Chronic patient symptom management method and system based on symptom network analysis

Abstract

The invention provides a chronic patient symptom management method and system based on symptom network analysis, and relates to the technical field of data processing. The method comprises the following steps: collecting a chronic disease patient symptom evaluation data set; a principal component analysis method and a maximum variance orthogonal rotation method are combined, and a symptom group in the symptom evaluation data set of the chronic disease patient is extracted; constructing a symptom network by taking the symptoms as nodes and the relationship between the symptoms as edges; core symptoms and bridge symptoms are identified by using synchronous network analysis; constructing a dynamic symptom network based on a cross-lag network model; the influence factors are introduced into the dynamic symptom network, subgroup analysis is carried out on the dynamic symptom network, and core influence factors are identified; constructing a chronic disease development trend prediction model based on LSTM; collecting clinical symptoms and clinical indexes of the chronic disease patient; extracting core influence factors; and predicting and outputting the development result of the chronic disease patient through the LSTM-based chronic disease development trend prediction model.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and system for symptom management of chronic patients based on symptom network analysis. Background Art

[0002] Symptom network analysis is a technique that studies how different symptoms are interrelated and identifies their relative importance and impact through graph theory. Symptom management refers to the effective monitoring, evaluation, and intervention of patients' symptoms to relieve pain and improve the quality of life of patients. The method for symptom management of chronic patients based on symptom network analysis is a method that helps doctors and patients identify core symptoms and potential influencing factors by constructing and analyzing the network between symptoms, so as to more effectively manage and treat the symptoms of chronic disease patients.

[0003] Managing the symptoms of chronic disease patients can well help doctors more accurately understand the symptom development trend of chronic diseases, improve the treatment effect of chronic diseases and the quality of life of patients, thereby effectively identifying potential risk factors and providing a basis for formulating more scientific intervention measures for patients.

[0004] Traditional methods for symptom management of chronic patients often ignore the complex connections between symptoms and individual differences among different patients, and cannot comprehensively identify the core influencing factors and potential risk factors between symptoms. In addition, traditional methods have weak predictive ability for chronic disease symptoms and often have difficulty formulating personalized treatment plans for each patient's specific situation, thus affecting the treatment effect and the quality of life of patients. Summary of the Invention

[0005] In order to solve the technical problems that traditional methods for symptom management of chronic patients often ignore the complex connections between symptoms and individual differences among different patients, cannot comprehensively identify the core influencing factors and potential risk factors between symptoms, and at the same time traditional methods have weak predictive ability for chronic disease symptoms and often have difficulty formulating personalized treatment plans for each patient's specific situation, thus affecting the treatment effect and the quality of life of patients, the present invention provides a method and system for symptom management of chronic patients based on symptom network analysis.

[0006] The technical solutions provided by the embodiments of the present invention are as follows:

[0007] First aspect:

[0008] A method for symptom management of chronic patients based on symptom network analysis provided by an embodiment of the present invention includes:

[0009] S1: Collect symptom data of chronic disease patients to obtain a symptom evaluation data set of chronic disease patients;

[0010] S2: Combine the principal component analysis method and the varimax orthogonal rotation method to extract symptom clusters from the symptom assessment dataset of chronic disease patients. The symptom clusters include multiple symptoms;

[0011] S3: Construct a symptom network with each symptom as a node and the relationships between symptoms as edges;

[0012] S4: Use the synchronous network analysis algorithm to identify the core symptoms and bridge symptoms in the symptom network;

[0013] S5: Based on the relationships between the core symptoms and bridge symptoms, construct a dynamic symptom network based on the cross-lagged network model;

[0014] S6: Introduce influencing factors into the dynamic symptom network, and based on the relationships between the influencing factors and symptoms, conduct subgroup analysis on the dynamic symptom network to identify the core influencing factors that lead to differences in the symptom network structure;

[0015] S7: According to the core influencing factors, construct a prediction model for the development trend of chronic diseases based on LSTM;

[0016] S8: Collect the real-time clinical symptom data and real-time clinical index data of chronic disease patients;

[0017] S9: Extract the real-time core influencing factors from the real-time clinical symptom data and real-time clinical index data;

[0018] S10: Input the real-time core influencing factors into the prediction model for the development trend of chronic diseases, and predict and output the development results of the chronic disease symptoms of chronic disease patients.

[0019] Second aspect:

[0020] A chronic patient symptom management system based on symptom network analysis provided by an embodiment of the present invention includes:

[0021] A processor;

[0022] A memory, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor, the chronic patient symptom management method based on symptom network analysis as described in the first aspect is implemented.

[0023] Third aspect:

[0024] A computer-readable storage medium provided by an embodiment of the present invention, on which a computer program is stored. When the program is executed by the processor, the chronic patient symptom management method based on symptom network analysis as described in the first aspect is implemented.

[0025] The beneficial effects brought by the technical solutions provided by the embodiments of the present invention at least include:

[0026] In the embodiments of the present invention, by collecting the symptom data of chronic disease patients, real inputs are provided for subsequent analysis. Symptom groups are extracted based on the principal component analysis method and the varimax orthogonal rotation method. Then, with symptoms as nodes and the relationships between symptoms as edges, a symptom network is constructed. Synchronous network analysis is used to identify the core symptoms and bridge symptoms in the symptom network. Based on the relationships between the core symptoms and bridge symptoms, a dynamic symptom network based on the cross-lagged network model is constructed, enabling symptom management not only to be limited to static analysis but also to capture the evolving trend of symptoms over time. Further, influencing factors are introduced into the dynamic symptom network, and subgroup analysis is performed on the dynamic symptom network to identify the core influencing factors that lead to differences in the symptom network structure. Based on the core influencing factors, a prediction model for the development trend of chronic diseases based on LSTM is constructed to provide prospective guidance for treatment. Finally, by collecting the clinical symptom data and clinical index data of patients and extracting the core influencing factors, the prediction model for the development trend of chronic diseases based on LSTM is used to predict and output the development results of chronic disease symptoms in chronic disease patients, providing a scientific basis for personalized treatment and long-term management. By analyzing the relationships and interactions between the symptoms of chronic patients, the present invention can effectively manage and treat the symptoms of chronic disease patients, help doctors and patients identify the core symptoms and potential influencing factors, improve the treatment effect and quality of life of chronic patients, and provide a personalized symptom management plan for the unique situation of each patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0028] Figure 1 It is a schematic flowchart of a method for managing chronic patient symptoms based on symptom network analysis provided by an embodiment of the present invention;

[0029] Figure 2 It is a schematic structural diagram of a system for managing chronic patient symptoms based on symptom network analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] The technical solutions in the present invention will be described below with reference to the drawings.

[0031] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to give examples, illustrations or explanations. Any embodiment or design described as an "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or designs. Rather, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0032] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meanings they express are the same. "Of", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meanings they express are the same.

[0033] In the embodiments of the present invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0034] To make the technical problems to be solved, technical solutions and advantages of the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0035] Refer to the attached Figure 1 , which shows a schematic flowchart of a method for managing chronic patients' symptoms based on symptom network analysis provided by the embodiments of the present invention.

[0036] The embodiments of the present invention provide a method for managing chronic patients' symptoms based on symptom network analysis. This method can be implemented by a device for managing chronic patients' symptoms based on symptom network analysis. The device for managing chronic patients' symptoms based on symptom network analysis can be a terminal or a server. The processing flow of the method for managing chronic patients' symptoms based on symptom network analysis can include the following steps:

[0037] S1: Collect symptom data of chronic disease patients to obtain a symptom assessment data set of chronic disease patients.

[0038] Among them, chronic disease patients refer to patients suffering from diseases that cannot be completely cured for a long time, such as hypertension, diabetes, chronic respiratory diseases, etc., who need long-term management and treatment. The symptom assessment data set refers to a set of information collected about the patients' symptoms, which is used for subsequent analysis to help evaluate and manage the patients' symptoms.

[0039] It should be noted that by collecting the symptom data, clinical data, and personal information of chronic disease patients, the health status of patients can be comprehensively understood. These data not only provide effective input for subsequent symptom group extraction and network construction, but also can provide a basis for personalized treatment and prediction, thereby improving the accuracy and effectiveness of chronic disease management.

[0040] In a possible implementation, the symptom assessment dataset of chronic disease patients specifically includes: symptom information, clinical data, and patient personal information.

[0041] Among them, the symptom information includes the symptom name, the severity of the symptom, the frequency of symptom occurrence, the duration of the symptom, and the change trend of the symptom. The clinical data includes the types of chronic diseases of the patient, the patient's past medical history, and imaging data. The patient personal information includes age, gender, weight, height, and lifestyle information.

[0042] S2: Combine the principal component analysis method and the varimax orthogonal rotation method to extract the symptom groups in the symptom assessment dataset of chronic disease patients. The symptom groups include multiple symptoms.

[0043] Among them, the principal component analysis method (PCA) is a statistical method used to transform high-dimensional data into low-dimensional data while retaining as much variability in the original data as possible. The varimax orthogonal rotation method is a rotation method aimed at making the interpretation of the principal components clearer. It rotates the eigenvectors so that each factor (principal component) can more strongly reflect certain specific variables, thus facilitating subsequent analysis and understanding. The symptom group refers to a set of interrelated symptoms extracted through principal component analysis.

[0044] It should be noted that through PCA, the most representative factors can be extracted from a large number of symptoms, reducing the dimension and retaining important information, while the Varimax rotation further enhances the interpretability of the factors. This method not only makes the symptom data more concise and understandable, but also can reveal the internal connections between symptoms, helping doctors better identify the symptom groups of patients and laying a solid foundation for subsequent symptom network construction and analysis.

[0045] In a possible implementation, S2 specifically includes:

[0046] S201: Standardize the symptom assessment dataset of chronic disease patients.

[0047] S202: Based on the principal component analysis method, calculate the covariance matrix of the standardized dataset:

[0048]

[0049] Among them, C represents the covariance matrix, Z iDenote the standardized eigenvector of the \(i\)-th symptom, where \(i = 1,\ldots,n\), \(n\) represents the total number of symptoms, and \(T\) represents the transpose.

[0050] S203: Calculate the eigenvalues and eigenvectors of the covariance matrix:

[0051] CV k =\(\lambda\) k V k

[0052] where \(V\) k denotes the \(k\)-th eigenvector, and \(\lambda\) k denotes the \(k\)-th eigenvalue.

[0053] S204: Select the principal components according to the magnitudes of the eigenvalues.

[0054] S205: Calculate the weights of the principal components and determine the eigenvectors to be rotated:

[0055] PC k =\(ZV\) k

[0056] where \(PC\) k denotes the \(k\)-th principal component, and \(Z\) represents the data matrix.

[0057] S206: Rotate the eigenvectors to be rotated by the varimax orthogonal rotation method:

[0058] V rot =\(VQ\)

[0059] where \(V\) rot denotes the rotated eigenvector, \(V\) denotes the eigenvector to be rotated, and \(Q\) represents the rotation matrix.

[0060] S207: Determine the factor loading matrix with the maximum rotation objective:

[0061]

[0062] where \(Varimax\) represents the rotation method, and \(F\) ij denotes the loading of the \(i\)-th variable on the \(j\)-th factor.

[0063] S208: Determine the symptom groups according to the factor loading matrix.

[0064] It should be noted that extracting symptom groups through principal component analysis and varimax orthogonal rotation method can effectively classify and simplify a large amount of symptom data. This step transforms a large amount of symptom data into more manageable symptom groups, facilitating subsequent analysis and management.

[0065] S3: Construct a symptom network with each symptom as a node and the relationships between symptoms as edges.

[0066] Among them, in graph theory, a node is a basic element in a graph, representing an entity. In a symptom network, each symptom serves as a node, representing a specific symptom. In a symptom network, an edge represents the relationship or interaction between different symptoms. A symptom network refers to a network structure that regards the symptoms of chronic disease patients as nodes and constructs through the relationships between symptoms.

[0067] It should be noted that by constructing a symptom network, different symptoms are regarded as nodes, and the mutual relationships between symptoms are regarded as edges, forming a complex network structure. This method can help reveal the internal connections and interactions between symptoms, and identify the dependence relationships and mutual influences between symptoms. Through the symptom network, doctors can clearly see which symptoms are interrelated and which symptoms may be the key points in the development of the disease, providing a powerful tool for subsequent symptom analysis, management, and prediction.

[0068] S4: Use the co-occurrence network analysis algorithm to identify the core symptoms and bridge symptoms in the symptom network.

[0069] Among them, the co-occurrence network analysis algorithm refers to analyzing the symptoms that are correlated with each other in time to find the interaction relationships between different symptoms. Core symptoms refer to the symptoms that occupy an important position in the symptom network and affect other symptoms or the progression of the disease. Bridge symptoms refer to the symptoms that connect different symptom groups, and these symptoms play a key bridging role between different symptoms. Understanding bridge symptoms helps to reveal the important links in the symptom network.

[0070] In a possible implementation manner, S4 specifically includes:

[0071] S401: Through linear projection, convert the node features and edge features into hidden features respectively, and input the hidden features into the GAT model:

[0072]

[0073] Among them, represents the feature vector of node i at the initial time, U 0 represents the initial node weight matrix, α i represents the input feature vector of node i, u 0 represents the initial node bias vector, represents the feature vector of edge (x,y) in the initial layer, V 0 represents the initial edge weight matrix, β ij represents the input feature vector of edge (x,y), v 0 represents the initial edge bias vector.

[0074] S402: Calculate the attention weights of each symptom node through the graph attention mechanism of the GAT model:

[0075]

[0076] Among them, represents the feature representation of edge (x, y) under the k-th head, LeakyReLU represents the non-linear activation function, and a l represents the scalar coefficient of the l-th layer, Concat represents the concatenation operation, and W l represents the weight matrix of the l-th layer, represents the feature vector of node i in the l-th layer, represents the feature vector of node j in the l-th layer.

[0077] It should be noted that the graph attention mechanism (GAT) is a mechanism based on graph neural networks, which can assign different attention weights to each node, so as to preferentially process the most important parts of the network for the model.

[0078] S403: Normalize the attention weights:

[0079]

[0080] Among them, represents the normalized attention weight of edge (x, y) under the k-th head and the l-th layer, exp represents the exponential function, and N i represents the set of neighbor nodes of node i.

[0081] S404: According to the normalized attention weights, introduce asymmetry into the neighborhood aggregation function to update the symptom nodes:

[0082]

[0083] Among them, represents the feature vector of node i in the l + 1-th layer, Concat represents the concatenation operation, ELU represents the activation function, and U k,l represents the weight matrix of the k-th attention head in the l-th layer, k = 1, 2,..., K, and K represents the total number of attention heads.

[0084] S405: According to the updated symptom nodes, calculate the degree centrality and betweenness centrality respectively, and determine the core symptoms and bridge symptoms:

[0085] CoreSymptoms = {i|DegreeCentrality(i) > θ}

[0086] BridgeSymptoms = {i|Betweenness(i) > θ′}

[0087] Among them, CoreSymptoms represents the set of core symptoms, DegreeCentrality(i) represents the degree centrality of node i, θ represents the core symptom threshold, BridgeSymptoms represents the set of bridge symptoms, Betweenness(i) represents the betweenness centrality of node i, and θ′ represents the bridge symptom threshold.

[0088] It should be noted that the introduction of the graph attention mechanism makes the analysis more accurate, can automatically focus on the most important parts of the symptom network, and improve the overall management efficiency and treatment accuracy. This makes the chronic disease symptom management not only more personalized, but also more scientific and efficient.

[0089] S5: Based on the relationship between core symptoms and bridge symptoms, construct a dynamic symptom network based on the cross-lagged network model.

[0090] Among them, the cross-lagged network model is a dynamic model used to study the symptom relationships between different time steps. The dynamic symptom network refers to the mutual connection and evolution of symptoms over time. Based on the cross-lagged network model, it can reflect the changing trend of symptoms over time.

[0091] It should be noted that by constructing a dynamic symptom network based on the cross-lagged network model, the changes of symptoms over time and their interactions can be revealed. Compared with static analysis, this method can capture the evolution process of symptoms and the causal relationships between symptoms, providing a more dynamic and real-time analysis for the symptom management of chronic diseases. By identifying core symptoms and bridge symptoms, doctors can more accurately understand the changes and influence paths between symptoms, and thus take timely intervention measures.

[0092] In a possible implementation, the dynamic symptom network based on the cross-lagged network model is specifically:

[0093]

[0094] Among them, represents the feature vector of node i at layer l - 1, represents the feature vector of node j at layer l - 1, and edgefeatures represents the edge feature.

[0095] It should be noted that the dynamic symptom network helps to develop personalized treatment plans for patients, predict the future development trend of diseases, and improve the treatment effect of chronic diseases and the quality of life of patients.

[0096] S6: Introduce influencing factors into the dynamic symptom network, and based on the relationship between influencing factors and symptoms, conduct subgroup analysis on the dynamic symptom network to identify the core influencing factors that lead to differences in the symptom network structure.

[0097] Among them, influencing factors refer to external or internal factors that may have a significant impact on the symptom network and its structure, such as the patient's lifestyle, eating habits, environmental factors, psychological state, etc. Subgroup analysis refers to dividing patients into different subgroups (such as the severity of the disease, age group, etc.) according to certain characteristics or conditions, and analyzing these subgroups separately to reveal the differences in the symptom network among different groups. The differences in symptom network structure refer to the fact that the relationships and network structures among symptoms may vary in different patient groups, and these differences may be related to the disease process, the individual characteristics of the patient, or external factors.

[0098] It should be noted that by introducing influencing factors and conducting subgroup analysis, the core factors affecting the differences in symptom network structure can be identified. This method can deeply explore the characteristics of different patient groups, reveal the impact of individual differences on the symptom network, and make symptom management more personalized and refined.

[0099] In one possible implementation, S6 specifically includes:

[0100] S601: Introduce influencing factors into the node features of the dynamic symptom network through feature splicing.

[0101] S602: Update the node features.

[0102] S603: Divide the updated node features into multiple subgroups.

[0103] S604: Analyze each subgroup to identify multiple different patient groups:

[0104]

[0105] Among them, Group k represents the patient group of the k-th group, represents the influencing factor of the l-th layer, and G k represents the conditional set of the k-th group.

[0106] S605: Calculate the differences between the symptom network structures of each subgroup to identify the core influencing factors:

[0107]

[0108] Among them, Difference(G k , G l ) represents the difference measure between the symptom network G k of the k-th group and the symptom network G l of the l-th group, and V k represents the number of nodes in the symptom network of the k-th group.

[0109] It should be noted that by considering multi-dimensional factors such as the patient's living habits and psychological state, it is possible to help doctors identify potential risk factors, thereby providing more precise treatment and intervention plans for different patient groups. In addition, subgroup analysis can also discover differences in symptom manifestations and network structures among different patient groups, further optimizing the management of chronic diseases and improving the treatment effect.

[0110] S7: Construct a prediction model for the development trend of chronic diseases based on LSTM according to the core influencing factors.

[0111] Among them, LSTM (Long Short-Term Memory) is a special recurrent neural network (RNN) that can process and predict time series data. The prediction model for the development trend of chronic diseases refers to a model based on the LSTM model that predicts the future symptom change trend of chronic diseases through the learning of historical symptom data.

[0112] S8: Collect real-time clinical symptom data and real-time clinical index data of chronic disease patients.

[0113] Among them, clinical symptom data refers to various clinical symptoms shown by patients during the disease process, such as pain, fatigue, shortness of breath, etc. These symptoms reflect the changes in the patient's health status and are important bases for disease management and prediction. Clinical index data refers to the data obtained through medical examinations and tests, such as blood pressure, blood glucose level, electrocardiogram, etc., which provide specific information about the patient's physiological state and help doctors diagnose and evaluate the severity of the disease.

[0114] It should be noted that by collecting the clinical symptoms and clinical index data of patients, a comprehensive information basis is provided for the symptom management of chronic diseases. It can ensure that the symptom management system has accurate input data, help doctors understand the patient's health status more comprehensively, and by combining the symptom data with the clinical index data, the patient's condition can be evaluated from multiple angles and in all-round ways, so as to make more scientific and personalized treatment decisions.

[0115] S9: Extract real-time core influencing factors from real-time clinical symptom data and real-time clinical index data.

[0116] S10: Input the real-time core influencing factors into the prediction model for the development trend of chronic diseases, and predict and output the development results of chronic disease symptoms of chronic disease patients.

[0117] Among them, the development results of chronic disease symptoms refer to the results of the future symptom changes or disease development of patients output based on the LSTM prediction model, which helps doctors judge the development trend of the disease and provides a basis for adjusting the treatment plan.

[0118] It should be noted that through the training and prediction of the LSTM model, the development results of the chronic disease symptoms of the patient are output, providing a scientific basis for personalized treatment and long-term management.

[0119] In a possible implementation manner, S10 specifically includes:

[0120] S1001: Smooth the core influencing factors to obtain a time parameter vector:

[0121]

[0122] where p Δt-1:t represents the time parameter vector from time step t - 1 to time step t, and Δ t-1:t represents the time interval from time step t - 1 to time step t.

[0123] It should be noted that the time parameter vector refers to the vector used to represent the change information between time steps in time series data. In the LSTM model, the time parameter vector helps the model capture the dynamic characteristics of symptom evolution over time by reflecting the changes between different time points.

[0124] S1002: Input the time parameter vector into the chronic disease development trend prediction model for training until the value of the loss function is less than the preset loss function value:

[0125]

[0126] where f t represents the output of the forget gate at time step t, σ represents the Sigmoid activation function, h t-1 represents the hidden state at time step t - 1, x t represents the input at time step t, P f represents the weight parameter of the forget gate, b f represents the bias term of the forget gate, W f represents the weight matrix of the forget gate, represents the candidate memory unit at time step t, tanh represents the hyperbolic tangent activation function, b c represents the bias term of the candidate memory unit, C t represents the memory unit at time step t, i t represents the output of the input gate, h t represents the hidden state output at time step t, ⊙ represents the element-wise multiplication operation, and o t represents the output of the output gate.

[0127] S1003: Output the development results of the chronic disease symptoms of the chronic disease patient.

[0128] Among them, during the training process of the machine learning model, the loss function is used to measure the difference between the predicted result and the actual result of the model.

[0129] Among them, those skilled in the art can set the size of the preset loss function value according to the actual situation, and the present invention does not make any limitations.

[0130] It should be noted that by training the LSTM model, the system can capture the evolution law of symptoms over time based on historical symptom data and predict future symptom changes. This enables chronic disease management to not only rely on the current symptom manifestations but also identify possible trends of the disease in advance, providing forward-looking treatment guidance for doctors. By optimizing the loss function, the model can continuously improve the prediction accuracy and further enhance the effectiveness of personalized treatment.

[0131] In a possible implementation manner, the calculation formula of the loss function value is specifically:

[0132]

[0133] Among them, loss represents the loss function value, α represents a hyperparameter, T represents the total number of time steps, represents the predicted value at time step t, y (t) represents the actual label at time step t, represents the loss function value at time step t, C represents the number of classification labels, y i represents the actual value of the i-th category, represents the predicted value of the i-th category.

[0134] Among them, the method based on time series prediction can help doctors make more scientific decisions, timely adjust the treatment plan, thereby effectively managing chronic diseases and improving the long-term health status of patients.

[0135] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0136] In the embodiments of the present invention, by collecting the symptom data of chronic disease patients, real inputs are provided for subsequent analysis. Symptom groups are extracted based on the principal component analysis method and the varimax orthogonal rotation method. Then, a symptom network is constructed with symptoms as nodes and the relationships between symptoms as edges. Synchronous network analysis is used to identify the core symptoms and bridge symptoms in the symptom network. Based on the relationships between the core symptoms and the bridge symptoms, a dynamic symptom network based on the cross-lagged network model is constructed, enabling symptom management not only to be limited to static analysis but also to capture the evolution trend of symptoms over time. Further, influencing factors are introduced into the dynamic symptom network, and subgroup analysis is performed on the dynamic symptom network to identify the core influencing factors that cause differences in the symptom network structure. Based on the core influencing factors, a prediction model for the development trend of chronic diseases based on LSTM is constructed to provide prospective guidance for treatment. Finally, by collecting the clinical symptom data and clinical index data of patients and extracting the core influencing factors, the prediction model for the development trend of chronic diseases based on LSTM predicts and outputs the development results of chronic disease symptoms of chronic disease patients, providing a scientific basis for personalized treatment and long-term management. By analyzing the relationships and interactions between the symptoms of chronic patients, the present invention can effectively manage and treat the symptoms of chronic disease patients, help doctors and patients identify the core symptoms and potential influencing factors, improve the treatment effect and quality of life of chronic patients, and provide a personalized symptom management plan for the unique situation of each patient.

[0137] Refer to the appended Figure 2 illustrates a schematic structural diagram of a chronic patient symptom management system based on symptom network analysis provided by the present invention.

[0138] The present invention also provides a chronic patient symptom management system 20 based on symptom network analysis, which is applied to the above-mentioned chronic patient symptom management method based on symptom network analysis and includes:

[0139] A processor 201.

[0140] A memory 202, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor 201, the chronic patient symptom management method based on symptom network analysis as in the method embodiment is implemented.

[0141] The chronic patient symptom management system 20 based on symptom network analysis provided by the present invention can execute the above-mentioned chronic patient symptom management method based on symptom network analysis and achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate further.

[0142] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0143] In the embodiments of the present invention, by collecting the symptom data of chronic disease patients, real inputs are provided for subsequent analysis. Symptom groups are extracted based on the principal component analysis method and the varimax orthogonal rotation method. Then, with symptoms as nodes and the relationships between symptoms as edges, a symptom network is constructed. Synchronous network analysis is used to identify the core symptoms and bridge symptoms in the symptom network. Based on the relationships between the core symptoms and the bridge symptoms, a dynamic symptom network based on the cross-lagged network model is constructed, enabling symptom management not only to be limited to static analysis but also to capture the evolution trend of symptoms over time. Further, influencing factors are introduced into the dynamic symptom network, and subgroup analysis is performed on the dynamic symptom network to identify the core influencing factors that lead to differences in the symptom network structure. Based on the core influencing factors, an LSTM-based chronic disease development trend prediction model is constructed to provide forward-looking guidance for treatment. Finally, by collecting the clinical symptom data and clinical index data of patients and extracting the core influencing factors, the LSTM-based chronic disease development trend prediction model predicts and outputs the development results of chronic disease symptoms of chronic disease patients, providing a scientific basis for personalized treatment and long-term management. By analyzing the relationships and interactions between the symptoms of chronic patients, the present invention can effectively manage and treat the symptoms of chronic disease patients, help doctors and patients identify the core symptoms and potential influencing factors, improve the treatment effect and quality of life of chronic patients, and provide a personalized symptom management plan for the unique situation of each patient.

[0144] It should be understood that the processor in the embodiments of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0145] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0146] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0147] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context before and after.

[0148] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0149] It should be understood that in various embodiments of the present invention, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0150] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0151] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the devices, apparatuses, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0152] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be in electrical, mechanical, or other forms.

[0153] 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 can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, the functional units in various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0155] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0156] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for chronic patient symptom management based on symptom network analysis as in the method embodiment.

[0157] The computer-readable storage medium provided by the present invention can implement the steps and effects of the method for chronic patient symptom management based on symptom network analysis in the above method embodiment. To avoid repetition, the present invention will not elaborate further.

[0158] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include:

[0159] In the embodiments of the present invention, by collecting the symptom data of chronic disease patients, real inputs are provided for subsequent analysis. Symptom groups are extracted based on the principal component analysis method and the varimax orthogonal rotation method. Then, a symptom network is constructed with symptoms as nodes and the relationships between symptoms as edges. Synchronous network analysis is used to identify the core symptoms and bridge symptoms in the symptom network. Based on the relationships between the core symptoms and bridge symptoms, a dynamic symptom network based on the cross-lagged network model is constructed, enabling symptom management not only to be limited to static analysis but also to capture the evolution trend of symptoms over time. Further, influencing factors are introduced into the dynamic symptom network, and subgroup analysis is performed on the dynamic symptom network to identify the core influencing factors that lead to differences in the symptom network structure. Based on the core influencing factors, a prediction model for the development trend of chronic diseases based on LSTM is constructed to provide prospective guidance for treatment. Finally, by collecting the clinical symptom data and clinical index data of patients and extracting the core influencing factors, the prediction model for the development trend of chronic diseases based on LSTM predicts and outputs the development results of chronic disease symptoms of chronic disease patients, providing a scientific basis for personalized treatment and long-term management. By analyzing the relationships and interactions between the symptoms of chronic patients, the present invention can effectively manage and treat the symptoms of chronic disease patients, help doctors and patients identify the core symptoms and potential influencing factors, improve the treatment effect and quality of life of chronic patients, and provide a personalized symptom management plan for the unique situation of each patient.

[0160] The above is only the specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0161] The following points need to be explained:

[0162] (1) The accompanying drawings of the embodiments of the present invention only relate to the structures involved in the embodiments of the present invention, and other structures can refer to the general design.

[0163] (2) For clarity, in the accompanying drawings used to describe the embodiments of the present invention, the thickness of layers or regions is enlarged or reduced, that is, these drawings are not drawn according to the actual scale. It can be understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element can be "directly" on or under the other element or there can be intermediate elements.

[0164] (3) Without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0165] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for symptom management of chronic patients based on symptom network analysis, characterized in that Including: S1: Collect symptom data of patients with chronic diseases to obtain a symptom assessment dataset for patients with chronic diseases; S2: Combine the principal component analysis method and the varimax orthogonal rotation method to extract symptom clusters from the symptom assessment dataset for patients with chronic diseases, where the symptom clusters include multiple symptoms; S3: Construct a symptom network with each of the symptoms as nodes and the relationships between symptoms as edges; S4: Use the synchronous network analysis algorithm to identify the core symptoms and bridge symptoms in the symptom network; S5: Based on the relationship between the core symptoms and the bridge symptoms, construct a dynamic symptom network based on the cross-lagged network model; S6: Introduce influencing factors into the dynamic symptom network, and based on the relationship between the influencing factors and the symptoms, perform subgroup analysis on the dynamic symptom network to identify the core influencing factors that cause differences in the symptom network structure; S7: According to the core influencing factors, construct a prediction model for the development trend of chronic diseases based on LSTM; S8: Collect real-time clinical symptom data and real-time clinical index data of patients with chronic diseases; S9: Extract real-time core influencing factors from the real-time clinical symptom data and the real-time clinical index data; S10: Input the real-time core influencing factors into the prediction model for the development trend of chronic diseases to predict and output the development result of the chronic disease symptoms of the patients with chronic diseases.

2. The chronic patient symptom management method based on symptom network analysis according to claim 1, characterized in that, The symptom assessment dataset for patients with chronic diseases specifically includes: symptom information, clinical data, and patient personal information.

3. The method for symptom management of chronic patients based on symptom network analysis according to claim 1, wherein The S2 specifically includes: S201: Perform standardization processing on the symptom assessment dataset for patients with chronic diseases; S202: Based on the principal component analysis method, calculate the covariance matrix of the standardized dataset; S203: Calculate the eigenvalues and eigenvectors of the covariance matrix; S204: Select the principal components according to the magnitudes of the eigenvalues; S205: Calculate the weights of the principal components to determine the eigenvectors to be rotated; S206: Rotate the eigenvectors to be rotated through the varimax orthogonal rotation method; S207: With the maximum rotation target as the goal, determine the factor loading matrix; S208: Determine the symptom clusters according to the factor loading matrix.

4. The method for symptom management of chronic patients based on symptom network analysis according to claim 1, characterized in that, The S4 specifically includes: S401: Through linear projection, convert the node features and edge features into hidden features respectively, and input the hidden features into the GAT model: Among them, represents the feature vector of node i at the initial time, U 0 represents the node weight matrix at the initial time, α i represents the input feature vector of node i, u 0 represents the node bias vector at the initial time, represents the feature vector of edge (x, y) in the initial layer, V 0 represents the edge weight matrix at the initial time, β ij represents the input feature vector of edge (x, y), v 0 represents the edge bias vector at the initial time; S402: Calculate the attention weights of each symptom node through the graph attention mechanism of the GAT model: Among them, represents the feature representation of edge (x, y) under the k-th head, LeakyReLU represents the non-linear activation function, and a l represents the scalar coefficient of the l-th layer, Concat represents the concatenation operation, and W l represents the weight matrix of the l-th layer, represents the feature vector of node i in the l-th layer, represents the feature vector of node j in the l-th layer; S403: Perform normalization processing on the attention weights: Among them, represents the normalized attention weight of edge (x, y) under the k-th head and the l-th layer. exp represents the exponential function, and N i represents the set of neighbor nodes of node i; S404: According to the normalized attention weights, update the symptom nodes by introducing asymmetry into the neighborhood aggregation function: Among them, represents the feature vector of node i at layer l + 1, Concat represents the concatenation operation, ELU represents the activation function, and U k,l represents the weight matrix of the k-th attention head at layer l; S405: According to the updated symptom nodes, calculate the degree centrality and betweenness centrality respectively, and determine the core symptoms and the bridge symptoms: CoreSymptoms = {i|DegreeCentrality(i) > θ} BridgeSymptoms = {i|Betweenness(i) > θ′} Among them, CoreSymptoms represents the set of core symptoms, DegreeCentrality(i) represents the degree centrality of node i, θ represents the core symptom threshold, BridgeSymptoms represents the set of bridge symptoms, Betweenness(i) represents the betweenness centrality of node i, and θ′ represents the bridge symptom threshold.

5. The method for symptom management of chronic patients based on symptom network analysis according to claim 1, wherein The dynamic symptom network based on the cross-lagged network model is specifically as follows: Among them, represents the feature vector of node i at the (l - 1)-th layer, represents the feature vector of node j at the (l - 1)-th layer, and edgefeatures represents edge features.

6. The chronic patient symptom management method based on symptom network analysis according to claim 1, characterized in that, The S6 specifically includes: S601: Introduce the influencing factor into the node features of the dynamic symptom network by means of feature splicing; S602: Update the node features; S603: Divide the updated node features into multiple subgroups; S604: Analyze each of the subgroups to identify multiple patient groups; S605: Calculate the differences between the symptom network structures of each of the subgroups to identify the core influencing factors.

7. The method for symptom management of chronic patients based on symptom network analysis according to claim 6, characterized in that The S10 specifically includes: S1001: Smooth the core influencing factor to obtain a time parameter vector; S1002: Input the time parameter vector into the chronic disease development trend prediction model for training until the loss function value is less than the preset loss function value; S1003: Output the development result of the chronic disease symptoms of the chronic disease patient.

8. The method for symptom management of chronic patients based on symptom network analysis according to claim 7, wherein The calculation formula of the loss function value is specifically: Among them, loss represents the value of the loss function, α represents the hyperparameter, T represents the total number of time steps, represents the predicted value at time step t, y (t) represents the actual label at time step t, represents the value of the loss function at time step t, C represents the number of classification labels, y i represents the actual value of the i-th category, represents the predicted value of the i-th category.

9. A symptom management system for chronic patients based on symptom network analysis, characterized in that, Including: A processor; A memory, on which computer-readable instructions are stored. When the computer-readable instructions are executed by the processor, the chronic patient symptom management method based on symptom network analysis as described in any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, the chronic patient symptom management method based on symptom network analysis as described in any one of claims 1 to 8 is implemented.

Images