Personalized hypertensive nephropathy risk assessment method and system based on graph convolutional network

Abstract

The present invention discloses a personalized hypertensive nephropathy risk assessment method and system based on graph convolutional network, the assessment method comprising: collecting modal data related to hypertensive nephropathy and performing preprocessing; constructing a patient similarity graph; performing multi-source information fusion and domain knowledge integration; performing sparse processing on the graph structure and normalizing the adjacency matrix; capturing the evolution trend of patient characteristics over time through temporal graph convolutional network and long short-term memory network modeling; strengthening the transmission of important features through self-attention mechanism and graph regularization term; adopting multi-head attention mechanism to capture different types of relationship patterns; optimizing model parameters using Adam optimization algorithm and regularization method; and optimizing risk assessment results using weighted stacking and adaptive attention mechanism. The present invention adopts the above-mentioned risk assessment method to achieve accurate assessment of hypertensive nephropathy risk by constructing a multi-dimensional, dynamically changing patient similarity graph.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a personalized hypertensive nephropathy risk assessment method and system based on graph convolutional networks. Background Art

[0002] Hypertensive nephropathy (HN) is kidney damage caused by long-term, uncontrolled hypertension. It has become one of the leading causes of chronic kidney disease (CKD) and end-stage renal disease (ESRD) worldwide. Pathophysiologically, hypertension-induced kidney damage primarily manifests as arteriosclerosis, glomerulosclerosis, and interstitial fibrosis. These lesions lead to a gradual decrease in glomerular filtration rate (GFR) and ultimately renal failure. Because early symptoms of HN are often subtle, many patients are not diagnosed until the disease has progressed to the late stage, missing the optimal time for intervention and treatment. Therefore, early identification of high-risk individuals and risk assessment are crucial to delaying disease progression and improving patient outcomes. Currently, risk assessment for hypertensive nephropathy primarily relies on traditional statistical methods and scoring systems based on clinical experience, such as the Framingham risk score and the Epidemiology Collaboration (Eq. ). However, these methods have limitations in processing high-dimensional, multimodal, and nonlinear data, making it difficult to accurately capture the complex similarities between patients and the dynamic characteristics of disease progression. Furthermore, heterogeneity among individual patients, such as genetic background, environmental factors, and lifestyle, also affects the accuracy of risk assessment. With the rapid development of artificial intelligence and deep learning technologies, advanced models such as deep feedforward neural networks (DFNNs) and graph neural networks (GNNs) have been introduced into the medical field, providing new insights for risk assessment and disease prediction. In particular, graph convolutional networks (GCNs) excel in processing non-Euclidean spatial data and capturing high-order neighborhood relationships, and have achieved success in fields such as bioinformatics, drug discovery, and social network analysis. However, single models still have shortcomings in handling multi-source heterogeneous data, dynamically changing disease characteristics, and personalized risk prediction, making them difficult to meet the needs of clinical practice. Summary of the Invention

[0003] The purpose of the present invention is to provide a personalized hypertensive nephropathy risk assessment method and system based on graph convolutional networks to solve the problems mentioned in the background technology.

[0004] To achieve the above objectives, the present invention provides a personalized hypertensive nephropathy risk assessment method based on graph convolutional networks, comprising the following steps:

[0005] Step 100: collecting modal data related to hypertensive nephropathy, preprocessing the modal data, and then establishing a patient feature matrix;

[0006] Step 200: Constructing a patient similarity graph by measuring and encoding the similarities between patients and enhancing patient feature representation using a graph convolutional network;

[0007] Step 300: performing multi-source information fusion and domain knowledge integration, wherein the multi-source information fusion includes structured data fusion, unstructured data fusion and image feature fusion;

[0008] Step 400: performing a sparse processing on the graph structure and then normalizing the adjacency matrix;

[0009] Step 500: Process the time series data and capture the temporal evolution trend of patient characteristics through temporal graph convolutional networks and long short-term memory networks.

[0010] Step 600: Strengthen the transfer of important features through the self-attention mechanism and graph regularization term;

[0011] Step 700: In each layer of the temporal graph convolutional network, a multi-head attention mechanism is used to capture different types of relationship patterns;

[0012] Step 800: Optimize GCN parameters, base model parameters, ensemble weights, and personalized attention parameters using the Adam optimization algorithm and regularization method.

[0013] Step 900: Using weighted stacking and adaptive attention mechanisms, and learning a personalized loss function, optimize the final risk assessment result;

[0014] Step 800 is specifically as follows:

[0015] The Adam optimization algorithm is used to optimize all model parameters through the gradient descent method. The formula is as follows:

[0016]

[0017] Where η is the learning rate, and are the first-order and second-order moment estimates, ∈ is a small constant to prevent division by zero, and θ t+1 and θ t denote the model parameters in t+1 and t iterations respectively;

[0018] At the same time, a variety of regularization techniques are used, including L2 regularization and Dropout method. Dropout is introduced in each hidden layer to randomly discard some neurons to enhance the generalization ability of the model:

[0019] h ′ =h⊙d,d~Bernoulli(p);

[0020] Among them, h ′ Represents the feature vector after Dropout, h represents the original feature vector, and d represents the Dropout mask vector, whose elements conform to the Bernoulli distribution.

[0021] Preferably, in step 100:

[0022] The modality data includes biomarker data, electronic health records, lifestyle data, and imaging data; the biomarker data includes serum creatinine, urine protein, blood pressure, blood glucose levels, and blood lipid indicators; the electronic health records include medical history, diagnosis information, medication records, and complication information; the lifestyle data includes dietary habits, exercise frequency, smoking and drinking habits, and sleep patterns; the imaging data includes renal ultrasound images and magnetic resonance imaging;

[0023] The pre-processing step comprises:

[0024] For different modal data, missing data are filled by using missing value filling and multiple imputation methods;

[0025] Identify and handle outliers through statistical method box plot analysis;

[0026] Perform Z-score standardization on numerical features to ensure that all features are on the same scale. The formula is as follows:

[0027]

[0028] Among them, f′ i is the standardized modal data of type i, f i represents the original i-th modal data, μ i and σ i are the mean and standard deviation of the i-th modal data respectively;

[0029] The patient feature matrix is Where N is the number of patients and D is the feature dimension.

[0030] Preferably, in step 200:

[0031] The construction of the patient similarity graph includes graph structure definition and similarity measurement;

[0032] The graph structure definition is: the nodes of the patient similarity graph represent each patient, and the feature vector of the node is the preprocessed patient feature, which is expressed as Among them, x i represents the patient features of the i-th modality data, and the edges of the patient similarity graph represent the similarities between patients;

[0033] Similarity measures include:

[0034] Biomarker similarity metric: Based on the similarity metric of patient biomarkers, the Gaussian similarity function is used, and the formula is as follows:

[0035]

[0036] Among them, S biomarker (i, j) represents the Gaussian similarity measure of the patient’s biomarkers, x i,biomarker 、x j,biomarker Represents the feature vectors of the patient's biomarkers at nodes i and j, σ biomarker represents the width of the Gaussian kernel of the patient's biomarker;

[0037] Lifestyle similarity measurement: Based on the similarity measurement of lifestyle data, cosine similarity is used, and the formula is as follows:

[0038]

[0039] Among them, S lifestyle (i, j) represents the similarity measure of patients’ lifestyles, x i,lifestyle 、x j,lifestyle Represent the feature vectors of the patient’s lifestyle at nodes i and j respectively;

[0040] Medical history similarity measurement: Based on the similarity measurement of the patient's medical history information, Jaccard similarity is used, and the formula is as follows:

[0041]

[0042] Among them, S history (i,j) The feature vector of the patient’s medical history information at nodes i and j, Represent the patient’s medical history set at nodes i and j respectively;

[0043] Dynamic similarity measurement: Using the dynamic weight adjustment mechanism, the weights of each similarity measurement are dynamically adjusted by learning weight parameters. The formula is as follows:

[0044]

[0045] Among them, A ijrepresents the variable weight between node i and node j in the patient similarity graph, x ik represents the kth eigenvalue of node i, x jk represents the kth eigenvalue of node j, K is the number of dimensions of similarity measurement, w k is the weight of the k-th similarity dimension.

[0046] Preferably, in step 300:

[0047] The structured data fusion integrates biomarkers, electronic health records, etc., and calculates the similarity between patients using the similarity measurement method; the unstructured data fusion uses natural language processing to encode text data and incorporates the extracted high-dimensional text features into the similarity measurement; the image feature fusion extracts deep features of image data through pre-trained convolutional neural networks, and fuses the deep features with other modal data to enhance the depth and diversity of the similarity measurement;

[0048] The domain knowledge integration includes defining the similarity between patients based on clinical pathways and progression trajectories, adjusting the similarity measurement weights through expert knowledge guidance, and dividing the similarity into different levels, calculating them separately and integrating them into the overall similarity score.

[0049] Preferably, in step 400:

[0050] The thinning process includes:

[0051] By setting a similarity threshold θ, only edges with similarity higher than θ are retained. The formula is as follows:

[0052]

[0053] The K nearest neighbor retention strategy is used. For each node, only its K most similar neighbors are retained. The formula is as follows:

[0054]

[0055] Among them, Top-K(A i,: ) represents the K nodes with the highest similarity between node i and other nodes;

[0056] The steps of the normalization process are as follows:

[0057]

[0058] Where I is the identity matrix, is the degree matrix, is the normalized adjacency matrix, represents the normalized adjacency matrix, Represents the normalized adjacency matrix The element at row i and column j in .

[0059] Preferably, step 500 is specifically as follows:

[0060] Step 510: Divide the patient data into multiple time windows according to time. Each time window t corresponds to a graph structure at a moment. In each time window t, recalculate the similarity measure based on the latest patient feature data and update the adjacency matrix A. t ,Through dynamic weight adjustment and multi-source information fusion, it reflects the latest patient characteristics and similarity relationships;

[0061] Step 520: Combined with the time series information, the graph structure data in the time window sequence is processed by the time series graph convolutional network to capture the time series evolution of the patient characteristics. The formula is as follows:

[0062]

[0063] in, is the node feature matrix of the l+1th layer, is the normalized adjacency matrix of the t-th time window, is the node feature matrix of the lth layer, W (l) represents the weight matrix of the lth layer;

[0064] Step 530: Introduce a memory mechanism to serialize the graph convolution outputs of different time windows to capture long-term dependencies and feature evolution trends.

[0065] Preferably, step 600 is specifically as follows:

[0066] Step 610: Use the self-attention mechanism to enhance node features and dynamically adjust the importance of different feature dimensions. The formula is as follows:

[0067]

[0068] Among them, α ij is the self-attention weight, h j is the feature vector of node j, W represents the weight matrix, h′ i Represents the new feature vector of node i after being enhanced by the self-attention mechanism;

[0069] Step 620: Add a graph regularization term to the loss function to make the features of similar nodes closer:

[0070]

[0071] Among them, ε is the edge set in the graph, h i The feature vector of node i, represents the graph regularization loss term;

[0072] Step 630: Randomly delete or add edge and node noise to enhance the patient similarity graph data, the method includes edge perturbation, node perturbation and feature perturbation.

[0073] Preferably, step 700 is specifically as follows:

[0074] Step 710: In each layer of the temporal graph convolutional network (GCN), a multi-head attention mechanism is used to calculate different attention weights in parallel through multiple attention heads to capture different types of relationship patterns. The formula is as follows:

[0075] MultiHead(Q,K,V)=Concat(head1,…,head h )W O ;

[0076] Among them, head i =Attention(Q i ,K i ,V i ), h is the number of attention heads, W O is the output weight matrix, MultiHead(Q,K,V) represents the multi-head attention mechanism, Q is the query vector, K is the key vector, and V is the value vector;

[0077] Step 720: Generate adversarial examples to improve the model's robustness to input perturbations. Specific methods include the fast gradient sign method and projected gradient descent:

[0078] Fast Gradient Sign Method: Generate adversarial samples using gradient information:

[0079]

[0080] Among them, x represents the original input sample, x ′ represents an adversarial sample, generated by adding a small perturbation to the original input sample. Denotes the loss function to measure the difference between the model prediction and the true label, ∈ denotes the perturbation amplitude, Represents the loss function Partial derivative with respect to the original input sample x;

[0081] Projected Gradient Descent: Iteratively generates adversarial samples, the formula is as follows:

[0082]

[0083] in, is the constraint set of the adversarial sample, α is the step size, x t ′ represents the adversarial sample generated by t iterations, x t′ +1 represents the adversarial sample generated by t+1 iterations;

[0084] Step 730: Perform deep learning integration and personalized risk assessment to ensure collaborative optimization of the model at the integration and personalization levels;

[0085] Step 730 includes the following steps:

[0086] Step 731: Randomly initialize all model parameters, including GCN parameters, base model parameters, integration weight coefficient α and personalized attention parameters;

[0087] Step 732: Optimize the integrated prediction loss and personalized loss to obtain the total loss function. for:

[0088]

[0089] in, is the loss function for ensemble prediction, is the loss function of personalized prediction, γ is the weight of personalized loss, and λ is the regularization parameter.

[0090] Preferably, the specific steps of step 900 are as follows:

[0091] Step 910: Design a secondary model as the weight coefficient for learning the base model. The final prediction result is the weighted sum of the prediction results of each base model:

[0092]

[0093] in, Represents the final prediction result, α i Represents the weight coefficient of the i-th basis model, y i represents the prediction result of the i-th base model, m represents the total number of base models in the ensemble method,

[0094] Through back-propagation and optimization algorithms, the weight coefficient α is learned to minimize the loss function of the integrated prediction;

[0095] Based on the individual characteristics of the patient, a customized risk prediction is made for each patient through an adaptive attention mechanism. The specific steps include feature extraction, attention weight calculation, and personalized weighting. The formula is as follows:

[0096] e=tanh(W1p+b1);

[0097] α=softmax(W2e+b2);

[0098]

[0099] Where e represents the intermediate representation vector obtained after linear transformation and nonlinear activation of the patient's individual features, p is the patient's individual feature vector, W1 and W2 are learnable weight matrices, b1 and b2 are bias vectors, and α is the attention weight;

[0100] Step 920: Design a personalized loss function, combining the error between the personalized prediction result and the true label. The formula is as follows:

[0101]

[0102] Where N is the number of samples, y k is the true label of the k-th sample, Personalized prediction results.

[0103] The present invention also provides a personalized hypertensive nephropathy risk assessment system based on graph convolutional networks, comprising:

[0104] Data acquisition and preprocessing module: used to collect modal data related to hypertensive nephropathy, preprocess the modal data, and then establish the patient feature matrix;

[0105] Patient Similarity Graph Construction Module: This module is used to construct a patient similarity graph by measuring and encoding the similarities between patients and enhancing the patient feature representation using a graph convolutional network.

[0106] Multi-source information fusion module: used for multi-source information fusion and domain knowledge integration;

[0107] Sparseness and normalization processing module: used to perform sparse processing on the graph structure and then normalize the adjacency matrix;

[0108] Time series processing module: used to process time series data and capture the temporal evolution trend of patient characteristics through temporal graph convolutional networks and long short-term memory networks;

[0109] Feature transfer enhancement module: used to enhance the transfer of important features through the self-attention mechanism and graph regularization term;

[0110] Relationship pattern capture module: It is used to capture different types of relationship patterns using a multi-head attention mechanism in each layer of the temporal graph convolutional network;

[0111] Parameter optimization module: used to optimize GCN parameters, base model parameters, ensemble weights, and personalized attention parameters through the Adam optimization algorithm and regularization method;

[0112] Personalized risk assessment module: It uses weighted stacking and adaptive attention mechanisms and learns personalized loss functions to optimize the final risk assessment results.

[0113] Therefore, the present invention adopts the above-mentioned personalized hypertensive nephropathy risk assessment method and system based on graph convolutional network, which has the following beneficial effects:

[0114] (1) By constructing a multi-dimensional, dynamically updated patient similarity graph, fusing multi-source heterogeneous data, and adopting a weighted stacking integration strategy and adaptive attention mechanism, we can achieve accurate and personalized assessment of the risk of hypertensive nephropathy.

[0115] (2) It not only fully utilizes the advantages of GCN in capturing high-order relationships and complex similarities, but also improves the model's sensitivity and adaptability to changes in patient characteristics through multimodal data fusion and dynamic graph structure updates.

[0116] (3) This method will provide clinicians with an efficient and reliable risk prediction tool for hypertensive nephropathy, assisting in early diagnosis and intervention, ultimately improving patient prognosis and reducing the burden on the medical system.

[0117] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0118] Figure 1 This is a flowchart of a personalized hypertensive nephropathy risk assessment method based on a graph convolutional network according to an embodiment of the present invention;

[0119] Figure 2 This is a model framework diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0120] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0121] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0122] Example

[0123] like Figure 1-2 As shown, the present invention provides a personalized hypertensive nephropathy risk assessment method based on a graph convolutional network (GCN), aiming to achieve accurate assessment of hypertensive nephropathy (HN) risk. The core innovation lies in constructing a patient similarity graph (PSG) and using a GCN to enhance patient feature representation, thereby achieving personalized risk prediction. The entire process runs through data collection and preprocessing, construction of a patient similarity graph, feature enhancement of a GCN, integration of a deep learning prediction module, and personalized risk assessment. Ultimately, efficient model collaboration is achieved through mathematical optimization and training strategies.

[0124] The risk assessment methodology includes the following steps:

[0125] Step 100: collecting modal data related to hypertensive nephropathy, preprocessing the modal data, and then establishing a patient feature matrix;

[0126] Modal data includes biomarker data, electronic health records, lifestyle data, and imaging data. Biomarker data includes serum creatinine, urine protein, blood pressure, blood sugar levels, blood lipid indicators, etc. Electronic Health Records (EHR) include medical history, diagnostic information, medication records, comorbidity information, etc. Lifestyle data includes dietary habits (e.g., whether a high-salt diet is consumed), exercise frequency, smoking and drinking habits, sleep patterns, etc. Imaging data includes renal ultrasound images, magnetic resonance imaging (MRI), etc.

[0127] The preprocessing steps include filling missing values, handling outliers, and standardization. The specific process is as follows:

[0128] For different modal data, missing data are filled by using missing value filling and multiple interpolation methods to ensure data integrity; outliers are identified and processed through statistical boxplot analysis to prevent them from having an adverse impact on model training; numerical features are Z-score standardized to ensure that all features are on the same scale. The formula is as follows:

[0129]

[0130] Among them, f′ i is the standardized modal data of type i, f irepresents the original i-th modal data, μ i and σ i are the mean and standard deviation of the i-th mode data respectively.

[0131] After preprocessing, generate the patient feature matrix Where N is the number of patients and D is the feature dimension.

[0132] Step 200: By systematically measuring and encoding the similarities between patients and enhancing patient feature representation using a graph convolutional network (GCN), a patient similarity graph (PSG) is constructed to achieve personalized risk assessment. The construction of the PSG is one of the core concepts of this method.

[0133] The construction of PSG involves several key steps, including graph structure definition and similarity measurement.

[0134] The graph structure definition refers to the definition of the PSG graph structure. The nodes of the PSG represent each patient, and the feature vector of the node is the preprocessed patient feature, which is expressed as Among them, x i Represents the patient characteristics of the i-th modality data. The edges of PSG represent the similarity between patients, and the weight of the edge reflects the degree of similarity between patients.

[0135] Similarity measurement is key to PSG construction, determining the closeness of connections between patients and their representation in the graph. Traditional similarity measures such as cosine similarity and Euclidean distance, while effective, may not fully reflect the true relationships between patients in complex, multidimensional data. Therefore, this method utilizes an innovative multidimensional similarity measurement with dynamic weight adjustment, comprehensively considering multiple dimensions such as biomarkers, lifestyle, and medical history to ensure the comprehensiveness and accuracy of similarity assessments. Specifically, it includes the following aspects:

[0136] Biomarker similarity: Based on the similarity measure of patients' biomarkers (such as serum creatinine, urine protein, etc.), a Gaussian similarity function is used:

[0137]

[0138] Among them, S biomarker (i, j) represents the Gaussian similarity measure of the patient’s biomarkers, x i,biomarker 、x j,biomarker Represents the feature vectors of the patient's biomarkers at nodes i and j, σ biomarker represents the width of the Gaussian kernel for the patient's biomarker.

[0139] Lifestyle similarity: A similarity measure based on lifestyle data (diet, exercise habits), using cosine similarity:

[0140]

[0141] Among them, S lifestyle (i, j) represents the similarity measure of patients’ lifestyles, x i,lifestyle 、x j,lifestyle Represent the feature vectors of the patient's lifestyle at nodes i and j respectively.

[0142] Medical history similarity: Based on the similarity measurement of the patient's medical history information (such as previous diseases and family medical history), the Jaccard similarity formula is as follows:

[0143]

[0144] Among them, S history (i,j) The feature vector of the patient’s medical history information at nodes i and j, Represent the medical history sets of the patient at nodes i and j respectively.

[0145] Dynamic similarity measurement: Taking into account the different importance of different feature dimensions, a dynamic weight adjustment mechanism is used to dynamically adjust the weights of each similarity measurement by learning weight parameters. The formula is as follows:

[0146]

[0147] Among them, A ij represents the variable weight between node i and node j in the patient similarity graph, which is used to quantify the similarity between patients, x ik represents the kth eigenvalue of node i, x jk represents the kth eigenvalue of node j, K is the number of dimensions of similarity measurement, w k is the weight of the k-th similarity dimension, which is adaptively learned through the training process to reflect the contribution of different features to the similarity.

[0148] Step 300: Multi-source information fusion and domain knowledge integration are performed. Multi-source information fusion includes structured data fusion, unstructured data fusion, and image feature fusion. During PSG construction, a single data source often cannot fully reflect the similarity relationships between patients. Therefore, this method further enhances the comprehensiveness and depth of similarity assessment by integrating multi-source information and domain knowledge.

[0149] Structured data fusion integrates biomarkers, electronic health records, etc., and calculates the similarity between patients through the above-mentioned multidimensional similarity measurement method; unstructured data fusion uses natural language processing technology (such as BERT) to encode text data (medical record descriptions) and incorporates the extracted high-dimensional text features into the similarity measurement; image feature fusion extracts deep features of image data through pre-trained convolutional neural networks, and fuses these features with other modal data to enhance the depth and diversity of similarity measurement.

[0150] In addition, domain knowledge integration includes defining the similarity between patients based on clinical pathways and progression trajectories, adjusting the similarity measurement weights through expert knowledge guidance, and hierarchical similarity, dividing the similarity into different levels (such as basic feature similarity and pathological progression similarity), calculating them separately and integrating them into the overall similarity score to ensure that the model fully captures the similarity between patients at different levels.

[0151] Step 400: Perform a sparse processing on the graph structure and then normalize the adjacency matrix.

[0152] In order to reduce the complexity of the graph and highlight the key similarity relationships, a graph structure sparsification strategy is introduced. Specifically, first, by setting a similarity threshold θ, only edges with similarity higher than θ are retained. The formula is as follows:

[0153]

[0154] At the same time, the K nearest neighbor retention strategy is adopted. For each node, only its K most similar neighbors are retained. The formula is as follows:

[0155]

[0156] Among them, Top-K(A i,: ) represents the K nodes with the highest similarity between node i and other nodes.

[0157] By setting the similarity threshold and retaining the K-nearest neighbor strategy, the edge structure of PSG is effectively sparse, which not only reduces the computational complexity but also ensures that only key similarity relationships are retained in the graph, thereby improving the efficiency and performance of the model.

[0158] Next, the adjacency matrix is ​​normalized to meet the input requirements of the graph convolutional network. The normalization steps are as follows:

[0159]

[0160]

[0161] Where I is the identity matrix, is the degree matrix, is the normalized adjacency matrix. Represents the adjacency matrix after normalization. Normalization aims to balance the degree of nodes, ensure the effective transmission of information in graph convolution, and avoid excessive influence of high-connectivity nodes on feature updates. Represents the normalized adjacency matrix This normalization method can balance the degree of nodes, ensure the effective transmission of information in graph convolution, and avoid the excessive influence of high-degree nodes on feature updates.

[0162] Step 500: Process the time series data and use a time series graph convolutional network and a long short-term memory network to model the patient's characteristics over time. This step can reflect the dynamic characteristics of the patient's health status and lifestyle over time. Specifically, it includes the following steps:

[0163] Step 510: Divide the patient data into multiple time windows (such as monthly, quarterly), each time window corresponds to a graph structure at a moment, there are T time windows, each time window corresponds to an adjacency matrix A t In each time window t, the similarity measure is recalculated based on the latest patient feature data and the adjacency matrix A is updated. t ,Through dynamic weight adjustment and multi-source information fusion, it ensures that the graph structure can reflect the latest patient characteristics and similarity relationships in real time.

[0164] Step 520: Combined with the temporal information, a temporal graph convolutional network (such as Dynamic GCN or Temporal GCN) is used to process the graph structure data in the time window sequence to capture the temporal evolution of the patient characteristics. The formula is as follows:

[0165]

[0166] in, is the node feature matrix of the l+1th layer, is the normalized adjacency matrix of the t-th time window, is the node feature matrix of the lth layer, W (l) Represents the weight matrix of layer l. Through the temporal graph convolutional network, the model can capture the evolution trend of patient characteristics over time, improving the timeliness and accuracy of risk assessment.

[0167] Step 530: Introduce a memory mechanism (such as a long short-term memory network, LSTM) to serialize the graph convolution outputs of different time windows, capture long-term dependencies and feature evolution trends, and further enhance the model's time series prediction capabilities.

[0168] Step 600: Through the self-attention mechanism and graph regularization term, the transmission of important features is strengthened, the expressiveness of PSG and the robustness of the model are further improved, and the adaptability of the model to changes in graph results is improved.

[0169] Step 610: Utilize the self-attention mechanism to enhance node features and dynamically adjust the importance of different feature dimensions:

[0170]

[0171] Among them, α ij is the self-attention weight, h j is the eigenvector of node j, W represents the weight matrix, h ′ i The new feature vector of node i after being enhanced by the self-attention mechanism combines important feature information from the node itself and its neighbors. The self-attention mechanism dynamically adjusts weights based on feature similarity between nodes, strengthening information transfer between important neighbors.

[0172] Step 620: Add a graph regularization term to the loss function to encourage similar nodes to have closer features:

[0173]

[0174] Where ε is the edge set in the graph, h i The feature vector of node i, The graph regularization loss term constrains the model to keep the feature vectors of connected nodes (i.e., similar patients) as close as possible. Through graph regularization, the model can better capture the similarity between nodes, improving the consistency and robustness of feature representation.

[0175] Step 630: Graph data augmentation methods such as randomly deleting or adding edge and node noise are used to improve the model's adaptability to graph structure changes and enhance its generalization capabilities. Specific methods include edge perturbation, node perturbation, and feature perturbation.

[0176] Step 700: In each layer of the temporal graph convolutional network, a multi-head attention mechanism is used to capture different types of relationship patterns and enhance the diversity of feature representation.

[0177] Step 710: To further enhance the expressiveness and flexibility of the graph convolutional network, a multi-head attention mechanism is used in each layer of the temporal graph convolutional network (GCN). Multiple attention heads are used to calculate different attention weights in parallel to capture different types of relationship patterns.

[0178] MultiHead(Q,K,V)=Concat(head1,…,head h )W O ;

[0179] Among them, head i =Attention(Q i ,K i ,V i ), h is the number of attention heads, W O is the output weight matrix. MultiHead(Q,K,V) represents a multi-head attention mechanism, where Q is the query vector, which is used to calculate the attention weight with the key vector, K is the key vector, which together with the query vector determines the attention weight, and V is the value vector, which is weighted and summed according to the attention weight to generate the final attention output.

[0180] MultiHead(Q,K,V) uses multiple parallel attention heads to capture diverse relational patterns from different representation subspaces. This multi-head attention mechanism allows the model to capture different relational patterns in parallel, improving the diversity and richness of feature representation.

[0181] Step 720: Improve the model's robustness to input perturbations by generating adversarial examples. Specific methods include the Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD).

[0182] FGSM (Fast Gradient Sign Method): Generate adversarial samples using gradient information:

[0183]

[0184] Here, x represents the original input sample, which is the unperturbed input data received by the model during training or prediction. ′ Adversarial examples are generated by adding small perturbations to the original input samples, aiming to fool the model into making incorrect predictions. Denotes the loss function that measures the difference between the model prediction and the true label, and ∈ denotes the perturbation amplitude, which is a small constant used to control the size of the perturbation added to the original input sample. Represents the loss function The partial derivative with respect to the original input sample x is used to indicate how to adjust x to increase or decrease the loss.

[0185] PGD ​​(Projected Gradient Descent): Iteratively generates adversarial samples. The formula is as follows:

[0186]

[0187] in, is the constraint set of the adversarial sample, α is the step size, x t ′ represents the adversarial sample generated by t iterations, x t ′ +1 represents the adversarial sample generated by t+1 iterations. Through adversarial training, the model can better adapt to the diversity and uncertainty of input data.

[0188] Step 730: Perform deep learning integration and personalized risk assessment to ensure collaborative optimization of the model at the integration and personalization levels, including the following steps:

[0189] Step 731: Randomly initialize all model parameters, including GCN parameters, base model parameters, integration weight coefficient α and personalized attention parameters;

[0190] Step 732: Optimize the integrated prediction loss and personalized loss to obtain the total loss function. for:

[0191]

[0192] in, is the loss function for ensemble prediction, is the loss function of personalized prediction, γ is the weight of personalized loss, and λ is the regularization parameter.

[0193] Step 800: Adopt the Adam optimization algorithm to optimize all model parameters by gradient descent method. The formula is as follows:

[0194]

[0195] Where η is the learning rate, and are the first-order and second-order moment estimates, ∈ is a small constant to prevent division by zero, and θ t+1 and θ t denote the model parameters in iterations t+1 and t, respectively.

[0196] At the same time, in order to prevent the model from overfitting, a variety of regularization techniques are used, including L2 regularization and Dropout method. Dropout is introduced in each hidden layer to randomly discard a part of neurons to enhance the generalization ability of the model:

[0197] h ′ =h⊙d,d~Bernoulli(p);

[0198] Among them, h ′ Represents the feature vector after Dropout, h represents the original feature vector, and d represents the Dropout mask vector, whose elements conform to the Bernoulli distribution.

[0199] Through joint optimization and collaborative training strategies, the model can simultaneously optimize GCN parameters, base model parameters, integration weights and personalized attention parameters, improving the performance and generalization ability of the overall model.

[0200] Step 900: Perform integrated and personalized risk assessment, ensuring the accuracy of personalized predictions through weighted stacking and adaptive attention mechanisms. Optimize the final risk assessment results by learning a personalized loss function. The specific steps are as follows:

[0201] Step 910: Use a weighted stacking strategy to weight the prediction results of multiple base models to improve the overall prediction performance. Specifically, a secondary model (Meta-Learner) is designed to learn the weight coefficients of the base models. The final prediction result is the weighted sum of the prediction results of each base model:

[0202]

[0203] in, Represents the final prediction result, α i Represents the weight coefficient of the i-th basis model, y i represents the prediction result of the i-th base model, m represents the total number of base models in the ensemble method,

[0204] Through back-propagation and optimization algorithms, the weight coefficient α is learned to minimize the loss function of the ensemble prediction.

[0205] By combining individual patient characteristics and using an adaptive attention mechanism, we can achieve customized risk prediction for each patient. The specific steps include feature extraction, attention weight calculation, and personalized weighting. The formula is as follows:

[0206] e=tanh(W1p+b1);

[0207] α=softmax(W2e+b2);

[0208]

[0209] Among them, e represents the intermediate representation vector obtained after linear transformation and nonlinear activation of the patient's individual features, which is used as the basis for calculating the attention weight, p is the patient's individual feature vector, W1 and W2 are learnable weight matrices, b1 and b2 are bias vectors, and α is the attention weight.

[0210] Step 920: To optimize the personalized prediction results, a personalized loss function is designed, combining the error between the personalized prediction results and the true label:

[0211]

[0212] Where N is the number of samples, y k is the true label of the k-th sample, The above method can ultimately provide a personalized prediction result for the risk of hypertensive nephropathy.

[0213] The present invention also provides a personalized hypertensive nephropathy risk assessment system based on graph convolutional networks, comprising:

[0214] Data acquisition and preprocessing module: used to collect modal data related to hypertensive nephropathy, preprocess the modal data, and then establish the patient feature matrix;

[0215] Patient Similarity Graph Construction Module: This module is used to construct a patient similarity graph by measuring and encoding the similarities between patients and enhancing the patient feature representation using a graph convolutional network.

[0216] Multi-source information fusion module: used for multi-source information fusion and domain knowledge integration;

[0217] Sparseness and normalization processing module: used to perform sparse processing on the graph structure and then normalize the adjacency matrix;

[0218] Time series processing module: used to process time series data and capture the temporal evolution trend of patient characteristics through temporal graph convolutional networks and long short-term memory networks;

[0219] Feature transfer enhancement module: used to enhance the transfer of important features through the self-attention mechanism and graph regularization term;

[0220] Relationship pattern capture module: It is used to capture different types of relationship patterns using a multi-head attention mechanism in each layer of the temporal graph convolutional network;

[0221] Parameter optimization module: used to optimize GCN parameters, base model parameters, ensemble weights, and personalized attention parameters through the Adam optimization algorithm and regularization method;

[0222] Personalized risk assessment module: It uses weighted stacking and adaptive attention mechanisms and learns personalized loss functions to optimize the final risk assessment results.

[0223] Therefore, the present invention adopts the above-mentioned personalized hypertensive nephropathy risk assessment method and system based on graph convolutional network, adopts multi-source information fusion and complex similarity modeling, combines time series information and personalized adjustment, and can provide accurate hypertensive nephropathy risk assessment.

[0224] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A personalized hypertensive nephropathy risk assessment method based on graph convolutional networks, characterized in that: The following steps are involved: Step 100: collecting modal data related to hypertensive nephropathy, preprocessing the modal data, and then establishing a patient feature matrix; Step 200: Constructing a patient similarity graph by measuring and encoding the similarities between patients and enhancing patient feature representation using a graph convolutional network; Step 300: performing multi-source information fusion and domain knowledge integration, wherein the multi-source information fusion includes structured data fusion, unstructured data fusion and image feature fusion; Step 400: performing a sparse processing on the graph structure and then normalizing the adjacency matrix; Step 500: Process the time series data and capture the temporal evolution trend of patient characteristics through temporal graph convolutional networks and long short-term memory networks. Step 600: Strengthen the transfer of important features through the self-attention mechanism and graph regularization term; Step 700: In each layer of the temporal graph convolutional network, a multi-head attention mechanism is used to capture different types of relationship patterns; Step 800: Optimize GCN parameters, base model parameters, ensemble weights, and personalized attention parameters using the Adam optimization algorithm and regularization method. Step 900: Using weighted stacking and adaptive attention mechanisms, and learning a personalized loss function, optimize the final risk assessment result; Step 800 is specifically as follows: The Adam optimization algorithm is used to optimize all model parameters through the gradient descent method. The formula is: ； in, is the learning rate, and are the first-order and second-order moment estimates, respectively. To prevent division by zero for small constants, and Respectively expressed in Second and Model parameters in iterations; At the same time, a variety of regularization techniques are used, including L2 regularization and Dropout method. Dropout is introduced in each hidden layer to randomly discard a part of neurons to enhance the generalization ability of the model, and the formula is obtained: ； in, represents the feature vector after Dropout, represents the original feature vector, Represents the Dropout mask vector, whose elements conform to the Bernoulli distribution.

2. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: In step 100: The modality data includes biomarker data, electronic health records, lifestyle data, and imaging data; the biomarker data includes serum creatinine, urine protein, blood pressure, blood glucose levels, and blood lipid indicators; the electronic health records include medical history, diagnosis information, medication records, and complication information; the lifestyle data includes dietary habits, exercise frequency, smoking and drinking habits, and sleep patterns; the imaging data includes renal ultrasound images and magnetic resonance imaging; The pre-processing step comprises: For different modal data, missing data are filled by using missing value filling and multiple imputation methods; Identify and handle outliers through statistical method box plot analysis; Perform Z-score standardization on numerical features to ensure that all features are on the same scale. The formula is as follows: ； in, After standardization modal data, Represents the original modal data, and Respectively The mean and standard deviation of the modal data; The patient feature matrix is ,in is the number of patients, is the feature dimension.

3. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: In step 200: The construction of patient similarity graph includes graph structure definition and similarity measurement; The graph structure definition is: the nodes of the patient similarity graph represent each patient, and the feature vector of the node is the preprocessed patient feature, which is expressed as ,in, Indicates the The edges of the patient similarity graph represent the similarities between patients. The similarity measure includes: Biomarker similarity metric: Based on the similarity metric of patient biomarkers, the Gaussian similarity function is used to obtain the formula: ； in, represents the Gaussian similarity measure of the patient's biomarkers, 、 Represents the patient's biomarkers at the nodes and The eigenvectors on , represents the width of the Gaussian kernel of the patient's biomarker; Lifestyle similarity measurement: Based on the similarity measurement of lifestyle data, cosine similarity is used to obtain the formula: ； in, represents the similarity measure of patients’ lifestyles, 、 Represents the patient's lifestyle at the node and The eigenvectors on ; Medical history similarity measurement: Based on the similarity measurement of the patient's medical history information, Jaccard similarity is used to obtain the formula: ； in, The patient's medical history information is in the node and The eigenvectors on , 、 Represents the patient at the node and Medical history collection; Dynamic similarity measurement: Using the dynamic weight adjustment mechanism, the weights of each similarity measurement are dynamically adjusted by learning weight parameters, and the formula is obtained: ； in, Represents a node in the patient similarity graph and nodes The variable weights between Representation node No. eigenvalues, Representation node No. eigenvalues, is the number of dimensions of similarity measurement, For the The weight of each similarity dimension.

4. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 3, characterized in that: In step 300: The structured data fusion integrates biomarkers, electronic health records, etc., and calculates the similarity between patients using the similarity measurement method; the unstructured data fusion uses natural language processing to encode text data and incorporates the extracted high-dimensional text features into the similarity measurement; the image feature fusion extracts deep features of image data through pre-trained convolutional neural networks, and fuses the deep features with other modal data to enhance the depth and diversity of the similarity measurement; The domain knowledge integration includes defining the similarity between patients based on clinical pathways and progression trajectories, adjusting the similarity measurement weights through expert knowledge guidance, and dividing the similarity into different levels, calculating them separately and integrating them into the overall similarity score.

5. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: In step 400: The thinning process includes: By setting a similarity threshold , only keep the similarity higher than The side of , the formula is as follows: ； The K nearest neighbor retention strategy is used. For each node, only the node with the closest similarity to it is retained. neighbors, the formula is as follows: ； in, Representation node The node with the highest similarity to other nodes nodes; The steps of the normalization process are as follows: ； ； ； in, is the identity matrix, is the degree matrix, is the normalized adjacency matrix, represents the normalized adjacency matrix, Represents the normalized adjacency matrix Middle Rank Elements of a column.

6. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: Step 500 includes: Step 510: Divide the patient data into multiple time windows according to time. The graph structure corresponding to a moment, in each time window , recalculate the similarity measure based on the latest patient characteristic data and update the adjacency matrix ,Through dynamic weight adjustment and multi-source information fusion, it reflects the latest patient characteristics and similarity relationships; Step 520: Combined with the time series information, the graph structure data in the time window sequence is processed by the time series graph convolutional network to capture the time series evolution of the patient characteristics. The formula is as follows: ； in, For the The node feature matrix of the layer, For the The normalized adjacency matrix of the time window, For the The node feature matrix of the layer, Indicates the The weight matrix of the layer; Step 530: Introduce a memory mechanism to serialize the graph convolution outputs of different time windows to capture long-term dependencies and feature evolution trends.

7. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: Step 600 includes: Step 610: Use the self-attention mechanism to enhance node features and dynamically adjust the importance of different feature dimensions. The formula is as follows: ； in, is the self-attention weight, For nodes The eigenvector of represents the weight matrix, Representation node The new feature vector after being enhanced by the self-attention mechanism; Step 620: Add a graph regularization term to the loss function to make the features of similar nodes closer: ； in, is the set of edges in the graph, node The eigenvector of represents the graph regularization loss term; Step 630: Randomly delete or add edge and node noise to enhance the patient similarity graph data, the method includes edge perturbation, node perturbation and feature perturbation.

8. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: Step 700 includes: Step 710: In each layer of the temporal graph convolutional network (GCN), a multi-head attention mechanism is used to calculate different attention weights in parallel through multiple attention heads to capture different types of relationship patterns. The formula is as follows: ； in, is the number of attention heads, is the output weight matrix, represents the multi-head attention mechanism, is the query vector, K is the key vector, and V is the value vector; Step 720: Generate adversarial examples to improve the model's robustness to input perturbations. Specific methods include the fast gradient sign method and projected gradient descent: Fast Gradient Sign Method: Generate adversarial samples using gradient information: ； in, represents the original input sample, represents an adversarial sample, generated by adding a small perturbation to the original input sample. The loss function measures the difference between the model prediction and the true label. represents the disturbance amplitude, Represents the loss function For the original input sample The partial derivative of Projected Gradient Descent: Iteratively generates adversarial samples, the formula is as follows: ； in, is the set of constraints for adversarial samples, is the step length, express The adversarial samples generated by iterations are express Adversarial samples generated by iterations; Step 730: Perform deep learning integration and personalized risk assessment to ensure collaborative optimization of the model at the integration and personalization levels; Step 730 includes the following steps: Step 731: Randomly initialize all model parameters, including GCN parameters, base model parameters, and integrated weight coefficients and personalized attention parameters; Step 732: Optimize the integrated prediction loss and personalized loss to obtain the total loss function. for: ； in, is the loss function for ensemble prediction, is the loss function for personalized prediction, is the weight of personalized loss, is the regularization parameter.

9. The personalized hypertensive nephropathy risk assessment method based on graph convolutional network according to claim 1, characterized in that: The specific steps of step 900 are as follows: Step 910: Design a secondary model as the weight coefficient for learning the base model. The final prediction result is the weighted sum of the prediction results of each base model: ； in, The final prediction result is represented by Indicates the The weight coefficient of each basis model, Indicates the The prediction results of the basic model are represents the total number of base models in the ensemble method, ; Learn weight coefficients through backpropagation and optimization algorithms , to minimize the loss function of integrated prediction; Based on the individual characteristics of the patient, a customized risk prediction is made for each patient through an adaptive attention mechanism. The specific steps include feature extraction, attention weight calculation, and personalized weighting. The formula is as follows: ； ； ； in, Represents the intermediate representation vector obtained after linear transformation and nonlinear activation of the patient's individual characteristics, is the individual feature vector of the patient, and is the learnable weight matrix, and is the bias vector, is the attention weight; Step 920: Design a personalized loss function, combining the error between the personalized prediction result and the true label. The formula is as follows: ； in, is the sample size, For the The true labels of samples, Personalized prediction results.

10. A personalized hypertensive nephropathy risk assessment system based on a graph convolutional network, applied to the personalized hypertensive nephropathy risk assessment method based on a graph convolutional network according to any one of claims 1 to 9, characterized in that: include: Data acquisition and preprocessing module: used to collect modal data related to hypertensive nephropathy, preprocess the modal data, and then establish the patient feature matrix; Patient Similarity Graph Construction Module: This module is used to construct a patient similarity graph by measuring and encoding the similarities between patients and enhancing the patient feature representation using a graph convolutional network. Multi-source information fusion module: used for multi-source information fusion and domain knowledge integration; Sparseness and normalization processing module: used to perform sparse processing on the graph structure and then normalize the adjacency matrix; Time series processing module: used to process time series data and capture the temporal evolution trend of patient characteristics through temporal graph convolutional networks and long short-term memory networks; Feature transfer enhancement module: used to enhance the transfer of important features through the self-attention mechanism and graph regularization term; Relationship pattern capture module: It is used to capture different types of relationship patterns using a multi-head attention mechanism in each layer of the temporal graph convolutional network; Parameter optimization module: used to optimize GCN parameters, base model parameters, ensemble weights, and personalized attention parameters through the Adam optimization algorithm and regularization method; Personalized risk assessment module: It uses weighted stacking and adaptive attention mechanisms and learns personalized loss functions to optimize the final risk assessment results.

Images