used to represent its weight in the superposition state; |ψ i > is the quantum state vector of the i-th potential treatment option; i is the index of the potential treatment option, from 1 to N; N is the total number of potential treatment options; η is the influence of the learnable parameter controlling the auxiliary feature; K is the auxiliary feature conversion matrix that maps the auxiliary feature to the quantum state space; ω j is the weight coefficient of each auxiliary feature; H j is a set of auxiliary feature vectors, including the personal preferences and environmental factors of the injured and sick; j is the index of the auxiliary feature vector, from 1 to M; M is the number of auxiliary feature vectors; ζ is a learnable parameter that controls the influence of the Gaussian kernel function; x k is the kth medical feature vector; μ k is the mean of the kth medical feature; σ k is the standard deviation of the kth medical feature; k is the index of the medical feature vector, from 1 to L; L is the number of medical feature vectors;

[0145] Based on the quantum state representation, the importance of different treatment options is adjusted using adaptive weight coefficients, the complex nonlinear relationship in medical data is captured by introducing a Gaussian kernel function, and the expression ability of the model is enhanced by combining advanced feature vectors to generate an enhanced treatment plan vector;

[0146] The enhanced treatment plan vector is calculated using the following formula:

[0147]

[0148] Where Q is the enhanced treatment plan vector; β i The adaptive weight coefficient for each treatment option is used to adjust the degree of influence of different treatment options; γ, δ are learnable parameters to control the influence of additional factors on the treatment plan; F is the feature matrix conversion function that maps the treatment-related feature matrix to the quantum state space; X is the treatment-related feature matrix containing the specific condition and historical treatment data of the injured and sick; G is the parameter range conversion function that defines the parameter range of the treatment option; P is the parameter range matrix that defines the specific parameter restrictions of each treatment option; ⊙ represents the element-by-element multiplication operation; σ is the activation function ReLU or tanh, which is used to introduce nonlinearity; θ k is the weight coefficient for each high-level feature; A k is the high-level feature conversion matrix that maps high-level features to quantum state space; Z is a set of high-level feature vectors including complex medical indicator combinations and multimodal data fusion results; ρ is another activation function, which is LeakyReLU or other nonlinear transformations; λ is a learnable parameter that controls the influence of high-order features; W m Y is the high-order feature conversion matrix that maps high-order features to the quantum state space; m is a set of high-order feature vectors, including complex interactive features and features extracted by deep networks; m is the index of the high-order feature vector, from 1 to D; D is the number of high-order feature vectors; k is the index of the high-order feature vector, from 1 to L; L is the number of high-order feature vectors; i is the index of the potential treatment option, from 1 to N; N is the total number of potential treatment options;

[0149] Based on the enhanced treatment plan vector, a multi-dimensional scoring standard is preset for comprehensive scoring, a threshold is set for screening to construct a preliminary candidate pool, and cross-validation technology is applied to evaluate the stability and generalization ability of each treatment plan to generate an initial treatment plan candidate pool.

[0150] This method aims to use quantum-enhanced learning algorithms to perform multi-layer information transmission on the medical information of the wounded and sick, consider the possibility of combining multiple treatment plans through the principle of quantum state superposition, initialize the treatment plan search space, define potential treatment options and parameter ranges, and generate an initial treatment plan candidate pool. This method can not only process high-dimensional data, but also enhance the focus on key medical information, ensure that the model accurately reflects the internal connections and patterns of the medical information of the wounded and sick, and provide a scientific basis for personalized treatment.

[0151] In the quantum state representation, potential treatment options are superposition terms This part aims to initialize the treatment plan search space through the principle of quantum state superposition, considering the possibility of multiple treatment plan combinations. The amplitude coefficient of each potential treatment option is used to represent its weight in the superposition state, ensuring that the model can fully explore different treatment paths and increase the probability of finding the optimal solution; auxiliary feature conversion item This part introduces auxiliary features (such as personal preferences and environmental factors of the injured and sick) and maps them into the quantum state space. By controlling the influence of auxiliary features, the model can better integrate non-medical factors and provide more personalized and comprehensive treatment suggestions; Gaussian kernel function term This part uses the Gaussian kernel function to capture the complex nonlinear relationships in medical data. By applying the Gaussian kernel function to the standardized medical features, the model can enhance the expression ability while maintaining the data distribution characteristics, thereby more accurately reflecting the specific conditions of the injured and sick.

[0152] Among them, S,α i ,|ψ i > Obtained by pre-setting potential treatment options; η, K, ω j ,H j Determined by weight allocation method based on mutual information; ζ,x k ,μ k ,σ k Preprocessing calculation by Z-score standardization;

[0153] In the enhanced treatment plan vector, the adaptive weight adjustment term This part dynamically adjusts the importance of different treatment options by introducing adaptive weight coefficients to ensure that the model can respond flexibly according to the specific condition. The influence of additional factors (such as parameter range and advanced features) is also regulated by learnable parameters, which enhances the flexibility and adaptability of the model. This part combines advanced feature vectors (such as complex combinations of medical indicators and multimodal data fusion results) and introduces nonlinearity through activation functions to improve the model's ability to understand and express complex medical data, which helps capture deep medical information patterns and provide support for generating precise treatment plans; High-order feature mapping items This part maps high-order features (such as complex interaction features and features extracted by deep networks) to quantum state space and further processes them through activation functions. The introduction of high-order features enables the model to capture more subtle data structures, thereby improving the scientificity and accuracy of treatment plans.

[0154] Among them, Q,β i ,S,γ,F,X,δ,G,P are calculated through quantum state representation; σ,θ k ,A k ,Z is determined by advanced feature selection techniques; λ,W m ,Y m Optimization through the training process;

[0155] Assume that the total number of potential treatment options N = 4; the amplitude coefficient α of each potential treatment option i =[0.1, 0.3, 0.2, 0.4]; quantum state vector of potential treatment options |ψ i >=[[0.4,0.6],[0.7,0.8],[0.5,0.9],[0.6,0.7]]; Learnable parameter controls the influence of auxiliary features η=0.15; Auxiliary feature conversion matrix K=[[0.3,0.4],[0.5,0.6]]; Auxiliary feature vector set H j =[[0.4,0.5],[0.6,0.7]] (including the patient's personal preferences and environmental factors); weight coefficient of each auxiliary feature ω j = [0.7, 0.3]; the influence of the Gaussian kernel function can be controlled by the learnable parameter ζ = 0.25; the medical feature vector X k =[0.6,0.8,0.7]; mean μ of medical characteristics k =[0.5,0.7,0.8]; standard deviation of medical characteristics σ k =[0.1,0.2,0.3];

[0156]

[0157] Assuming the adaptive weight coefficient β i =[0.5,0.4,0.6,0.3]; the learnable parameter controls the impact of additional factors on the treatment plan γ = 0.2; the feature matrix conversion function F = [[0.3,0.4], [0.5,0.6], [0.7,0.8], [0.9,0.1]]; the parameter range conversion function G = [[0.2,0.3], [0.4,0.5], [0.6,0.7], [0.8,0.9]]; the parameter range matrix P = [[0.2,0.3], [0.4,0.5], [0.6,0.7], [0.8,0.9]]; the weight coefficient of the advanced feature θ k=[0.6,0.4,0.5]; High-level feature transformation matrix A k =[[0.4,0.5],[0.6,0.7],[0.8,0.9]]; high-level feature vector set Z = [[0.5,0.6],[0.7,0.8],[0.9,0.1]]; learnable parameter controls the influence of high-order features λ = 0.35; high-order feature conversion matrix W m =[[0.3,0.4],[0.5,0.6],[0.7,0.8]]; high-order eigenvector set Y m =[[0.5,0.6],[0.7,0.8],[0.9,0.1]];

[0158]

[0159] Assuming the threshold is 0.6, since the calculated enhanced treatment plan vector Q = [0.65, 0.78, 0.82, 0.71], all elements are greater than the set threshold, these results exceeding the threshold not only verify the effectiveness of the quantum enhanced learning algorithm, but also ensure the accuracy and reliability of subsequent diagnosis and personalized treatment recommendations, which shows that the model successfully enhances the information characteristics of certain treatment plans. Through the above steps, the system can more comprehensively explore potential effective treatment paths, while efficiently processing high-dimensional and complex medical data, comprehensively considering the specific conditions of the injured and sick, historical treatment data, personal preferences and environmental factors, thereby generating a more scientific and personalized treatment plan candidate pool.

[0160] 104. Based on the personalized treatment plan recommendation, develop data backup and recovery functions to ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendation to a third-party system.

[0161] Data backup and recovery function refers to the mechanism that ensures that data can maintain integrity and availability under any circumstances to prevent data loss or damage. It includes the ability to back up data regularly and restore data quickly when necessary.

[0162] Data integrity refers to the accuracy and consistency of data, ensuring that the data has not been tampered with or lost, which is crucial to ensuring the authenticity and reliability of medical information.

[0163] Data availability means that data can be accessed and used at any time. Even in the event of system failure or other abnormal circumstances, data accessibility should be ensured so as not to affect the continuity of medical services.

[0164] The integrated health record management system is a platform for centralized management and storage of personal health information. It not only supports medical staff to conveniently access and update the medical records of the injured and sick, but also provides powerful data analysis and report generation functions.

[0165] In the embodiment of this application, it is assumed that it is necessary to ensure the security and convenient access of data in a long-term medical environment; first, based on the personalized treatment plan recommendation, the personalized treatment plan recommendation is sent to the third-party system. Secondly, develop an automated data backup and recovery function to regularly save data copies; further, protect the security of backup data through encryption technology and security protocols; thirdly, build an integrated health record management system to support medical staff and the injured and sick to easily access and update personal health records; finally, ensure the high availability and reliability of the system to provide continuous support for medical services.

[0166] Optionally, the step 104 develops a data backup and recovery function based on the personalized treatment plan recommendation, ensures data integrity and availability, and generates an integrated health record management system, including: based on the personalized treatment plan recommendation, the medical information and treatment plans of the injured and sick are structured to ensure the orderliness and accessibility of the information, and a standardized medical information library is generated; based on the standardized medical information library, an efficient data backup mechanism is developed, incremental backup operations are automatically performed regularly, encryption technology is used to protect the security of backup data, and a secure backup data set is generated; based on the secure backup data set, a fast data recovery process is established to support rapid recovery to the most recent state when the system fails, and a data recovery strategy is generated; based on the data recovery strategy, data management and user authority control function modules are integrated to generate an integrated health record management system.

[0167] In this step, the standardized medical information database refers to the data set generated after the medical information and treatment plans of the injured and sick are structured and organized.

[0168] An efficient data backup mechanism refers to the regular and automatic execution of incremental backup operations to protect important medical data from loss or damage.

[0169] A secure backup data set refers to a collection of backup data that has been encrypted and stored in a secure environment.

[0170] The rapid data recovery process refers to a mechanism that is established to quickly restore the system to its most recent state when a system fails, supports efficient recovery operations, and reduces service interruption time caused by data loss.

[0171] A data recovery strategy is a detailed recovery plan developed for different types of data loss or system failures, including the selection of backup data, recovery steps, and methods for verifying recovery results.

[0172] The integrated health record management system is a platform for centralized management and storage of personal health information, providing convenient and fast access, update and management functions, including data management and user authority control modules to ensure high availability and security of the system.

[0173] The user authority control function module refers to the system component used to manage the permissions of different users to access and operate health records, ensuring that only authorized personnel can view or modify sensitive data and ensure information security.

[0174] In the embodiments of the present application, firstly, based on the personalized treatment plan recommendations, the medical information and treatment plans of the injured and sick are structured and organized to ensure the orderliness and accessibility of the information, and to generate a standardized medical information database; secondly, based on the standardized medical information database, an efficient data backup mechanism is developed, incremental backup operations are automatically performed regularly, encryption technology is used to protect the security of the backup data, and a secure backup data set is generated; thirdly, based on the secure backup data set, a fast data recovery process is established to support rapid recovery to the most recent state in the event of a system failure, and a data recovery strategy is generated; finally, based on the data recovery strategy, the data management and user authority control functional modules are integrated to generate an integrated health record management system.

[0175] Suppose that efficient information management of the wounded and sick is required in a transnational network of medical institutions. First, medical information and personalized treatment plans of the wounded and sick from branches around the world are collected and organized in a structured manner to ensure that all information is clear and easily accessible, and a standardized medical information database is generated. Secondly, based on this standardized database, an efficient automatic backup mechanism is developed to automatically perform incremental backups every morning, and use advanced encryption standards to protect backup data to ensure its security and privacy. Thirdly, in order to cope with possible system failures, a detailed rapid data recovery process is established to ensure that the most recent state can be quickly restored in the event of a failure, minimizing service interruptions. Finally, combined with data recovery strategies, data management and user authority control functional modules are integrated to ensure that only authorized personnel can access sensitive information, generating an integrated health record management system to support transnational team collaboration and data sharing, and improve the overall level of medical services.

[0176] In summary, steps 101 to 104 cover the entire process from medical data preprocessing to generating an initial treatment plan candidate pool, aiming to provide an efficient and personalized treatment plan recommendation system to meet the needs of precision medicine. By introducing quantum-enhanced learning algorithms and the principle of quantum state superposition, these steps ensure that the model can fully explore potential effective treatment paths and comprehensively consider the specific conditions, historical treatment data, personal preferences, and environmental factors of the injured and sick, thereby providing strong technical support for achieving more scientific and personalized medical services.

[0177] Figure 2 A schematic diagram of the structure of an electronic management system for the information of the injured and sick is provided for the embodiment of the present application, such as Figure 2 As shown, the device comprises:

[0178] The receiving module 21 is used to receive basic information and detailed medical records of the injured and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set;

[0179] The processing module 22 is used to perform multi-level information analysis and entity association processing based on the comprehensive medical data set using a hypergraph neural network algorithm, model the high-order interaction between the medical information of the injured and the sick through a hyperedge structure, and dynamically adjust the importance weight of each information node using an adaptive attention mechanism technology to generate a precise medical information model;

[0180] The analysis module 23 is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick based on the precision medical information model, using the quantum enhanced learning algorithm and the efficient parallel processing capability of quantum computing, and adopt the quantum random forest method to perform efficient classification and regression analysis to generate personalized treatment plan recommendations;

[0181] The generation module 24 is used to develop data backup and recovery functions based on the personalized treatment plan recommendation, ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendation to a third-party system.

[0182] Figure 2 The electronic management system for the information of the injured and sick can be implemented Figure 1 The implementation principle and technical effect of the electronic management method for the patient information described in the embodiment are not described in detail. The specific way in which each module and unit performs operations in the electronic management system for the patient information in the above embodiment has been described in detail in the embodiment of the method, and will not be described in detail here.

[0183] In one possible design, Figure 2 An electronic management system for patient information in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0184] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0185] The processing component 32 is used to: receive basic information and detailed medical records of the wounded and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set; based on the comprehensive medical data set, use the hypergraph neural network algorithm to perform multi-level information analysis and entity association processing, model the high-order interaction between the medical information of the wounded and sick through the hyperedge structure, and use the adaptive attention mechanism technology to dynamically adjust the importance weight of each information node to generate a precise medical information model; based on the precise medical information model, use the quantum enhanced learning algorithm, through the efficient parallel processing capability of quantum computing, accelerate and optimize the selection process of personalized treatment plans for the wounded and sick, use the quantum random forest method to perform efficient classification and regression analysis, and generate personalized treatment plan recommendations; based on the personalized treatment plan recommendations, develop data backup and recovery functions to ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendations to a third-party system.

[0186] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0187] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0188] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0189] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0190] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0191] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0192] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 A method for electronic management of information of the sick and injured in the embodiment shown.

[0193] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0194] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0195] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0196] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for electronic management of information of the wounded and sick, characterized in that: include: Receive basic information and detailed medical records of the injured and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set; Based on the comprehensive medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and a hyperedge structure is used to model the high-order interaction between the medical information of the injured and the sick. The adaptive attention mechanism technology is used to dynamically adjust the importance weight of each information node to generate a precise medical information model. Based on the precision medical information model, the quantum enhanced learning algorithm is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick through the efficient parallel processing capabilities of quantum computing, and the quantum random forest method is used for efficient classification and regression analysis to generate personalized treatment plan recommendations; Based on the personalized treatment plan recommendation, develop data backup and recovery functions to ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendation to a third-party system.

2. The method according to claim 1, characterized in that Based on the comprehensive medical data set, the hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, the high-order interaction between the medical information of the injured and the sick is modeled through the hyperedge structure, and the importance weight of each information node is dynamically adjusted using the adaptive attention mechanism technology to generate a precise medical information model, including: Based on the comprehensive medical data set, data cleaning and standardization are performed to remove irrelevant information and noise, ensure that the data has a high degree of consistency, and generate a clean medical data set; Based on the clean medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and a preliminary medical information model is generated by modeling high-order interactions between medical information of the injured and sick through a hyperedge structure; Based on the preliminary medical information model, an adaptive attention mechanism technology is used to evaluate the importance, dynamically adjust the importance weight of each information node, enhance the attention to key medical information, and generate an optimized medical information model; Based on the optimized medical information model, the model parameters are further adjusted to ensure that the model accurately captures the internal connections and patterns of the medical information of the injured and sick, and generates a precise medical information model.

3. The method according to claim 2, characterized in that Based on the clean medical data set, a hypergraph neural network algorithm is used to perform multi-level information analysis and entity association processing, and a high-order interaction between the medical information of the injured and the sick is modeled through a hyperedge structure to generate a preliminary medical information model, including: Based on the clean medical data set, preprocess the text and structured data, extract key features, construct initial representations of nodes and edges, and generate node feature vectors; Based on the node feature vectors, a hypergraph neural network algorithm is used to perform multi-layer information transmission on the medical information of the injured and sick, and the complex dependency relationship between different medical entities is captured through the hyperedge structure to generate an intermediate layer feature representation; Based on the intermediate layer feature representation, a hypergraph convolution operation is used to further strengthen the interaction between nodes, context information is introduced to enhance the model's understanding ability, and an enhanced feature representation is generated; Based on the enhanced feature representation, all levels of information are aggregated to form a complete medical information map of the injured and sick, and generate a preliminary medical information model.

4. The method according to claim 2, characterized in that Based on the preliminary medical information model, the adaptive attention mechanism technology is used to evaluate the importance, dynamically adjust the importance weight of each information node, enhance the attention to key medical information, and generate an optimized medical information model, which specifically includes: Based on the preliminary medical information model, feature extraction is performed on each entity and its relationship in the medical information of the injured and sick to generate an entity feature vector; Based on the entity feature vector, the importance of each entity is evaluated using the adaptive attention mechanism technology, a relative attention score is obtained, and an attention weight vector is generated; Based on the attention weight vector, dynamically adjust the importance weight of each information node, weaken the influence of irrelevant information, and generate a weighted medical information model; Based on the weighted medical information model, the model parameters are further corrected to ensure accurate reflection of the key features and internal connections of the medical information of the injured and sick, thereby generating an optimized medical information model.

5. The method according to claim 1, characterized in that: Based on the precision medical information model, the quantum enhanced learning algorithm is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick through the efficient parallel processing capabilities of quantum computing, and the quantum random forest method is used for efficient classification and regression analysis to generate personalized treatment plan recommendations, including: Based on the precision medical information model, deeply analyze the patient's condition and historical treatment data, extract key medical indicators and treatment response characteristics, and generate detailed medical feature vectors; Based on the detailed medical feature vector, a quantum enhanced learning algorithm is used to construct a treatment plan search space, and the optimal treatment path is quickly explored through the efficient parallel processing capability of quantum computing to generate a preliminary treatment plan set; Based on the preliminary treatment plan set, the quantum random forest method is used to classify and regress the effects of different treatment plans, evaluate potential effects and risks, and generate a treatment plan evaluation report; Based on the treatment plan evaluation report, a personalized treatment plan recommendation is generated by comprehensively considering the specific conditions and personalized preferences of the injured and sick.

6. The method according to claim 5, characterized in that Based on the detailed medical feature vector, the quantum enhanced learning algorithm is used to construct a treatment plan search space, and the efficient parallel processing capability of quantum computing is used to quickly explore the optimal treatment path and generate a preliminary treatment plan set, including: Based on the detailed medical feature vector, quantitatively analyze the historical treatment response, determine the key influencing factors of the treatment effect, and generate a treatment-related feature matrix; Based on the treatment-related feature matrix, a quantum enhanced learning algorithm is used to initialize the treatment plan search space, define potential treatment options and parameter ranges, consider the possibility of combining multiple treatment plans through the principle of quantum state superposition, and generate an initial treatment plan candidate pool; Based on the initial treatment plan candidate pool, using the efficient parallel processing capability of quantum computing, multiple rounds of iterative optimization processes are performed to perform step-by-step screening and generate an optimized treatment plan list; Based on the optimized treatment plan list, the specific implementation steps and expected results are further refined to ensure the feasibility and effectiveness of the plan, and generate a preliminary treatment plan set.

7. The method according to claim 1, characterized in that Based on the personalized treatment plan recommendation, the data backup and recovery function is developed to ensure data integrity and availability, and an integrated health record management system is generated, including: Based on the personalized treatment plan recommendations, the medical information and treatment plans of the injured and sick are structured to ensure the orderliness and accessibility of the information and generate a standardized medical information database; Based on the standardized medical information database, develop an efficient data backup mechanism, automatically perform incremental backup operations regularly, use encryption technology to protect the security of backup data, and generate a secure backup data set; Based on the secure backup data set, a fast data recovery process is established to support rapid recovery to the most recent state in case of system failure and generate a data recovery strategy; Based on the data recovery strategy, the data management and user authority control functional modules are integrated to generate an integrated health record management system.

8. An electronic management system for the information of the wounded and sick, characterized in that: include: The receiving module is used to receive basic information and detailed medical records of the injured and sick from different sources, ensure the comprehensiveness and accuracy of the information, and generate a comprehensive medical data set; A processing module, which is used to perform multi-level information analysis and entity association processing based on the comprehensive medical data set using a hypergraph neural network algorithm, model the high-order interaction between the medical information of the injured and the sick through a hyperedge structure, and dynamically adjust the importance weight of each information node using an adaptive attention mechanism technology to generate a precise medical information model; An analysis module is used to accelerate and optimize the selection process of personalized treatment plans for the injured and sick based on the precision medical information model, using a quantum enhanced learning algorithm and the efficient parallel processing capabilities of quantum computing, and to use a quantum random forest method for efficient classification and regression analysis to generate personalized treatment plan recommendations; A generation module is used to develop data backup and recovery functions based on the personalized treatment plan recommendation, ensure data integrity and availability, generate an integrated health record management system, and send the personalized treatment plan recommendation to a third-party system.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an electronic management method for the information of the injured and the sick as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, an electronic management method for the information of the injured and sick as described in any one of claims 1 to 7 is implemented.

Images