A Monitoring and Early Warning Method for the Spatiotemporal Transmission of Infectious Diseases Oriented to Regional Medical Big Data

Abstract

This application relates to the technical field of digital medicine, and in particular to an infectious disease spatio-temporal transmission monitoring and early warning method for regional medical big data, which includes: obtaining multi-source data based on data sources and integrating the data to obtain continuous spatio-temporal data, and constructing a behavior chain database according to the continuous spatio-temporal data; dividing the monitoring areas and combining the behavior chain database to count the population density corresponding to each monitoring area; analyzing the behavior chain database to extract key knowledge data associated with the monitoring areas and converting it into knowledge triples, and storing a number of knowledge triples in a graph database; traversing the graph database to obtain population behavior knowledge to construct a behavior graph database, predicting the change characteristics of the population density based on a neural network learning model, and analyzing the transmission risk based on the change characteristics in combination with the behavior database to conduct corresponding early warnings. This application has the effect of improving the accuracy and timeliness of infectious disease transmission monitoring and early warning.

Description

Technical Field

[0001] The present application relates to the technical field of digital medicine, and in particular to a method for monitoring, warning of the spatio-temporal spread of infectious diseases for regional medical big data. Background Art

[0002] Intelligent medicine uses advanced information technology and data analysis methods to promote the modernization of the medical and health field. In particular, it has been widely used in the monitoring and warning of infectious diseases in recent years.

[0003] In traditional infectious disease monitoring and prevention systems, it usually relies on rule-based expert systems or simple statistical methods to analyze the spread trend of infectious diseases. These systems generally build simple prediction models based on historical data, or judge the current spread situation through regular population sampling and monitoring reports.

[0004] Traditional methods have great limitations in data collection and processing. Firstly, the data sources are single, and they can only rely on the data provided by hospitals or public health departments, lacking a comprehensive analysis of multi-dimensional information such as patient flow, gathering area places, and population behavior. In addition, traditional models often lack sufficient flexibility and accuracy in dealing with complex spatio-temporal dynamic changes, and are unable to respond to the rapidly changing infectious environment in real time. At the same time, many traditional systems rely too much on empirical judgment and are difficult to achieve data-driven accurate prediction, resulting in the timeliness and accuracy of their warnings being unable to meet the requirements of emergency response. Summary of the Invention

[0005] In order to improve the accuracy and timeliness of infectious disease spread monitoring and warning, the present application provides a method for monitoring, warning of the spatio-temporal spread of infectious diseases for regional medical big data.

[0006] The present application provides a method for monitoring, warning of the spatio-temporal spread of infectious diseases for regional medical big data, adopting the following technical solutions:

[0007] A method for monitoring, warning of the spatio-temporal spread of infectious diseases for regional medical big data includes the following steps:

[0008] Obtain multi-source data based on data sources and perform data integration to obtain continuous spatio-temporal data, and construct a behavior chain database according to the continuous spatio-temporal data, wherein the behavior chain database consists of several behavior chains independently corresponding to patients;

[0009] Divide the monitoring areas and combine the behavior chain database to count the population density corresponding to each monitoring area;

[0010] Analyze the behavior chain database to extract key knowledge data associated with the monitoring areas and convert it into knowledge triples, and store several of the knowledge triples in a graph database;

[0011] Traverse the graph database to obtain crowd behavior knowledge to construct a behavior graph database, predict the change characteristics of the crowd density based on a neural network learning model, and analyze the transmission risk based on the change characteristics in combination with the behavior database for corresponding early warning.

[0012] In some embodiments, multi-source data is obtained based on data sources and data integration is performed to obtain continuous spatio-temporal data, and a behavior chain database is constructed according to the continuous spatio-temporal data, including the following steps:

[0013] Obtain the hospital data table and extract the patient information and its associated medical visit information and sort them by time to generate continuous medical visit records;

[0014] Obtain the travel data associated with the patient information, extract the residence information corresponding to the stay time exceeding a preset duration based on the travel data to construct a residence chain, and extract the travel destination information corresponding to the stay time exceeding a preset duration based on the travel data to construct a travel chain;

[0015] When the total stay time corresponding to the residence chain and the travel chain is greater than a preset time threshold, merge the residence chain and the travel chain to form the behavior chain associated with the patient information, and the behavior chain includes behavior location information and stay time;

[0016] Integrate and store the behavior chains corresponding to several pieces of the patient information to obtain the behavior chain database.

[0017] In some embodiments, a monitoring area is divided and the crowd density corresponding to each monitoring area is counted in combination with the behavior chain database, including the following steps:

[0018] Divide each area based on the division type option to obtain several monitoring areas with corresponding results, where the division type option includes area option, function option, and representative feature option;

[0019] Map the behavior location information stored in the behavior chain database to the monitoring area to calculate the number of overlapping patients, and generate the crowd density corresponding to each monitoring area in each time period in combination with the corresponding stay time, where the calculation formula is:

[0020] ,

[0021] where, represents the crowd density of the monitoring area s in the time period t, represents the number of overlapping patients in the monitoring area s in the time period t, represents the area of the monitoring area s.

[0022] In some of these embodiments, after counting the population density corresponding to each of the monitoring areas, the following steps are further included:

[0023] Determine whether the population density is greater than a preset density. If so, mark the monitoring area as a high-risk area.

[0024] In some of these embodiments, analyzing the behavior chain database to extract key knowledge data associated with the monitoring area and converting it into knowledge triples includes the following steps:

[0025] Generate a number of triple structures corresponding to each of the monitoring areas based on the behavior chain database. The triple structure includes a head entity, a relationship, and a tail entity;

[0026] Extract the patient information in the behavior chain database and correspond it to the head entity, extract the behavior location information within the monitoring area and correspond it to the tail entity, and extract the behavior action and correspond it to the relationship;

[0027] Determine whether the head entity and the tail entity have a namespace;

[0028] If it exists, follow the existing naming rules to assign identifiers. If not, create a new namespace and assign the identifiers based on the naming rules. The identifiers are used to identify and describe the associated objects.

[0029] In some of these embodiments, storing a number of the knowledge triples in a graph database includes the following steps:

[0030] Store the head entity and the tail entity in the nodes of the graph database, and store the relationship in the edges of the graph database;

[0031] Define the expression relationship between the associated nodes and edges. The expression relationship characterizes the relationship strength between the nodes and edges. Specifically,

[0032] ,

[0033] Among them, Characterizes the relationship strength from node i to node j, Characterizes the attribute related to relationship m in node i, Characterizes the attribute weight from relationship m to node j. n is the number of relationships involved.

[0034] In some of these embodiments, storing a number of the knowledge triples in a graph database further includes the following steps:

[0035] Obtain the quantity of the knowledge triples;

[0036] If the quantity is less than a preset value, create a triple table structure, store the head entity corresponding to the subject, the relationship corresponding to the predicate, and the tail entity corresponding to the object, and establish a combined index for the subject and object through a B+Tree and store them;

[0037] If the quantity is greater than the preset value, record the column increment or row increment in the triple table, and use the smallest bit-width integer type to store the row index and column index to achieve compression, and allocate the columns or rows where the relationship strength between the nodes and the edges in the knowledge triples exceeds the preset value to the same shard for storage.

[0038] In some embodiments, traverse the graph database to obtain crowd behavior knowledge to construct a behavior graph database, including the following steps:

[0039] Traverse the nodes in the graph database layer by layer through a queue structure to obtain the shortest behavior paths and the shortest spatio-temporal paths between the nodes to complete breadth-first search;

[0040] Traverse the non-direct paths between the nodes through recursion or a stack structure to obtain cross-regional behavior associations and explore continuous behavior sequences based on the residence time to obtain long-term behavior associations to complete depth-first exploration;

[0041] Integrate the search results of the breadth-first search and the depth-first exploration into crowd behavior knowledge to construct the behavior graph database.

[0042] In some embodiments, predict the change characteristics of the crowd density based on a neural network learning model, and analyze the spread risk based on the change characteristics in combination with the behavior database for corresponding early warning, including the following steps:

[0043] Screen several behavior triples in the behavior graph database as training samples to train a first neural network model, input the crowd density corresponding to the monitoring area at the current time into the trained first neural network model, and obtain the predicted potential crowd density based on the output result. Specifically,

[0044] ,

[0045] Among them, represents the potential crowd density for the predicted time period ; represents the crowd density at the current time; is an external influencing factor associated with the monitoring area; represents the model parameters of the first neural network model;

[0046] Obtain the density spatial distribution corresponding to each of the monitoring regions in the historical data and convert it into an image sample, and use a number of the image samples to train a second neural network model. After obtaining the input population density distribution map, the second neural network model outputs a classification prediction result based on image analysis combined with the potential population density. The classification prediction result includes potential high risk and potential low risk;

[0047] When the output potential population density exceeds a preset value or the output is potential high risk, an alarm is triggered.

[0048] In some of the embodiments, when predicting the change characteristics of the population density based on the neural network learning model, the following steps are further included:

[0049] Select high-risk patients based on the population density, obtain the behavior chain corresponding to the high-risk patients and match it with the behavior graph database, and generate iterative data based on the matching result to iteratively update the model parameters of the first neural network model and the second neural network model.

[0050] The technical solutions provided in the embodiments of the present application have the following technical effects:

[0051] (1) Realize multi-source data fusion and synergy effect, can effectively depict the spatio-temporal activity trajectory of patients, and combine the personnel gathering situation in the regional venue, providing a more accurate basis for the assessment and early warning of the spread of infectious diseases;

[0052] (2) Accurate spatio-temporal dynamic monitoring and analysis, by constructing a behavior chain database of patients and combining the division of regional venues, comprehensively master the spatio-temporal dynamics of patients and the changes in regional population density;

[0053] (3) Efficient population density prediction and early warning, based on the neural network of time series, can predict the changes in regional population density in real time and dynamically, and judge whether the early warning threshold for the outbreak of infectious diseases is reached, identify the population gathering trend in different regions and time periods, and accurately evaluate the potential risks;

[0054] (4) Improve the scientificity and accuracy of prevention and control decisions: Through the reasoning function of the knowledge graph and the prediction ability of the neural network learning model, it can accurately simulate the spatio-temporal dynamic migration of the population and provide a scientific basis for the transmission path of infectious diseases. Description of the Drawings

[0055] Figure 1 It is a schematic diagram of the steps of a method for monitoring and early warning of the spatio-temporal transmission of infectious diseases for regional medical big data provided by an embodiment of the present application.

[0056] Figure 2It is a schematic flowchart of S100 in an embodiment of the present application.

[0057] Figure 3 It is a schematic flowchart of S200 in an embodiment of the present application.

[0058] Figure 4 It is a schematic flowchart of S300 in an embodiment of the present application.

[0059] Figure 5 It is a schematic flowchart of S400 in an embodiment of the present application. Detailed implementation manners

[0060] To more clearly understand the purpose, technical solution and advantages of the present application, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. However, those of ordinary skill in the art should understand that the present application can be implemented without these details. In some cases, in order to avoid unnecessary descriptions from obscuring various aspects of the present application, well-known methods, processes, systems, components and / or circuits that have been described at a higher level will not be described in detail. For those of ordinary skill in the art, it is obvious that various changes can be made to the disclosed embodiments of the present application, and without departing from the principles and scope of the present application, the general principles defined in the present application can be applied to other embodiments and application scenarios. Therefore, the present application is not limited to the illustrated embodiments, but conforms to the broadest scope consistent with the scope claimed in the present application.

[0061] It should be noted here that the description of these implementation manners is used to help understand the present invention, but does not constitute a limitation to the present invention. In addition, the technical features involved in the various implementation manners of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0062] In the description of the present application, the meaning of several is one or more, the meaning of multiple is more than two, greater than, less than, exceeding, etc. are understood as not including the present number, above, below, within, etc. are understood as including the present number. If there is a description of first and second, it is only used to distinguish technical features and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0063] In the description of this application, the description with reference to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a combined manner.

[0064] As Figure 1 shown, an embodiment of this application discloses an infectious disease spatio-temporal transmission monitoring and early warning method for regional medical big data, including the following steps:

[0065] S100, obtaining multi-source data based on data sources and performing data integration to obtain continuous spatio-temporal data, and constructing a behavior chain database according to the continuous spatio-temporal data, wherein the behavior chain database is composed of behavior chains independently corresponding to several patients.

[0066] First, obtain multi-source data associated with patients at multiple data sources. At the same time, perform continuous time recording based on the time corresponding to various types of multi-source data of the patients, and obtain continuous travel locations based on the behavior locations corresponding to various types of multi-source data of the patients. Correlate the above-mentioned time information and location information analyzed to obtain continuous spatio-temporal data.

[0067] Build behavior chains corresponding to each independent patient through the integrated continuous spatio-temporal data. The behavior chains are characterized as chain data required for subsequent analysis of the movement trajectories of patients after integrating multi-source data.

[0068] Unify and store the behavior chains of several patients to obtain a behavior chain database.

[0069] S200, dividing the monitoring areas and combining the behavior chain database to statistically calculate the population density corresponding to each monitoring area.

[0070] Divide the area, taking each specific place within the area as a unit, and statistically calculate the population density matrix corresponding to each monitoring area in the constructed behavior chain database according to time, which is beneficial to grasping the population aggregation situation in different areas and different time periods, and providing basic data for the assessment of the risk of infectious disease transmission.

[0071] S300, analyzing the behavior chain database to extract key knowledge data associated with the monitoring areas and converting them into knowledge triples, and storing several of the knowledge triples in a graph database.

[0072] The behavior chain database is processed based on semantic analysis and grammatical concepts to construct a knowledge graph corresponding to regions and scenes, and a data model of crowd behavior knowledge is established based on this.

[0073] Analyze the patient behavior chain, identify the patient's mobile behavior data as structured knowledge triples and store them in a graph database that intuitively displays the relationship logic. This link effectively organizes and stores the behavioral knowledge of the crowd, facilitating subsequent reasoning and prediction.

[0074] S400, traversing the graph database to obtain crowd behavior knowledge to construct a behavior graph database, predicting the change characteristics of the crowd density based on a neural network learning model, and analyzing the transmission risk based on the change characteristics in combination with the behavior database to make corresponding warnings.

[0075] First, the corresponding traversal algorithm is used to traverse the data in the graph database to deeply explore the propagation path and chain and capture potential propagation risks. According to the traversal results, the knowledge triples of the graph database are converted into crowd behavior knowledge and stored again to construct a behavior graph database.

[0076] The data in the behavior graph database is used as historical data samples to train the deep learning prediction model. When new patient behavior data is obtained later, the knowledge triples of the corresponding behavior are input into the dynamic prediction model, and the crowd behavior graph database is used to simulate the dynamic spatiotemporal migration of the crowd to obtain the distribution of the spatiotemporal movement state of the crowd. At the same time, various parameters of the model are iterated in real time to refine the model accuracy to predict the regional population density distribution in the next period of time, and realize intelligent monitoring of the spatiotemporal spread of infectious diseases. At the same time, corresponding warnings are issued when a higher risk of infection is obtained after prediction.

[0077] Through the above method, multi-source data integration and synergy effects can be achieved, which can effectively depict the spatiotemporal activity trajectory of patients, and combined with the gathering of people in regional places, provide a more accurate basis for the assessment and early warning of infectious disease transmission; accurate spatiotemporal dynamic monitoring and analysis, by building a patient behavior chain database and combining the division of regional places, comprehensively grasp the spatiotemporal dynamics of patients and changes in regional population density; efficient population density prediction and early warning, based on time series neural networks, can predict changes in regional population density in real time and dynamically, and determine whether the warning threshold of infectious disease outbreaks has been reached, especially under complex spatiotemporal dynamics, the model can identify the trend of population gathering in different regions and time periods, and accurately assess potential risks; improve the scientificity and accuracy of prevention and control decisions: through the reasoning function of knowledge graphs and the predictive ability of neural network learning models, it can accurately simulate the spatiotemporal dynamic migration of people, and provide a scientific basis for the transmission path of infectious diseases.

[0078] like Figure 2As shown, in some other embodiments, multi-source data is obtained based on data sources and data integration is performed to obtain continuous spatio-temporal data. A behavior chain database is constructed according to the continuous spatio-temporal data, including the following steps:

[0079] S110, obtain a hospital data table, extract patient information and its associated medical visit information, and sort them by time to generate continuous medical visit records.

[0080] First, by extracting patient information from data tables of different hospitals and performing data association based on the "outpatient number", each different disease type is regarded as an independent outpatient number, so that the specific information of each medical visit can be accurately traced in the data table corresponding to the patient's medical visit record.

[0081] At the same time, in order to ensure the timeliness of medical visit information, the medical visit information of patients is sorted according to the medical visit time to obtain continuous medical visit records of each patient. These records specifically include outpatient number, medical visit time, diagnosis results, etc.

[0082] S120, obtain the travel data associated with the patient information, extract the residence information corresponding to the stay time exceeding a preset duration based on the travel data to construct a residence chain, and extract the travel destination information corresponding to the stay time exceeding a preset duration based on the travel data to construct a travel chain.

[0083] Next, the travel data of the patient is associated to further construct the residence spatio-temporal information and travel spatio-temporal information of the patient.

[0084] Specifically, the travel data of the patient can provide the continuous residence information and travel destination information of the patient. At the same time, the location information of each location with valid behavior also corresponds to the time the patient stays at these locations.

[0085] The travel data can be the travel information actively registered by the patient through APPs, mini-programs, networked notebooks, etc., and can also include the travel information analyzed by associating networked data such as ticket data, subway ticket data, shopping receipt data, and catering data generated by the patient.

[0086] Combining the location information and the stay time to construct the chain. Specifically, if the patient's stay time at a certain location exceeds 0.5 days, it is considered that the patient has a long stay at this location, and these locations will be included in the chain corresponding to the patient's residence or travel.

[0087] Then, all locations with a stay time exceeding 0.5 days are screened and corresponding to chain nodes based on their types (residence / travel) and stay time to construct a residence chain and a travel chain.

[0088] S130. When the total residence time corresponding to the residence chain and the travel chain is greater than a preset time threshold, merge the residence chain and the travel chain to form the behavior chain associated with the patient information.

[0089] For the residence chain and travel chain of a patient, when the total residence time of the two exceeds a preset threshold, it is considered that the patient's movement between the two locations has a high spatio-temporal correlation. Therefore, the two chains can be merged into an overall patient movement behavior chain.

[0090] The behavior chain includes information on the behavior location and residence time. At the same time, in order to further accurately construct the patient's movement trajectory, the calculation method of the residence time is represented by the following formula:

[0091] ,

[0092] represents the total residence time of the patient at a certain location, represents the timestamp when the patient records the itinerary for the i-th time, represents the number of patient residence records, - represents the residence time between the i-th recorded location and the (i + 1)-th recorded location of the patient.

[0093] Among them, the residence time specifically corresponds to the type of behavior, such as medical treatment time, shopping time, eating time, class time, and so on.

[0094] Through the above formula, the residence time of the patient at different locations can be effectively calculated, so as to judge whether the conditions for merging the residence chain and the travel chain are met.

[0095] S140. Integrate and store the behavior chains corresponding to several pieces of the patient information to obtain the behavior chain database.

[0096] Systematically store and manage the behavior chains of several patients to construct a behavior chain database, which contains the spatio-temporal trajectory information of the patients. This information can not only provide basic data for subsequent infectious disease transmission analysis, but also provide key inputs for subsequent model training and prediction.

[0097] The patient's behavior chain can help us identify potential sources of infection and transmission paths, and thus provide strong support for the spatio-temporal transmission prediction and early warning of infectious diseases. In subsequent links such as knowledge graph construction, population density statistics, and population status prediction, the patient's behavior chain data will serve as an important spatio-temporal information basis to guide the model to perform accurate spatio-temporal dynamic prediction. Therefore, the data integration and behavior chain construction at this stage are not only the first step in infectious disease transmission monitoring and early warning, but also the core data source for subsequent analysis.

[0098] As shown Figure 3 In some other embodiments, a monitoring area is demarcated, and the population density corresponding to each monitoring area is counted in combination with the behavior chain database, including the following steps:

[0099] S210. Demarcate each area based on the demarcation type option to obtain a plurality of the monitoring areas with corresponding results.

[0100] First, demarcate the area. The entire area is divided into multiple sub-areas or places as the monitoring areas. At the same time, in order to ensure the analysis of the aggregation of people at a finer level, different demarcation type criteria can be used in different scenarios.

[0101] Among them, the demarcation type option includes an area option, a function option, and a representative feature option.

[0102] Demarcation by area means that without considering the specific building types in each area, the entire area is divided into several monitoring areas with equal areas only by area.

[0103] Demarcation by function means that the division is carried out according to the overall functional role of the building complex, such as commercial areas, school areas, residential areas, industrial parks, etc.

[0104] Demarcation by representative feature means that the division is carried out based on the specific representative features of the building complex, such as shopping malls, hospitals, squares, subway stations, etc.

[0105] S220. Map the information of each behavior location stored in the behavior chain database to the monitoring area to calculate the number of overlapping patients, and generate the population density corresponding to each monitoring area at each time period in combination with the corresponding residence time.

[0106] Analyze the number of patients staying in different places at a certain time period according to the already constructed behavior chain database of patients. Specifically, after selecting a monitoring area, map the name of the monitoring area to the behavior chain database to analyze the number of matching behavior chains to calculate the aggregation density corresponding to the number of overlapping patients.

[0107] Among them, the calculation formula is:

[0108] ,

[0109] Among them, represents the population density of the monitoring area s within the time period t, represents the number of overlapping patients corresponding to the monitoring area s within the time period t, represents the area of the monitoring area s.

[0110] That is, calculate the number of people staying in the monitoring area by a patient (or other person) within a specific time period, and analyze the crowd gathering situation in different places at different times by calculating the patient density per unit area.

[0111] In some other embodiments, after counting the population density corresponding to each of the monitoring areas, the following steps are further included:

[0112] S230, determine whether the population density is greater than a preset density. If so, mark the monitoring area as a high-risk area.

[0113] Furthermore, when the population density in a certain monitoring area at a specific time period is calculated to be greater than the preset density, mark the monitoring area as a high-risk area. If the population density in the monitoring area is not greater than the preset density, make a regular record and continue the statistical calculation.

[0114] The significance of calculating the population density lies in that it can provide key data support for the subsequent assessment of the risk of infectious disease transmission. For example, when the population density in a certain area is relatively high during peak hours (8:00 - 9:00 am, 11:00 - 12:00 noon, 5:00 - 6:00 pm), it means that the transmission risk in this area is relatively high. Through this density-based analysis method, high-risk areas can be more clearly identified, further guiding the allocation of resources and the formulation of prevention and control strategies. In model prediction, based on this data, the changing trend of the population in the next period of time can be accurately estimated, especially in densely populated areas, potential epidemic outbreak points can be predicted in advance.

[0115] In addition, in population density statistics, we also need to consider the spatio-temporal change characteristics of patients' behaviors. Since the activities of patients are affected by various factors, and the change of their behavior chains in different time periods and different regions is non-linear, therefore, through the statistical analysis of population density, the gathering trend of people in a specific area can be more accurately identified, and it provides a basis for the subsequent spatio-temporal transmission model. By continuously updating the behavior chain data of patients and continuously adjusting and optimizing the statistical model, accurate prediction of population density can be achieved in different time periods and regions, thus providing important decision-making basis for the monitoring and early warning of the spatio-temporal transmission of infectious diseases.

[0116] As Figure 4 shown, in some other embodiments, analyzing the behavior chain database to extract key knowledge data associated with the monitoring area and converting it into knowledge triples includes the following steps:

[0117] S310, generate a number of triple structures corresponding to each of the monitoring areas based on the behavior chain database.

[0118] In order to achieve a structured and systematic expression of the patient's movement behavior, so as to provide strong support for subsequent reasoning and prediction, it is first necessary to disassemble and analyze the behavior chain database through semantic analysis and the application of grammatical concepts, and extract knowledge such as regional locations, scenarios, and population behaviors involved therein, and transform them into structured knowledge triples. The structure of the triple includes a head entity, a relationship, and a tail entity.

[0119] S311, Extract the patient information in the behavior chain database and correspond it to the head entity, extract the behavior location information within the monitoring area and correspond it to the tail entity, and extract the behavior actions and correspond them to the relationship.

[0120] The head entity and the tail entity respectively represent specific elements related to the patient's behavior, while the relationship is used to describe the association between these entities.

[0121] Specifically, taking the example of a patient staying in a certain place, assuming that patient A is shopping in shopping mall X, then the corresponding knowledge triple structure is (patient A, shopping, shopping mall X). This way expresses the knowledge of population behavior very intuitively and can be efficiently stored and queried through a graph database.

[0122] S312, Determine whether the head entity and the tail entity have a namespace.

[0123] S313, If it exists, follow the existing naming rules to assign identifiers. If it does not exist, create a new namespace and assign the identifier based on the naming rules. The identifier is used to identify and explain the associated object.

[0124] At the same time, in order to ensure that entities in different fields and types can be accurately distinguished, each head entity and tail entity will carry the identifier corresponding to the namespace, and use a slash and a real number to distinguish elements of the same type. For example, the identifier of the head entity can be "medical / patient A", and the tail entity may be "functional place / shopping mall X". This way can effectively avoid the confusion of different fields and entity types.

[0125] At the same time, before setting the identifier, it is first necessary to determine whether the head entity and the tail entity have a namespace. The namespace indicates whether there is room for adding an identifier to the entity. If the corresponding namespace of itself exists, it can be directly followed according to the existing naming rules. If there is no namespace, create a new namespace and assign an identifier according to the corresponding naming rules.

[0126] In some other embodiments, storing a plurality of the knowledge triples in a graph database includes the following steps:

[0127] S320. Store the head entity and the tail entity in the nodes of the graph database, and store the relationship in the edges of the graph database.

[0128] The graph database is an ideal choice for storing the above knowledge triples. The graph database contains nodes and edges. The head entity and the tail entity are stored in the nodes of the graph database, while the relationship is stored in the edges of the graph database.

[0129] Nodes contain the relevant attributes of the entity and identifiers for description, while edges contain the relationship types between entities and pointers to entities.

[0130] S321. Define the expression relationship between the associated nodes and edges. The expression relationship represents the relationship strength between the nodes and the edges. Specifically,

[0131] ,

[0132] where, represents the relationship strength from node i to node j, represents the attribute in node i related to relationship m, represents the attribute weight from relationship m to node j, and n is the number of relationships involved.

[0133] Through the above formula, the relationship strength between two nodes can be evaluated, so as to know the reasoning process and help with efficient path reasoning and knowledge query in the graph database.

[0134] When the relationship strength between nodes is strong, it indicates a high degree of association between the two nodes, and the possibility of an outbreak on the transmission path of infectious diseases is high.

[0135] In subsequent intelligent monitoring of the spatio-temporal spread of infectious diseases, with the help of the entity relationship network in the knowledge graph, the model can quickly obtain, analyze, and infer the spatio-temporal behavior characteristics of the population, and then predict potential transmission risks. In addition, the query and reasoning capabilities of the graph database enable us to start from different dimensions for complex simulation and optimization decision-making of infectious disease transmission. This not only improves the utilization efficiency of data but also provides an accurate prediction basis for infectious disease prevention and control strategies.

[0136] In some other embodiments, storing a plurality of the knowledge triples in the graph database further includes the following steps:

[0137] S330. Obtain the number of the knowledge triples.

[0138] To optimize the storage performance and query efficiency, different storage methods need to be selected according to the number of knowledge triples.

[0139] S331, if the quantity is less than the preset value, create a triple table structure, store the head entity corresponding to the subject, the relationship corresponding to the predicate, and the tail entity corresponding to the object, and establish a combined index for the subject and object through B+Tree and store it.

[0140] If the number of knowledge triples is small, they are stored based on a simple indexing method. Specifically, a triple table structure of (subject, predicate, object) is adopted, and a combined index for the subject and object is established through B+Tree.

[0141] B+Tree is a multi-way balanced search tree. Each node can have multiple child nodes. The root node and internal nodes do not store data and are only used for indexing. All data is stored in the leaf nodes.

[0142] Through B+Tree, each head entity and tail entity are combined with the subject and object to obtain the index of the table structure for storage.

[0143] S332, if the quantity is greater than the preset value, record the column increment or row increment in the triple table, and use the minimum bit-width integer type to store the row index and column index to achieve compression. Allocate the columns or rows in the knowledge triples where the relationship strength between the nodes and the edges exceeds the preset value to the same shard for storage.

[0144] If the number of knowledge triples is large, on the one hand, the simple indexing storage method will result in a large amount of redundant and duplicate data, and on the other hand, it is not convenient for accurate search.

[0145] Therefore, more complex compression and sharding techniques need to be adopted at this time to improve the storage and query speed. Specifically, record the row / column increment instead of the absolute value (for example, the difference in row numbers of adjacent elements is stored using a variable-length integer, and the minimum bit-width integer type is used to store the row and column indexes (such as uint16_t to store a matrix with a maximum size of 65535)).

[0146] Allocate the columns or rows with strong relationship strength to the same shard, and determine the storage node through (col_index % shard_num), which is suitable for the column traversal scenario.

[0147] Through the above method, while minimizing the storage space as much as possible, several knowledge triples are stored in a sharded manner, and several knowledge triples with high relevance are stored together, which is convenient for subsequent search.

[0148] Such as Figure 5 As shown, in some other embodiments, traverse the graph database to obtain crowd behavior knowledge to construct a behavior graph database, including the following steps:

[0149] S410, traverse the nodes in the graph database layer by layer through a queue structure to obtain the shortest behavior paths and the shortest spatio-temporal paths between the nodes to complete breadth-first search.

[0150] Use the BFS (breadth-first search) traversal method to achieve hierarchical traversal, which can effectively capture the shortest paths from one node to other nodes. The shortest paths specifically include the shortest behavior paths and the shortest spatio-temporal paths. The shortest behavior path represents the shortest path of population migration and diffusion, so as to identify key transmission nodes. The shortest spatio-temporal path represents the behaviors carried out in a specific area at a specific time in the analysis of population movement trajectories, such as subway transfers, emergency evacuations, etc.

[0151] S420, traverse the non-direct paths between the nodes through recursion or a stack structure to obtain cross-regional behavior associations and explore continuous behavior sequences based on the residence time to obtain long-term behavior associations to complete depth-first exploration.

[0152] DFS (depth-first search) is used for depth traversal to help better explore the relationships between different regions and potential transmission chains.

[0153] Specifically, depth-first search deeply explores branch paths through recursion or a stack structure. For example, analyzing the non-direct path selection of people at nodes across regions reveals the indirect transmission mechanism.

[0154] Track the same or similar behavior sequences of patients for consecutive days to obtain fixed long-term behaviors.

[0155] Secondly, DFS can also connect loops and branches in the graph database to help discover periodic aggregation times, such as weekly market activities and festival activities. It can also analyze the hierarchical interactions within a specific group through the attributes and relationships between nodes, such as families, communities, and cities.

[0156] S430, integrate the search results of the breadth-first search and the depth-first exploration into population behavior knowledge to construct the behavior graph database.

[0157] BFS extracts immediate and short-distance interactions, and DFS supplements long-cycle and cross-regional behavior associations to form a complete spatio-temporal feature matrix. This spatio-temporal feature matrix is used as population behavior knowledge to construct the behavior graph database.

[0158] In some other embodiments, predict the change characteristics of the population density based on a neural network learning model, and analyze the transmission risk based on the change characteristics in combination with the behavior database to perform corresponding warnings, including the following steps:

[0159] S440. Screen several behavior triples in the behavior graph database and use them as training samples to train the first neural network model. Input the population density corresponding to the monitoring area at the current time into the trained first neural network model and obtain the predicted potential population density based on the output result.

[0160] Once the behavior graph database is constructed, the population behavior knowledge stored in it, as historical data, can serve as good samples for training neural network models. The dynamic prediction model based on time series can calculate the change characteristics of the population density in sub-regions and associate them with the population quantity. According to historical data, these change characteristics can help establish warning thresholds and predict the change trend of population density.

[0161] In the embodiment of the present application, the first neural network model based on the time series prediction method is a model of a recurrent neural network (RNN) or a long short-term memory network (LSTM). Such models can capture long-term dependencies in time series data. By training historical data samples, the first neural network model can predict the change of the regional population density in a future period.

[0162] Specifically, to improve the prediction accuracy, the following formula is used to describe the change prediction of population density:

[0163] ,

[0164] where, represents the potential population density for the prediction time period , represents the population density at the current time, is the external influencing factor associated with the monitoring area, represents the model parameters of the first neural network model, and the model parameters are obtained by training several historical data samples from the behavior graph database.

[0165] Through this formula, the model can predict future density changes based on current and historical density data and external factors, and provide a basis for whether to trigger a warning.

[0166] S450. Obtain the density spatial distribution corresponding to each monitoring area in the historical data and convert it into image samples. Use several image samples to train the second neural network model. After the second neural network model obtains the input population density distribution map, it outputs a classification prediction result based on image analysis combined with the potential population density. The classification prediction result includes potential high risk and potential low risk.

[0167] In this application, in addition to time series dynamic prediction, a convolutional neural network (CNN) is also used as a second neural network model for migration prediction.

[0168] Convolutional neural networks are good at processing image data, and in this solution, they are used for the classification prediction of image samples of the spatial distribution of population density. By converting the population density distribution within a region into an image sample to train the second neural network model, the model can classify different spatial distribution patterns and predict the risk of infectious disease transmission in different scenarios.

[0169] Specifically, the second neural network model can distinguish different transmission patterns in potential high-risk areas and potential risk areas after migration prediction in the future, so that it can determine which monitoring areas are more likely to become potential epidemic outbreak points after training.

[0170] S460, trigger an alarm when the output potential population density exceeds a preset value or the output is a potential high-risk.

[0171] When judged by the fusion of two different neural network models, if the model outputs a potential population density exceeding the preset value or the output is a potential high-risk area, an alarm is generated to remind that there is a greater risk of infection in this area.

[0172] The time series model provides the long-term trend of density change, while the convolutional neural network helps with detailed classification in spatial distribution, thus achieving precise monitoring and early warning of the spatio-temporal transmission of infectious diseases. This method of multi-model collaborative work not only enhances the accuracy of the infectious disease early warning system but also provides a scientific basis for the formulation of actual prevention and control measures.

[0173] In some other embodiments, when predicting the change characteristics of the population density based on the neural network learning model, the following steps are further included:

[0174] S470, select high-risk patients based on the population density, obtain the behavior chains corresponding to the high-risk patients and match them with the behavior graph database, and generate iterative data based on the matching results to iteratively update the model parameters of the first neural network model and the second neural network model.

[0175] During the prediction process, the model continuously selects high-risk patients based on the population density in the monitored area and matches their behavior chains with the population behavior map database. Through the comparison of knowledge triples, this process highly correlates the activities of patients with the population density data of the area, thus providing more accurate predictions. After a series of iterative operations, the model can continuously update the movement trajectories of patients, the graph database, and the population density matrix to predict the regional population density distribution in the next time period in real time. The iteration of this prediction process not only improves the accuracy of spatio-temporal propagation prediction but also enhances the real-time performance and sensitivity of the early warning system.

[0176] The implementation principle is as follows:

[0177] It realizes the integration of multi-source data and the synergy effect, can effectively depict the spatio-temporal activity trajectories of patients, and provides a more accurate basis for the assessment and early warning of infectious disease transmission by combining the personnel gathering situation in regional venues; precise spatio-temporal dynamic monitoring and analysis, through constructing a behavior chain database of patients and combining the division of regional venues, comprehensively grasps the spatio-temporal dynamics of patients and the changes in regional population density; efficient population density prediction and early warning, based on a neural network of time series, can predict the changes in regional population density in real time and dynamically, and judge whether the early warning threshold for the outbreak of infectious diseases is reached. Especially in complex spatio-temporal dynamics, the model can identify the population gathering trends in different regions and time periods and accurately assess potential risks; improve the scientificity and accuracy of prevention and control decisions: through the reasoning function of the knowledge graph and the prediction ability of the neural network learning model, it can accurately simulate the spatio-temporal dynamic migration of the population and provide a scientific basis for the transmission path of infectious diseases.

[0178] It should be understood that although the steps in the flowchart of the accompanying drawings are shown sequentially according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, the execution of these steps has no strict order limit and can be executed in other orders.

[0179] The above are all preferred embodiments of this application. It does not limit the protection scope of this application accordingly. Therefore, all equivalent changes made according to the structure, shape, and principle of this application should be covered within the protection scope of this application.

Claims

1. A method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data, characterized in that: The following steps are involved: Acquire multi-source data based on data sources and integrate the data to obtain continuous spatiotemporal data, and construct a behavior chain database based on the continuous spatiotemporal data, wherein the behavior chain database is composed of behavior chains corresponding to a number of patients independently; Divide the monitoring areas and count the crowd density corresponding to each monitoring area in combination with the behavior chain database; Analyzing the behavior chain database to extract key knowledge data associated with the monitoring area and converting the key knowledge data into knowledge triples, and storing a number of the knowledge triples in a graph database; Traverse the graph database to obtain crowd behavior knowledge to build a behavior graph database, predict the change of crowd density based on the neural network learning model, combine the characteristics of the behavior graph database to analyze the transmission risk and make corresponding warnings. Specifically, In the behavior graph database, several behavior triplets are selected and used as training samples to train the first neural network model, the crowd density corresponding to the monitoring area at the current time is input into the trained first neural network model, and the predicted potential crowd density is obtained based on the output result. Specifically, ， in, Characterized as the forecast period The potential population density, Characterized by the density of the crowd at the current time, is an external influencing factor associated with the monitoring area, Characterized as model parameters of the first neural network model; Obtain the density spatial distribution corresponding to each monitoring area in the historical data and convert it into an image sample, use a number of the image samples to train a second neural network model, the second neural network model outputs a classification prediction result based on image analysis combined with the potential crowd density after obtaining the input crowd density distribution map, and the classification prediction result includes potential high risk and potential low risk; When the output potential crowd density exceeds a preset value or the output is a potential high risk, an early warning is triggered.

2. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 1 is characterized in that: Acquiring multi-source data based on data sources and integrating the data to obtain continuous spatiotemporal data, and constructing a behavior chain database based on the continuous spatiotemporal data, including the following steps: Obtain hospital data tables and extract patient information and their associated visit information and sort them by time to generate a continuous visit record; Acquire the travel data associated with the patient information, extract the residence information corresponding to the stay time exceeding the preset time based on the travel data to construct a residence chain, and extract the travel information corresponding to the stay time exceeding the preset time based on the travel data to construct a travel chain; When the sum of the stay times corresponding to the residence chain and the travel chain is greater than a preset time threshold, the residence chain and the travel chain are merged to form the behavior chain associated with the patient information, wherein the behavior chain includes behavior location information and stay time; The behavior chains corresponding to the patient information are integrated into a database to obtain the behavior chain database.

3. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 2 is characterized in that: Dividing the monitoring areas and counting the crowd density corresponding to each monitoring area in combination with the behavior chain database includes the following steps: Dividing each area based on a division type option to obtain a number of monitoring areas of corresponding results, wherein the division type option includes an area option, a function option, and a representative feature option; Based on the mapping of each behavior location information stored in the behavior chain database with the monitoring area to calculate the number of overlapping patients, the crowd density corresponding to each monitoring area in each time period is generated in combination with the corresponding stay time, wherein the calculation formula is: ， in, It is represented by the population density of the monitoring area s in the time period t, It is represented by the number of overlapping patients corresponding to the monitoring area s within the time period t, Characterized by the area of ​​the monitoring area s.

4. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 3 is characterized in that: After counting the crowd density corresponding to each monitoring area, the following steps are also included: Determine whether the crowd density is greater than a preset density, and if so, mark the monitored area as a high-risk area.

5. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 2 is characterized in that: Analyzing the behavior chain database to extract key knowledge data associated with the monitoring area and converting it into knowledge triples includes the following steps: Generate a plurality of triple structures corresponding to each of the monitoring areas based on the behavior chain database, wherein the triple structures include a head entity, a relationship, and a tail entity; Extract the patient information in the behavior chain database and correspond it to the head entity, extract the behavior location information belonging to the monitoring area and correspond it to the tail entity, extract the behavior action and correspond it to the relationship; Determine whether the head entity and the tail entity exist in a namespace; If it exists, the existing naming rules are used to assign the identifier. If it does not exist, a new namespace is created and the identifier is assigned based on the naming rules. The identifier is used to identify and describe the associated object.

6. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 5 is characterized in that: Storing the plurality of knowledge triples in a graph database comprises the following steps: Storing the head entity and the tail entity in nodes of the graph database, and storing the relationship in edges of the graph database; An expression relationship between the associated nodes and the edges is defined, wherein the expression relationship is characterized by the strength of the relationship between the nodes and the edges. Specifically, ， in, It is characterized by the strength of the relationship from node i to node j. Represented as the attributes of node i related to relation m, It is represented as the attribute weight from relationship m to node j, and n is the number of relationships involved.

7. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 6 is characterized in that: Storing the plurality of knowledge triples in a graph database further comprises the following steps: Obtaining the number of the knowledge triples; If the number is less than the preset value, a triple table structure is created, the head entity is stored corresponding to the subject, the relationship is stored corresponding to the predicate, the tail entity is stored corresponding to the object, and a joint index is established for the subject and the object through B+Tree and stored; If the number is greater than the preset value, the column increment or row increment in the triple table is recorded, and the row index and column index are stored using the minimum bit width integer type to achieve compression, and the columns or rows in the knowledge triples whose relationship strength between the node and the edge exceeds the preset value are allocated to the same shard for storage.

8. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 6 is characterized in that: Traversing the graph database to obtain crowd behavior knowledge to build a behavior graph database includes the following steps: Traversing the nodes in the graph database layer by layer through a queue structure to obtain the shortest behavior path and the shortest space-time path between the nodes to complete a breadth-first search; Traversing the indirect paths between the nodes through recursion or stack structure to obtain cross-region behavior associations and exploring continuous behavior sequences based on the stay time to obtain long-term behavior associations to complete depth-first exploration; The search results of the breadth-first search and the depth-first exploration are integrated into crowd behavior knowledge to construct the behavior graph database.

9. The method for monitoring and early warning of the spatiotemporal spread of infectious diseases based on regional medical big data according to claim 8, characterized in that: When predicting the changing characteristics of the crowd density based on the neural network learning model, the following steps are also included: High-risk patients are selected based on the population density, the behavior chains corresponding to the high-risk patients are obtained and matched with the behavior graph database, and iterative data are generated based on the matching results to iteratively update the model parameters of the first neural network model and the second neural network model.

Images