Sudden infectious disease intelligent positioning system based on spatial-temporal distribution anomaly detection

Abstract

The invention relates to the technical field of public health information, in particular to an emergent infectious disease intelligent positioning system based on spatial-temporal distribution anomaly detection, which comprises a data acquisition module, a spatial-temporal feature modeling module, a spatial-temporal distribution anomaly detection module, an infectious disease intelligent positioning module, a propagation trend prediction module and a visualization module, according to the scheme, a multi-dimensional Gaussian ternary modeling scheme is provided, expectation, entropy and hyper-entropy parameters are introduced to finely describe the center position, change range and instability of spatial and temporal distribution characteristics, and accurate capture of weak anomalies is realized in combination with multi-criterion similarity calculation and a dynamic cluster merging mechanism; by introducing a graph structure modeling mode, constructing an explanatory graph structure and a layer-by-layer attribution trajectory and introducing trajectory volatility measurement in a hierarchical clustering process, the robust detection capability of abnormal points is improved, high-precision abnormal identification and infection source positioning of sudden infectious diseases are realized, and the intelligent level and timeliness of epidemic situation response are improved.

Description

Technical Field

[0001] The present invention relates to the field of public health information technology, and in particular to an intelligent positioning system for infectious diseases based on spatiotemporal distribution anomaly detection. Background Art

[0002] With the increasing mobility of the population and the advancement of urbanization, the frequency and speed of outbreaks of infectious diseases have increased significantly, posing a serious threat to public health and safety. Some existing methods attempt to use clustering or anomaly detection techniques for epidemic monitoring, but most of them have the following shortcomings:

[0003] (1) Existing clustering methods ignore the dynamic uncertainty of spatiotemporal characteristics and lack modeling of the uncertainty and evolution trend of spatiotemporal distribution characteristics. Clustering is only based on the similarity of static characteristics, which makes it difficult to effectively identify the weak but critical spatiotemporal anomaly signals hidden in the development of the epidemic;

[0004] (2) Existing clustering methods have the problem of insufficient modeling of similarity structure and susceptibility to noise. They lack in-depth modeling of similarity structure and topological relationship between nodes. The clustering process is easily disturbed by blurred boundaries, isolated points and structural noise, resulting in unstable anomaly detection results and low accuracy of infection source location and transmission prediction. Summary of the Invention

[0005] In response to the above situation, in order to overcome the shortcomings of the existing technology, the present invention provides an intelligent positioning system for infectious diseases based on spatiotemporal distribution anomaly detection. In response to the problem that the existing clustering method ignores the dynamic uncertainty of spatiotemporal characteristics, the present invention proposes a multi-dimensional Gaussian triplet modeling scheme, introduces expectation, entropy, and super entropy parameters to finely describe the central position, variation range and instability of the spatiotemporal distribution characteristics, and combines multi-criteria similarity calculation with a dynamic cluster merging mechanism to achieve accurate capture of weak anomalies; in response to the problem that the existing clustering method has insufficient modeling of similarity structure and is easily affected by noise, the present invention introduces a graph structure modeling method, constructs an explanatory graph structure and a layer-by-layer attribution trajectory, and introduces a trajectory volatility measure in the hierarchical clustering process to improve the robust detection capability of anomalies and avoid boundary fuzziness or interference from isolated points.

[0006] The technical solution adopted by the present invention is as follows: The system for intelligently locating infectious diseases based on spatiotemporal distribution anomaly detection provided by the present invention includes a data acquisition module, a spatiotemporal feature modeling module, a spatiotemporal distribution anomaly detection module, an infectious disease intelligent positioning module, a transmission trend prediction module, and a visualization module, specifically including the following:

[0007] The data collection module collects infectious disease related data, including case reporting data, personnel activity trajectory data, environmental meteorological data and regional population density data;

[0008] The spatiotemporal feature modeling module uses a spatiotemporal graph convolutional network to perform spatiotemporal encoding and modeling on infectious disease-related data to extract the spatiotemporal distribution characteristics of infectious diseases;

[0009] The spatiotemporal distribution anomaly detection module designs a hierarchical clustering method that integrates multi-criteria similarity and dynamic cluster merging mechanism to perform anomaly detection on spatiotemporal distribution features and obtain anomaly detection results;

[0010] The infectious disease intelligent positioning module uses a graph neural network model to construct an infection source traceability map based on the anomaly detection results and locate the location of potential infection sources;

[0011] The transmission trend prediction module constructs an infectious disease transmission location model based on the graph diffusion mechanism based on possible infection sources to obtain future risk transmission areas;

[0012] The visualization module graphically displays abnormal detection results, potential infection source locations, and future risk transmission areas.

[0013] Furthermore, the spatiotemporal distribution anomaly detection module designs a hierarchical clustering method that integrates multi-criteria similarity and a dynamic cluster merging mechanism to perform anomaly detection on spatiotemporal distribution features, specifically including the following steps:

[0014] Step S1: Input acquisition, receiving spatiotemporal distribution features, expressed as follows:

[0015] ;

[0016] in, represents a set of spatiotemporal distribution features, Indicates the spatiotemporal distribution feature vectors, Represents an index;

[0017] Step S2: Extend the modeling and design a Gaussian distribution mapping model to map each spatiotemporal distribution feature vector to the corresponding multidimensional Gaussian triplet parameter, as shown below:

[0018] ;

[0019] in, For the The expected value of the spatiotemporal distribution feature vector represents the center of the spatiotemporal distribution, and For the The entropy and super entropy of the spatiotemporal distribution feature vector represent the uncertainty range of the spatiotemporal distribution and the instability of the spatiotemporal distribution change. Indicates the spatiotemporal distribution feature vectors;

[0020] Step S3: Calculate the similarity matrix. Based on the multidimensional Gaussian triplet parameters, design a density similarity function to calculate the comprehensive similarity score between any two feature points. The density similarity function used is expressed as follows:

[0021] ;

[0022] ;

[0023] in, Indicates the and The comprehensive similarity of the spatiotemporal distribution feature vectors, is a smoothing factor, a constant to avoid division by zero, It is a regulation function based on super entropy;

[0024] Step S4: Anomaly clustering. Based on the correlation structure between spatiotemporal features and the similarity matrix between multi-dimensional Gaussian triples, a hierarchical clustering method that integrates multi-criteria similarity and dynamic cluster merging mechanism is designed to perform anomaly detection on the similarity matrix and obtain anomaly clustering results.

[0025] Step S5: Fusion feature construction, combining the abnormal clustering results with the spatiotemporal distribution feature set to construct a fusion feature vector;

[0026] Step S6: Output the result, input the fused feature vector into the classification model based on the gradient boosting tree, perform anomaly discrimination, and output the anomaly detection result corresponding to each spatiotemporal distribution feature vector.

[0027] Furthermore, in step S4, a hierarchical clustering method is designed that integrates multi-criteria similarity and a dynamic cluster merging mechanism to perform anomaly detection on the similarity matrix, which specifically includes the following steps:

[0028] Step S41: Multi-strategy similarity matrix expansion, based on the similarity matrix, a multi-strategy similarity fusion mechanism is introduced. The fused similarity matrix is ​​expressed as:

[0029] ;

[0030] in, is the similarity matrix after fusion, 、 and Represents an adjustable parameter, satisfying , satisfy , represents the scale adjustment parameter of the Gaussian kernel function;

[0031] Step S42: Initial clustering is constructed, based on the fused similarity matrix, defining each spatiotemporal distribution feature vector to correspond to an initial clustering cluster;

[0032] Step S43: Constructing an explanatory similarity view and constructing an explanatory graph structure , where the node set is , the edge set is ;

[0033] Step S44: Multi-level similarity fusion, introduce clustering hierarchy definition, record the first Level indicates execution Based on the cluster division state formed after the round of similarity merging, a hierarchical clustering strategy based on multi-round similarity threshold adaptive adjustment is designed to merge the explanatory graph structure layer by layer to obtain the final clustering structure;

[0034] Step S45: abnormal boundary detection: based on the final cluster structure, an abnormality detection method based on the analysis of inter-layer cluster structure changes is designed to identify abnormal points;

[0035] Step S46: outlier output, the outliers identified in step S45 are sorted into an outlier set as potential spatiotemporal distribution outliers.

[0036] Furthermore, step S44 specifically includes the following steps:

[0037] Step S441: Construction of inter-cluster similarity metric. Each node in the explanatory graph structure represents a spatiotemporal distribution feature vector. According to step S42, each node also constitutes a cluster. The similarity between any two clusters is defined as:

[0038] ;

[0039] in, express Lower level cluster and clusters The average similarity of

[0040] Step S442: Merge the graph structure levels and set the initial similarity threshold. satisfy When and Merge into new cluster , and update the explanatory graph structure to the next level , where the node set is the merged cluster;

[0041] Step S443: Adaptive threshold adjustment, introducing cluster change index to update threshold adjustment factor, the threshold adjustment factor update formula is as follows:

[0042] ;

[0043] in, express Similarity threshold at the level, express Similarity threshold at the level, is the threshold adjustment factor, is the cluster change index;

[0044] Step S444: Terminate the judgment, set the maximum number of iterations, and output the final clustering structure when the maximum number of iterations is reached. , passed to step S45 to perform abnormal boundary detection.

[0045] Furthermore, step S45 specifically includes the following steps:

[0046] Step S451: Construction of attribution trajectory. In the expansion step of step S44, the label changes of the cluster to which each spatiotemporal distribution feature vector belongs in each level are recorded, and the hierarchical attribution trajectory is defined as:

[0047] ;

[0048] in, Indicates the The spatiotemporal distribution feature vectors are from the first layer to the The clustering label sequence of the layer, Indicates the The spatiotemporal distribution feature vector is The cluster number to which the layer belongs, Indicates the maximum number of levels representing the clustering process;

[0049] Step S452: constructing a trajectory volatility metric, defining a trajectory volatility index, which is expressed as follows:

[0050] ;

[0051] in, Indicates the The trajectory volatility index of the spatiotemporal distribution characteristic vector, is an indicator function, which is 1 if the cluster labels of the current layer and the next layer are inconsistent, otherwise it is 0;

[0052] Step S453: Outlier screening: set a fluctuation threshold to screen out the spatiotemporal distribution feature vectors whose volatility is significantly higher than the overall mean, which is recorded as:

[0053] ;

[0054] in, is an outlier, indicating the subset of spatiotemporal distribution features identified as an abnormal boundary point. The anomaly discrimination threshold representing trajectory volatility.

[0055] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0056] (1) To address the problem that existing clustering methods ignore the dynamic uncertainty of spatiotemporal features, this paper proposes a multi-dimensional Gaussian triplet modeling scheme, which introduces expectation, entropy, and super entropy parameters to accurately describe the central position, variation range, and instability of spatiotemporal distribution features. It combines multi-criteria similarity calculation with a dynamic cluster merging mechanism to accurately capture weak anomalies.

[0057] (2) To address the problem that existing clustering methods are insufficient in modeling similarity structures and are easily affected by noise, this paper introduces a graph structure modeling method to construct an explanatory graph structure and layer-by-layer attribution trajectories. In the hierarchical clustering process, a trajectory volatility measure is introduced to improve the robust detection capability of outliers and avoid boundary ambiguity or interference from isolated points. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a schematic diagram of the intelligent positioning system for infectious diseases based on spatiotemporal distribution anomaly detection proposed by the present invention;

[0059] Figure 2 This is a flow chart of the method for performing anomaly detection using the spatiotemporal distribution anomaly detection module proposed in the present invention.

[0060] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0061] 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 only part of the embodiments of the present invention, rather than all the embodiments; 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.

[0062] Example 1, see Figure 1 The present invention provides an intelligent positioning system for infectious diseases based on spatiotemporal distribution anomaly detection, comprising a data acquisition module, a spatiotemporal feature modeling module, a spatiotemporal distribution anomaly detection module, an infectious disease intelligent positioning module, a transmission trend prediction module, and a visualization module. Specifically, the system includes the following:

[0063] The data collection module collects infectious disease related data, including case reporting data, personnel activity trajectory data, environmental meteorological data and regional population density data;

[0064] The spatiotemporal feature modeling module uses a spatiotemporal graph convolutional network to perform spatiotemporal encoding and modeling on infectious disease-related data to extract the spatiotemporal distribution characteristics of infectious diseases;

[0065] The spatiotemporal distribution anomaly detection module designs a hierarchical clustering method that integrates multi-criteria similarity and dynamic cluster merging mechanism to perform anomaly detection on spatiotemporal distribution features and obtain anomaly detection results;

[0066] The infectious disease intelligent positioning module uses a graph neural network model to construct an infection source traceability map based on the anomaly detection results and locate the location of potential infection sources;

[0067] The transmission trend prediction module constructs an infectious disease transmission location model based on the graph diffusion mechanism based on possible infection sources to obtain future risk transmission areas;

[0068] The visualization module graphically displays abnormal detection results, potential infection source locations, and future risk transmission areas.

[0069] Example 2, see Figure 2 This embodiment is based on the above embodiment. The spatiotemporal distribution anomaly detection module designs a hierarchical clustering method that integrates multi-criteria similarity and a dynamic cluster merging mechanism to perform anomaly detection on spatiotemporal distribution features. Specifically, the following steps are included:

[0070] Step S1: Input acquisition, receiving spatiotemporal distribution features, expressed as follows:

[0071] ;

[0072] in, represents a set of spatiotemporal distribution features, Indicates the spatiotemporal distribution feature vectors, Represents an index;

[0073] Step S2: Extend the modeling and design a Gaussian distribution mapping model to map each spatiotemporal distribution feature vector to the corresponding multidimensional Gaussian triplet parameter, as shown below:

[0074] ;

[0075] in, For the The expected value of the spatiotemporal distribution feature vector represents the center of the spatiotemporal distribution, and For the The entropy and super entropy of the spatiotemporal distribution feature vector represent the uncertainty range of the spatiotemporal distribution and the instability of the spatiotemporal distribution change. Indicates the spatiotemporal distribution feature vectors;

[0076] Step S3: Calculate the similarity matrix. Based on the multidimensional Gaussian triplet parameters, design a density similarity function to calculate the comprehensive similarity score between any two feature points. The density similarity function used is expressed as follows:

[0077] ;

[0078] ;

[0079] in, Indicates the and The comprehensive similarity of the spatiotemporal distribution feature vectors, is a smoothing factor, a constant to avoid division by zero, It is a regulation function based on super entropy;

[0080] Step S4: Anomaly clustering. Based on the correlation structure between spatiotemporal features and the similarity matrix between multi-dimensional Gaussian triples, a hierarchical clustering method that integrates multi-criteria similarity and dynamic cluster merging mechanism is designed to perform anomaly detection on the similarity matrix and obtain anomaly clustering results.

[0081] Step S5: Fusion feature construction, combining the abnormal clustering results with the spatiotemporal distribution feature set to construct a fusion feature vector;

[0082] Step S6: Output the result, input the fused feature vector into the classification model based on the gradient boosting tree, perform anomaly discrimination, and output the anomaly detection result corresponding to each spatiotemporal distribution feature vector.

[0083] In this example, the spatiotemporal distribution data of cases in five regions were collected. After ST-GCN modeling, five spatiotemporal distribution feature vectors were extracted, which are defined as: , including the following vectors:

[0084] (0.85, 0.40, 0.70);

[0085] (0.80, 0.42, 0.68);

[0086] (0.10, 0.85, 0.15);

[0087] (0.12, 0.82, 0.20);

[0088] (0.60, 0.38, 0.65);

[0089] Construct a Gaussian mapping model through the sample set and obtain the triplet parameters of each vector:

[0090] Triple parameter table

[0091]

[0092] ;

[0093] Pick ;

[0094] The similarity merging threshold is set to decrease from 0.95 to 0.70, and point pairs with similarity higher than the current threshold are merged in each round. The system generates the following hierarchical attribution tracks:

[0095] Hierarchical attribution trajectory table

[0096]

[0097] Indicates the In the hierarchy, Cluster identification of points;

[0098] Construct the trajectory volatility of each point and set the volatility threshold;

[0099] ;

[0100] ;

[0101] ;

[0102] ;

[0103] The system sets the volatility threshold to 1 and obtains a set of abnormal boundary points;

[0104] The original It is concatenated with its trajectory fluctuation index ∆ᵢ to form a fusion feature vector.

[0105] Example 3: Based on the above example, in step S4, a hierarchical clustering method is designed that integrates multi-criteria similarity and a dynamic cluster merging mechanism to perform anomaly detection on the similarity matrix. Specifically, the method includes the following steps:

[0106] Step S41: Multi-strategy similarity matrix expansion, based on the similarity matrix, a multi-strategy similarity fusion mechanism is introduced. The fused similarity matrix is ​​expressed as:

[0107] ;

[0108] in, is the similarity matrix after fusion, 、 and Represents an adjustable parameter, satisfying , satisfy , represents the scale adjustment parameter of the Gaussian kernel function;

[0109] Step S42: Initial clustering is constructed, based on the fused similarity matrix, defining each spatiotemporal distribution feature vector to correspond to an initial clustering cluster;

[0110] Step S43: Constructing an explanatory similarity view and constructing an explanatory graph structure , where the node set is , the edge set is ;

[0111] Step S44: Multi-level similarity fusion, introduce clustering hierarchy definition, record the first Level indicates execution Based on the cluster division state formed after the round of similarity merging, a hierarchical clustering strategy based on multi-round similarity threshold adaptive adjustment is designed to merge the explanatory graph structure layer by layer to obtain the final clustering structure;

[0112] Step S45: abnormal boundary detection: based on the final cluster structure, an abnormality detection method based on the analysis of inter-layer cluster structure changes is designed to identify abnormal points;

[0113] Step S46: outlier output, the outliers identified in step S45 are sorted into an outlier set as potential spatiotemporal distribution outliers.

[0114] In this embodiment, the core code used is as follows:

[0115] import numpy as np

[0116] import networkx as nx

[0117] from sklearn.metrics.pairwise import cosine_similarity

[0118] from scipy.spatial.distance import cdist

[0119] # --------------------------

[0120] # 1. Initialize data (5 samples)

[0121] # --------------------------

[0122] F = np.array([

[0123] [0.80, 0.40, 0.70],

[0124] [0.82, 0.39, 0.72],

[0125] [0.12, 0.85, 0.10],

[0126] [0.10, 0.82, 0.13],

[0127] [0.60, 0.35, 0.68] ])

[0129] # Simulate Gaussian triplet parameters (Ex, En, He)

[0130] Ex = F.copy()

[0131] En = np.random.rand(5) * 0.1 + 0.05 # Random entropy value En_i ∈ [0.05, 0.15]

[0132] He = np.random.rand(5) * 0.2 + 0.1 # Random super entropy value He_i ∈ [0.1, 0.3]

[0133] epsilon = 1e-5

[0134] λ = 0.5

[0135] α, β, γ = 0.4, 0.3, 0.3

[0136] # --------------------------

[0137] # 2. Multi-criteria similarity fusion

[0138] # --------------------------

[0139] n = F.shape[0]

[0140] Sim = np.zeros((n, n))

[0141] Cos = cosine_similarity(F)

[0142] K = np.exp(-λ * cdist(F, F, metric='sqeuclidean'))

[0143] for i in range(n):

[0144] for j in range(n):

[0145] diff_sq = np.sum((Ex[i] - Ex[j]) ** 2)

[0146] density_sim = np.exp(-diff_sq / (En[i] + En[j] + epsilon))

[0147] phi = 1 / (1 + abs(He[i] - He[j]))

[0148] Sim[i, j] = density_sim * phi

[0149] S_fused = α * Sim + β * Cos + γ * K

[0150] # --------------------------

[0151] # 3. Construct the interpretable graph structure

[0152] # --------------------------

[0153] [[ID=*37]]θ = 0.98 # Similarity threshold

[0154] G = nx.Graph()

[0155] for i in range(n):

[0156] G.add_node(i)

[0157] for j in range(i + 1, n):

[0158] if S_fused[i, j] > θ:

[0159] G.add_edge(i, j, weight=S_fused[i, j]) Note: In the translation of line 37, since the original text uses Chinese, it is translated into English here. If this is not allowed according to the strict requirements, please let me know and I will adjust accordingly.

[0160] # --------------------------

[0161] # 4. Hierarchical Clustering Simulation

[0162] # --------------------------

[0163] def hierarchical_clustering(graph, max_level=3):

[0164] clusters_per_level = []

[0165] current_graph = graph.copy()

[0166] for l in range(max_level):

[0167] communities = list(nx.connected_components(current_graph))

[0168] label_map = {node: idx for idx, com in enumerate(communities)for node in com}

[0169] clusters_per_level.append(label_map)

[0170] # Dynamically adjust the threshold to reduce density and merge weak edges

[0171] new_graph = nx.Graph()

[0172] new_graph.add_nodes_from(current_graph.nodes)

[0173] for u, v, d in current_graph.edges(data=True):

[0174] if d["weight"] > θ - (l + 1) * 0.03:

[0175] new_graph.add_edge(u, v, weight=d["weight"])

[0176] current_graph = new_graph

[0177] return clusters_per_level

[0178] cluster_labels = hierarchical_clustering(G)

[0179] # --------------------------

[0180] # 5. Trajectory volatility detection (anomaly recognition)

[0181] # --------------------------

[0182] def calc_volatility(cluster_labels):

[0183] volatility = np.zeros(n)

[0184] for i in range(n):

[0185] labels = [cl[i] for cl in cluster_labels]

[0186] volatility[i] = sum(labels[l] != labels[l + 1] for l in range(len(labels) - 1))

[0187] return volatility

[0188] volatility = calc_volatility(cluster_labels)

[0189] θ_delta = np.mean(volatility) + np.std(volatility)

[0190] anomaly_indices = np.where(volatility >= θ_delta)[0]

[0191] print("Anomaly indices:", anomaly_indices.tolist())。

[0192] Embodiment 4: This embodiment is based on the above embodiment, and step S44 specifically includes the following steps:

[0193] Step S441: Construction of inter-cluster similarity metric. Each node in the explanatory graph structure represents a spatiotemporal distribution feature vector. According to step S42, each node also constitutes a cluster. The similarity between any two clusters is defined as:

[0194] ;

[0195] in, express Lower level cluster and clusters The average similarity of

[0196] Step S442: Merge the graph structure levels and set the initial similarity threshold. satisfy When and Merge into new cluster , and update the explanatory graph structure to the next level , where the node set is the merged cluster;

[0197] Step S443: Adaptive threshold adjustment, introducing cluster change index to update threshold adjustment factor, the threshold adjustment factor update formula is as follows:

[0198] ;

[0199] in, express Similarity threshold at the level, express Similarity threshold at the level, is the threshold adjustment factor, is the cluster change index;

[0200] Step S444: Terminate the judgment, set the maximum number of iterations, and output the final clustering structure when the maximum number of iterations is reached. , passed to step S45 to perform abnormal boundary detection.

[0201] Embodiment 5, based on the above embodiment, step S45 specifically includes the following steps:

[0202] Step S451: Construction of attribution trajectory. In the expansion step of step S44, the label changes of the cluster to which each spatiotemporal distribution feature vector belongs in each level are recorded, and the hierarchical attribution trajectory is defined as:

[0203] ;

[0204] in, Indicates the The spatiotemporal distribution feature vectors are from the first layer to the The clustering label sequence of the layer, Indicates the The spatiotemporal distribution feature vector is The cluster number to which the layer belongs, Indicates the maximum number of levels representing the clustering process;

[0205] Step S452: constructing a trajectory volatility metric, defining a trajectory volatility index, which is expressed as follows:

[0206] ;

[0207] in, Indicates the The trajectory volatility index of the spatiotemporal distribution characteristic vector, is an indicator function, which is 1 if the cluster labels of the current layer and the next layer are inconsistent, otherwise it is 0;

[0208] Step S453: Outlier screening: set a fluctuation threshold to screen out the spatiotemporal distribution feature vectors whose volatility is significantly higher than the overall mean, which is recorded as:

[0209] ;

[0210] in, is an outlier, indicating the subset of spatiotemporal distribution features identified as an abnormal boundary point. The anomaly discrimination threshold representing trajectory volatility.

[0211] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0212] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

[0213] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. An intelligent positioning system for infectious diseases based on spatiotemporal distribution anomaly detection, characterized by: It includes data acquisition module, spatiotemporal feature modeling module, spatiotemporal distribution anomaly detection module, infectious disease intelligent positioning module, transmission trend prediction module and visualization module, specifically including the following contents: The data acquisition module collects data related to infectious diseases; The spatiotemporal feature modeling module uses a spatiotemporal graph convolutional network to perform spatiotemporal encoding and modeling on infectious disease-related data to extract the spatiotemporal distribution characteristics of infectious diseases; The spatiotemporal distribution anomaly detection module designs a hierarchical clustering method that integrates multi-criteria similarity and dynamic cluster merging mechanism to perform anomaly detection on spatiotemporal distribution features and obtain anomaly detection results; The infectious disease intelligent positioning module uses a graph neural network model to construct an infection source traceability map based on the anomaly detection results and locate the location of potential infection sources; The transmission trend prediction module constructs an infectious disease transmission location model based on the graph diffusion mechanism based on possible infection sources to obtain future risk transmission areas; The visualization module graphically displays abnormal detection results, potential infection source locations, and future risk transmission areas.

2. The intelligent positioning system for infectious disease emergencies based on spatiotemporal distribution anomaly detection according to claim 1 is characterized by: The spatiotemporal distribution anomaly detection module designs a hierarchical clustering method that integrates multi-criteria similarity and a dynamic cluster merging mechanism to detect anomalies in spatiotemporal distribution features. Specifically, it includes the following steps: Step S1: Input acquisition, receiving spatiotemporal distribution features, expressed as follows: ； in, represents a set of spatiotemporal distribution features, Indicates the spatiotemporal distribution feature vectors, Represents an index; Step S2: Extend the modeling and design a Gaussian distribution mapping model to map each spatiotemporal distribution feature vector to the corresponding multidimensional Gaussian triplet parameter, as shown below: ； in, For the The expected value of the spatiotemporal distribution feature vector represents the center of the spatiotemporal distribution, and For the The entropy and super entropy of the spatiotemporal distribution feature vector represent the uncertainty range of the spatiotemporal distribution and the instability of the spatiotemporal distribution change. Indicates the spatiotemporal distribution feature vectors; Step S3: Calculate the similarity matrix. Based on the multidimensional Gaussian triplet parameters, design a density similarity function to calculate the comprehensive similarity score between any two feature points. The density similarity function used is expressed as follows: ； ； in, Indicates the and The comprehensive similarity of the spatiotemporal distribution feature vectors, is a smoothing factor, a constant to avoid division by zero, It is a regulation function based on super entropy; Step S4: Anomaly clustering. Based on the correlation structure between spatiotemporal features and the similarity matrix between multi-dimensional Gaussian triples, a hierarchical clustering method that integrates multi-criteria similarity and dynamic cluster merging mechanism is designed to perform anomaly detection on the similarity matrix and obtain anomaly clustering results. Step S5: Fusion feature construction, combining the abnormal clustering results with the spatiotemporal distribution feature set to construct a fusion feature vector; Step S6: Output the result, input the fused feature vector into the classification model based on the gradient boosting tree, perform anomaly discrimination, and output the anomaly detection result corresponding to each spatiotemporal distribution feature vector.

3. The intelligent positioning system for infectious disease emergencies based on spatiotemporal distribution anomaly detection according to claim 2 is characterized by: Step S4: design a hierarchical clustering method that integrates multi-criteria similarity and a dynamic cluster merging mechanism to perform anomaly detection on the similarity matrix, specifically including the following steps: Step S41: Multi-strategy similarity matrix expansion, based on the similarity matrix, a multi-strategy similarity fusion mechanism is introduced. The fused similarity matrix is ​​expressed as: ； in, is the similarity matrix after fusion, 、 and Represents an adjustable parameter, satisfying , satisfy , represents the scale adjustment parameter of the Gaussian kernel function; Step S42: Initial clustering is constructed, based on the fused similarity matrix, defining each spatiotemporal distribution feature vector to correspond to an initial clustering cluster; Step S43: Constructing an explanatory similarity view and constructing an explanatory graph structure , where the node set is , the edge set is ; Step S44: Multi-level similarity fusion, introduce clustering hierarchy definition, record the first Level means execution of Based on the cluster division state formed after the round of similarity merging, a hierarchical clustering strategy based on multi-round similarity threshold adaptive adjustment is designed to merge the explanatory graph structure layer by layer to obtain the final clustering structure; Step S45: abnormal boundary detection: based on the final cluster structure, an abnormality detection method based on the analysis of inter-layer cluster structure changes is designed to identify abnormal points; Step S46: outlier output, the outliers identified in step S45 are sorted into an outlier set as potential spatiotemporal distribution outliers.

4. The intelligent positioning system for infectious disease emergencies based on spatiotemporal distribution anomaly detection according to claim 3 is characterized by: Step S44 specifically includes the following steps: Step S441: Construction of inter-cluster similarity metric. Each node in the explanatory graph structure represents a spatiotemporal distribution feature vector. According to step S42, each node also constitutes a cluster. The similarity between any two clusters is defined as: ； in, express Lower level cluster and clusters The average similarity of Step S442: Merge the graph structure levels and set the initial similarity threshold. satisfy When and Merge into new cluster , and update the explanatory graph structure to the next level , where the node set is the merged cluster; Step S443: Adaptive threshold adjustment, introducing cluster change index to update threshold adjustment factor, the threshold adjustment factor update formula is as follows: ； in, express Similarity threshold at the level, express Similarity threshold at the level, is the threshold adjustment factor, is the cluster change index; Step S444: Terminate the judgment, set the maximum number of iterations, and output the final clustering structure when the maximum number of iterations is reached. , passed to step S45 to perform abnormal boundary detection.

Images