Infectious disease risk warning method, system, computer equipment and medium

Abstract

The present invention relates to the fields of computer technology and digital medicine, and discloses an infectious disease risk warning method, system, computer equipment, and medium. The method includes: collecting and preprocessing infectious disease report cards of a geographic space to be warned to generate infectious disease report card data; performing matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned; inputting the two-dimensional matrix of the geographic space to be warned into a pre-trained infectious disease spatial warning model to output the risk probability of each area in the geographic space to be warned; and performing infectious disease risk warning based on the risk probability of each area. Because the present application performs matrix conversion on the infectious disease report card data, characterizing the data features of the infectious disease report card in the form of a two-dimensional matrix, and combining it with a deep learning model for warning, it is possible to determine complex spatial abnormal growth through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space.

Description

Technical Field

[0001] The present invention relates to the fields of machine learning technology and digital medicine, and in particular to an infectious disease risk early warning method, system, computer equipment, and medium. Background Art

[0002] Infectious diseases are a broad range of diseases caused by pathogens that can be transmitted between humans, animals, or both. Infectious diseases have always been a major health hazard, and some have a tendency to spread. The continued spread of infectious diseases often leads to major public health and safety risks, posing a serious threat not only to human health and life but also to the broader economy and society. With the continuous advancement of machine learning, researchers are eager to use it to improve early warning capabilities for infectious diseases.

[0003] In the existing technology, infectious disease early warning methods are based on statistical methods, such as the cumulative sum control chart method, the moving percentile method, the spatiotemporal scanning statistics method, etc. The above methods can detect relatively simple abnormal growth patterns, but lack the ability to detect complex abnormal growth patterns, thereby reducing the early warning performance of infectious diseases in geographic space. Summary of the Invention

[0004] Based on this, it is necessary to provide an infectious disease risk warning method, system, computer equipment and medium to eliminate the problem of low efficiency of null pointers.

[0005] A method for early warning of infectious disease risks comprises: collecting and preprocessing infectious disease report cards of a geographic space to be warned to generate infectious disease report card data; performing matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned; inputting the two-dimensional matrix of the geographic space to be warned into a pre-trained infectious disease spatial early warning model to output the risk probability of each area in the geographic space to be warned; and performing an infectious disease risk warning based on the risk probability of each area.

[0006] In one embodiment, infectious disease report cards of the geographic space to be warned are collected and preprocessed to generate infectious disease report card data, including: collecting infectious disease report cards of the geographic space to be warned within a preset period; when the number of collected infectious disease report cards is equal to a preset threshold, generating multiple infectious disease report cards; extracting address information of infectious disease cases in each infectious disease report card, and using a modeling method to perform data cleaning operations on the address information of infectious disease cases to generate multiple address information; and determining the multiple address information as infectious disease report card data.

[0007] In one embodiment, a modeling method is used to perform data cleaning operations on the address information of infectious disease cases to generate multiple address information, including: extracting attribute values ​​from multiple address information; constructing a data verification model based on the extracted attribute values ​​and combined with a decision tree algorithm; performing data verification on the address information of infectious disease cases one by one according to the data verification model, and correcting the erroneous parameters found to obtain multiple address information.

[0008] In one embodiment, infectious disease reporting card data is matrix-converted to generate a two-dimensional matrix of the geographic space to be warned, including: dividing the geographic space to be warned into grids according to preset grid parameters and longitude and latitude to obtain a grid matrix; determining the target position of each address information in the grid matrix among multiple address information to obtain the target position of each address information; associating each address information with the grid matrix according to the target position of each address information to obtain a target grid matrix; counting the number of associated people in each grid in the target grid matrix, and filling the counted number of associated people in each grid into the corresponding grid to obtain a two-dimensional matrix of the geographic space to be warned.

[0009] In one embodiment, a pre-trained infectious disease spatial warning model includes a convolution layer, a pooling layer, a fully connected layer, and a normalized exponential function; the two-dimensional matrix of the geographic space to be warned is input into the pre-trained infectious disease spatial warning model, and the risk probability of each area in the geographic space to be warned is output, including: the two-dimensional matrix of the geographic space to be warned is processed in sequence through the convolution layer, the pooling layer, and the fully connected layer to obtain a multidimensional feature vector; based on the multidimensional feature vector, and in combination with the normalized exponential function, a normalization calculation is performed to obtain and output the risk probability of each area in the geographic space to be warned.

[0010] In one embodiment, an infectious disease risk warning is performed based on the risk probability of each area, including: arranging the risk probability of each area in descending order to obtain a sorted risk probability; taking the starting point of the sorted risk probability as a starting point, sequentially intercepting a preset percentage of the risk probability value to obtain multiple target risk probabilities; determining the areas corresponding to the multiple target risk probabilities as high-risk areas; or, comparing the risk probability of each area with a preset value to determine a risk probability greater than the preset value, and determining the area corresponding to the risk probability greater than the preset value as a high-risk area; and performing an infectious disease risk warning for the high-risk area.

[0011] In one embodiment, a pre-trained infectious disease spatial early warning model is generated according to the following steps, including: using a convolutional neural network to construct an infectious disease spatial early warning model; obtaining historical infectious disease report cards and historical early warning data of the infectious disease to be predicted in a preset period in the target area; constructing a two-dimensional matrix of the target area; extracting the location information of the infected person in the historical infectious disease report card, and determining the number of reported cases in each grid in the two-dimensional matrix based on the mapping relationship between the infected person location information and the grids in the two-dimensional matrix; filling the number of reported cases in each grid into the two-dimensional matrix to obtain a first matrix; constructing a dimension vector of the first matrix based on the historical early warning data; associating the dimension vector as a label to the first matrix to obtain a training sample; inputting the training sample into the infectious disease spatial early warning model and outputting a loss value; when the loss value reaches a preset threshold, generating a pre-trained infectious disease spatial early warning model.

[0012] An infectious disease risk warning system includes: an infectious disease report card data generation module, which is used to collect and pre-process infectious disease report cards of a geographic space to be warned to generate infectious disease report card data; a two-dimensional matrix generation module, which is used to perform matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned; a risk probability output module, which is used to input the two-dimensional matrix of the geographic space to be warned into a pre-trained infectious disease space warning model to output the risk probability of each area in the geographic space to be warned; and an infectious disease risk warning module, which is used to issue an infectious disease risk warning based on the risk probability of each area.

[0013] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor executes the steps of the above-mentioned infectious disease risk warning method.

[0014] A medium storing computer-readable instructions, which, when executed by one or more processors, causes the one or more processors to execute the steps of the above-mentioned infectious disease risk warning method.

[0015] The above-mentioned infectious disease risk warning method, system, equipment and medium, the infectious disease risk warning system first collects and pre-processes the infectious disease report card of the geographic space to be warned, generates infectious disease report card data, and then performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned. Secondly, the two-dimensional matrix of the geographic space to be warned is input into the pre-trained infectious disease space warning model, and the risk probability of each area in the geographic space to be warned is output. Finally, an infectious disease risk warning is performed based on the risk probability of each area. Since this application performs matrix conversion on the infectious disease report card data, characterizes the data characteristics of the infectious disease report card in the form of a two-dimensional matrix, and combines the deep learning model for warning, it makes it possible to determine the abnormal growth of complex space through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0017] Figure 1 This is a diagram of the implementation environment of the infectious disease risk early warning method provided in one embodiment of the present application;

[0018] Figure 2 This is a schematic diagram of the internal structure of a computer device in one embodiment of the present application;

[0019] Figure 3 A schematic diagram of a method for early warning of infectious disease risk provided in one embodiment of the present application;

[0020] Figure 4 This is a schematic block diagram of the processing flow of a pre-trained infectious disease spatial early warning model provided in one embodiment of the present application;

[0021] Figure 5 A schematic diagram of a method flow of a spatial early warning model training method for infectious diseases provided in one embodiment of the present application;

[0022] Figure 6 A schematic diagram of a two-dimensional matrix provided in one embodiment of the present application;

[0023] Figure 7 This is a schematic diagram of the system structure of an infectious disease risk warning system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0025] It will be understood that the terms "first," "second," etc. used herein may be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish a first element from another element.

[0026] Figure 1 FIG. 1 is an implementation environment diagram of an infectious disease risk early warning method provided in an embodiment, such as Figure 1 As shown, in this implementation environment, a server 110 and a client 120 are included.

[0027] The server 110 can be a server, which can be a standalone server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. For example, it can be a server device that runs a pre-trained infectious disease spatial early warning model. The client 120 collects and pre-processes infectious disease report cards for the geographic space to be warned to generate infectious disease report card data; the client 120 performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix for the geographic space to be warned; the client 120 inputs the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial early warning model running on the server 110, and outputs the risk probability of each area in the geographic space to be warned and returns it to the client 120; the client 120 performs an infectious disease risk warning based on the risk probability of each area.

[0028] It should be noted that the client 120 can be a smart phone, tablet computer, laptop computer, desktop computer, etc., but is not limited thereto. The server 110 and the client 120 can be connected via Bluetooth, USB (Universal Serial Bus) or other communication connection methods, and the present invention does not limit this.

[0029] Figure 2 FIG. 1 is a schematic diagram of the internal structure of a computer device in one embodiment. Figure 2As shown, the computer device includes a processor, a medium, a memory and a network interface connected via a system bus. Among them, the medium of the computer device stores an operating system, a database and computer-readable instructions, and the database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor may implement an infectious disease risk warning method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor may execute an infectious disease risk warning method. The network interface of the computer device is used to connect and communicate with the terminal. Those skilled in the art will understand that Figure 2 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application and does not limit the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement. The medium is a readable storage medium.

[0030] The following will be combined with the Figure 3-6 This article details the infectious disease risk warning method provided in the embodiments of this application. This method can be implemented using a computer program and run on an infectious disease risk warning system based on the von Neumann architecture. This computer program can be integrated into an application or run as a standalone tool application.

[0031] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0032] Fundamental AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, pandemic risk warning technology, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0033] See Figure 3 , provides a flow chart of an infectious disease risk warning method according to an embodiment of the present application. Figure 3 As shown, the method of the embodiment of the present application may include the following steps:

[0034] S101, collecting and pre-processing infectious disease report cards in the geographic space to be warned, and generating infectious disease report card data;

[0035] The geographic space to be warned is the spatial area that requires warning, such as a designated city, a designated province, or the entire region of a country. The infectious disease report card is a card that the responsible reporter of an epidemic situation must fill out in a timely manner when discovering a notifiable infectious disease.

[0036] Typically, when collecting data, it can be infectious disease report cards reported from specific channels or downloaded from the cloud. Preprocessing involves cleaning the data in the infectious disease report cards to extract the correct required data.

[0037] In an embodiment of the present application, when collecting and preprocessing infectious disease report cards of the geographic space to be warned and generating infectious disease report card data, the infectious disease report cards of the geographic space to be warned within a preset period are first collected, and then when the number of collected infectious disease report cards is equal to a preset threshold, multiple infectious disease report cards are generated, and then the address information of the infectious disease case in each infectious disease report card is extracted, and the address information of the infectious disease case is cleaned by a modeling method to generate multiple address information, and finally the multiple address information is determined as infectious disease report card data.

[0038] Specifically, when using the modeling method to perform data cleaning operations on the address information of infectious disease cases and generate multiple address information, the attribute values ​​in the multiple address information are first extracted, and then a data verification model is constructed based on the extracted attribute values ​​and combined with the decision tree algorithm. Finally, the address information of the infectious disease cases is verified one by one according to the data verification model, and the erroneous parameters found are corrected to obtain multiple address information.

[0039] In one possible implementation, first, the user terminal connects to the cloud server, logs in to the infectious disease report card download system with the specified account and password, and then sets the required download period and data download parameters such as the geographic space to be warned. Secondly, after the cloud server receives the download request instruction, the cloud server collects the infectious disease report cards of the geographic space to be warned in the database according to the set data download parameters, and sends the infectious disease report cards of the geographic space to be warned to the user terminal. The user terminal collects the infectious disease report cards of the geographic space to be warned within the preset period, and after data cleaning of the collected infectious disease report cards, obtains multiple address information, and finally determines the multiple address information as infectious disease report card data.

[0040] S102, performing matrix conversion on infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned;

[0041] Among them, matrix transformation is to mathematically quantify the number of infectious diseases in the infectious disease reporting data combined with the spatial position.

[0042] In an embodiment of the present application, when the infectious disease report card data is matrix-converted to generate a two-dimensional matrix of the geographic space to be warned, the geographic space to be warned is first grid-divided according to preset grid parameters and longitude and latitude to obtain a grid matrix, and then the target position of each address information in the grid matrix is ​​determined to obtain the target position of each address information. Secondly, each address information is associated with the grid matrix according to the target position of each address information to obtain a target grid matrix. Finally, the number of associated people in each grid in the target grid matrix is ​​counted, and the counted number of associated people in each grid is filled into the corresponding grid to obtain a two-dimensional matrix of the geographic space to be warned.

[0043] S103, inputting the two-dimensional matrix of the geographic space to be warned into a pre-trained infectious disease spatial warning model, and outputting the risk probability of each area in the geographic space to be warned;

[0044] The infectious disease spatial early warning model is a mathematical model for predicting the development trend of infectious diseases. The pre-trained infectious disease spatial early warning model includes convolutional layers, pooling layers, fully connected layers, and a normalized exponential function.

[0045] In an embodiment of the present application, when the two-dimensional matrix of the geographic space to be warned is input into a pre-trained infectious disease spatial warning model and the risk probability of each area in the geographic space to be warned is output, the two-dimensional matrix of the geographic space to be warned is first processed in sequence through a convolution layer, a pooling layer, and a fully connected layer to obtain a multidimensional feature vector, and then a normalization calculation is performed based on the multidimensional feature vector and combined with a normalized exponential function to obtain and output the risk probability of each area in the geographic space to be warned.

[0046] For example Figure 4 As shown, Figure 4 This is a schematic diagram of the processing flow of the pre-trained infectious disease spatial warning model provided in this application. First, after obtaining the two-dimensional matrix of the geographic space to be warned, for example, the two-dimensional matrix is ​​1024*1024*1, the two-dimensional matrix is ​​input into the convolution layer for convolution processing to obtain 1024*1024*64 feature data, and then the 1024*1024*64 feature data is input into the pooling layer to be processed into 256*256*64 feature data, and the 256*256*64 feature data is processed through two convolution layers and pooling layers, and the 256*256*64 feature data is converted into a feature vector of 16384 units. The feature vector of 16384 units is processed through three fully connected layers to obtain 256 units. Finally, the normalization layer (Softmax function) is set to perform data normalization processing, and the probability of 256 warning areas is obtained in combination with the loss function of the model.

[0047] Furthermore, the normalized exponential function, or Softmax function, is a generalization of the logistic function. It can "compress" a K-dimensional vector z containing arbitrary real numbers into another K-dimensional real vector σ(z), such that each element ranges between (0, 1) and the sum of all elements is 1.

[0048] S104: Conduct infectious disease risk warning based on the risk probability of each area.

[0049] In one possible implementation, when issuing infectious disease risk warnings based on the risk probabilities of various regions, the risk probabilities of various regions are first arranged in descending order to obtain sorted risk probabilities. Then, starting from the starting point of the sorted risk probabilities, preset percentages of risk probability values ​​are intercepted in sequence to obtain multiple target risk probabilities. Finally, the regions corresponding to the multiple target risk probabilities are determined as high-risk areas, and infectious disease risk warnings are issued for the high-risk areas.

[0050] In another possible implementation, when issuing infectious disease risk warnings based on the risk probabilities of various regions, the risk probabilities of various regions are first compared with preset values ​​to determine risk probabilities greater than the preset values, and regions corresponding to risk probabilities greater than the preset values ​​are determined as high-risk regions, and then infectious disease risk warnings are issued for the high-risk regions.

[0051] In an embodiment of the present application, the infectious disease risk warning system first collects and pre-processes the infectious disease report cards of the geographic space to be warned, generates infectious disease report card data, and then performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned. Secondly, the two-dimensional matrix of the geographic space to be warned is input into a pre-trained infectious disease space warning model, and the risk probability of each area in the geographic space to be warned is output. Finally, an infectious disease risk warning is performed based on the risk probability of each area. Since the present application performs matrix conversion on the infectious disease report card data, characterizes the data features of the infectious disease report card in the form of a two-dimensional matrix, and combines the deep learning model for warning, it makes it possible to determine the abnormal growth of complex space through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space.

[0052] See Figure 5 , which is a flow chart of a method for training a spatial early warning model for infectious diseases according to an embodiment of the present application. Figure 5 As shown, the method of the embodiment of the present application may include the following steps:

[0053] S201, using convolutional neural networks to build a spatial early warning model for infectious diseases;

[0054] S202, obtaining historical infectious disease report cards and historical warning data of the infectious disease to be predicted in the target area within a preset period;

[0055] S203, constructing a two-dimensional matrix of the target area;

[0056] S204, extracting the location information of infected persons from the historical infectious disease report cards, and determining the number of reported cases in each grid in the two-dimensional matrix based on the mapping relationship between the location information of infected persons and the grids in the two-dimensional matrix;

[0057] S205, filling the number of reported cases in each grid into a two-dimensional matrix to obtain a first matrix;

[0058] S206, constructing a dimension vector of the first matrix based on the historical warning data;

[0059] S207, associating the dimension vector as a label into the first matrix to obtain a training sample;

[0060] In one possible implementation, a large amount of historical warning data for the infectious disease to be predicted is collected as model training data. Specifically, this historical warning data includes, for example, report cards for the infectious disease from the past five or ten years. Each report card contains case data including the patient's address. This collected historical warning data includes actual warning information—that is, confirmed warnings issued on a specific day for a specific area. This warning information is organized into labeled data for use as marking data.

[0061] Specifically, the geographical space of the warning, such as a city, is divided into N*N grids according to longitude and latitude, for example, N is 1024, so as to obtain a 1024*1024 size matrix. A certain day (or other time units, such as hours, etc.) corresponds to a matrix representing infectious disease report card data. For example, on January 10, the infectious disease report card data within a fixed time window, such as 7 days, that is, the infectious disease report card data from January 4 to January 10, is placed into the matrix according to the address of each report card data, such as Figure 6 As shown. Figure 6 In the table, each number represents the number of reported cases of infectious diseases in this area.

[0062] Furthermore, a two-dimensional matrix is ​​generated for each day to represent the current spatial infectious disease case information. Since each day corresponds to a spatial warning message, that is, whether there is a warning for a certain area that day, warnings are generally issued at the street level (or other spatial dimensions such as districts and counties). For example, if there are M streets, the corresponding warning information for each day is converted into an M-dimensional vector, where 1 indicates that the corresponding street has a spatial warning message that day, and 0 indicates that the corresponding street does not have a spatial warning message that day. After using the M-dimensional vector as the label data, the resulting N*N two-dimensional matrix is ​​used as the input of the model.

[0063] S208, inputting the training sample into the infectious disease spatial early warning model and outputting the loss value;

[0064] S209: When the loss value reaches a preset threshold, a pre-trained infectious disease spatial early warning model is generated.

[0065] In one possible implementation, when the loss value reaches a preset threshold, a pre-trained infectious disease spatial early warning model is generated; or when the loss value is less than or equal to the preset threshold, the loss value is fed back and optimized to update the model weights, and the step of inputting the training sample into the infectious disease spatial early warning model is continued until the training is stopped when the loss value reaches the preset threshold.

[0066] In an embodiment of the present application, the infectious disease risk warning system first collects and pre-processes the infectious disease report cards of the geographic space to be warned, generates infectious disease report card data, and then performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned. Secondly, the two-dimensional matrix of the geographic space to be warned is input into a pre-trained infectious disease space warning model, and the risk probability of each area in the geographic space to be warned is output. Finally, an infectious disease risk warning is performed based on the risk probability of each area. Since the present application performs matrix conversion on the infectious disease report card data, characterizes the data features of the infectious disease report card in the form of a two-dimensional matrix, and combines the deep learning model for warning, it makes it possible to determine the abnormal growth of complex space through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space.

[0067] The following are system embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the system embodiments of the present invention, please refer to the method embodiments of the present invention.

[0068] See Figure 7 , which shows a schematic diagram of the structure of an infectious disease risk warning system provided by an exemplary embodiment of the present invention. This infectious disease risk warning system can be implemented as all or part of a device through software, hardware, or a combination of both. The system 1 includes an infectious disease report data generation module 10, a two-dimensional matrix generation module 20, a risk probability output module 30, and an infectious disease risk warning module 40.

[0069] The infectious disease report card data generation module 10 is used to collect and pre-process the infectious disease report cards of the geographical space to be warned and generate infectious disease report card data;

[0070] A two-dimensional matrix generation module 20 is used to perform matrix conversion on infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned;

[0071] The risk probability output module 30 is used to input the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial warning model and output the risk probability of each area in the geographic space to be warned;

[0072] The infectious disease risk warning module 40 is used to provide infectious disease risk warning based on the risk probability of each area.

[0073] It should be noted that the infectious disease risk warning system provided in the above embodiment only uses the division of the above-mentioned functional modules as an example when executing the infectious disease risk warning method. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the infectious disease risk warning system provided in the above embodiment and the infectious disease risk warning method embodiment belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0074] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0075] In an embodiment of the present application, the infectious disease risk warning system first collects and pre-processes the infectious disease report cards of the geographic space to be warned, generates infectious disease report card data, and then performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned. Secondly, the two-dimensional matrix of the geographic space to be warned is input into a pre-trained infectious disease space warning model, and the risk probability of each area in the geographic space to be warned is output. Finally, an infectious disease risk warning is performed based on the risk probability of each area. Since the present application performs matrix conversion on the infectious disease report card data, characterizes the data features of the infectious disease report card in the form of a two-dimensional matrix, and combines the deep learning model for warning, it makes it possible to determine the abnormal growth of complex space through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space.

[0076] In one embodiment, a computer device is provided. The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0077] Collect and pre-process infectious disease report cards in the geographic space to be warned, and generate infectious disease report card data;

[0078] Perform matrix conversion on infectious disease reporting data to generate a two-dimensional matrix of the geographic space to be warned;

[0079] Input the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial warning model, and output the risk probability of each area in the geographic space to be warned;

[0080] Conduct infectious disease risk warnings based on the risk probability of each region.

[0081] In one embodiment, when the processor collects and pre-processes infectious disease report cards for the geographic space to be warned and generates infectious disease report card data, the processor specifically performs the following operations:

[0082] Collect infectious disease report cards within a preset period in the geographical space to be warned;

[0083] When the number of collected infectious disease report cards equals a preset threshold, generating multiple infectious disease report cards;

[0084] Extract the address information of infectious disease cases from each infectious disease report card, and use the modeling method to perform data cleaning operations on the address information of infectious disease cases to generate multiple address information;

[0085] Multiple address information is determined as infectious disease reporting card data.

[0086] In one embodiment, the processor performs a data cleansing operation on the address information of infectious disease cases using a modeling method to generate multiple address information by specifically performing the following operations:

[0087] Extract attribute values ​​from multiple address information;

[0088] Based on the extracted attribute values, a data verification model is constructed in combination with the decision tree algorithm;

[0089] The address information of infectious disease cases is verified one by one according to the data verification model, and the erroneous parameters found are corrected to obtain multiple address information.

[0090] In one embodiment, when the processor performs matrix conversion on infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned, the following operations are specifically performed:

[0091] Divide the geographical space to be warned into grids according to preset grid parameters and longitude and latitude to obtain a grid matrix;

[0092] Determining a target position of each address information in the grid matrix among the plurality of address information to obtain the target position of each address information;

[0093] Associating each address information with a grid matrix according to a target position of each address information to obtain a target grid matrix;

[0094] The number of people associated with each grid in the target grid matrix is ​​counted, and the number of people associated with each grid is filled into the corresponding grid to obtain a two-dimensional matrix of the geographic space to be warned.

[0095] In one embodiment, when the processor inputs the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial warning model and outputs the risk probability of each area in the geographic space to be warned, the processor specifically performs the following operations:

[0096] The two-dimensional matrix of the geographic space to be warned is processed sequentially through the convolution layer, pooling layer, and fully connected layer to obtain a multi-dimensional feature vector;

[0097] Based on the multidimensional feature vector and combined with the normalized exponential function, normalization calculation is performed to obtain and output the risk probability of each area in the geographical space to be warned.

[0098] In one embodiment, when the processor performs an infectious disease risk warning based on the risk probability of each area, the processor specifically performs the following operations:

[0099] Arrange the risk probabilities of each area in descending order to obtain the sorted risk probabilities;

[0100] Taking the starting point of the ranked risk probabilities as the starting point, sequentially intercepting the risk probability values ​​of preset percentages to obtain multiple target risk probabilities;

[0101] Identify areas corresponding to multiple target risk probabilities as high-risk areas;

[0102] or,

[0103] Compare the risk probability of each area with a preset value to determine a risk probability greater than the preset value, and determine the area corresponding to the risk probability greater than the preset value as a high-risk area;

[0104] Conduct infectious disease risk warnings for high-risk areas.

[0105] In one embodiment, before the processor collects and pre-processes infectious disease report cards for the geographic space to be warned and generates infectious disease report card data, it further performs the following operations:

[0106] Constructing a spatial early warning model for infectious diseases using convolutional neural networks;

[0107] Obtain historical infectious disease report cards and historical early warning data for the infectious disease to be predicted within a preset period in the target area;

[0108] Construct a two-dimensional matrix of the target area;

[0109] Extract the location information of infected persons from historical infectious disease report cards, and determine the number of reported cases in each grid in the two-dimensional matrix based on the mapping relationship between the location information of infected persons and the grids in the two-dimensional matrix;

[0110] Fill the number of reported cases in each grid into the two-dimensional matrix to obtain the first matrix;

[0111] Construct the dimension vector of the first matrix according to the historical warning data;

[0112] Associate the dimension vector as a label to the first matrix to obtain a training sample;

[0113] Input the training samples into the infectious disease spatial early warning model and output the loss value;

[0114] When the loss value reaches the preset threshold, a pre-trained infectious disease spatial early warning model is generated.

[0115] In an embodiment of the present application, the infectious disease risk warning system first collects and pre-processes the infectious disease report cards of the geographic space to be warned, generates infectious disease report card data, and then performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned. Secondly, the two-dimensional matrix of the geographic space to be warned is input into a pre-trained infectious disease space warning model, and the risk probability of each area in the geographic space to be warned is output. Finally, an infectious disease risk warning is performed based on the risk probability of each area. Since the present application performs matrix conversion on the infectious disease report card data, characterizes the data features of the infectious disease report card in the form of a two-dimensional matrix, and combines the deep learning model for warning, it makes it possible to determine the abnormal growth of complex space through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space.

[0116] In one embodiment, a medium storing computer-readable instructions is provided. When the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the following steps:

[0117] Collect and pre-process infectious disease report cards in the geographic space to be warned, and generate infectious disease report card data;

[0118] Perform matrix conversion on infectious disease reporting data to generate a two-dimensional matrix of the geographic space to be warned;

[0119] Input the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial warning model, and output the risk probability of each area in the geographic space to be warned;

[0120] Conduct infectious disease risk warnings based on the risk probability of each region.

[0121] In one embodiment, when the processor collects and pre-processes infectious disease report cards for the geographic space to be warned and generates infectious disease report card data, the processor specifically performs the following operations:

[0122] Collect infectious disease report cards within a preset period in the geographical space to be warned;

[0123] When the number of collected infectious disease report cards equals a preset threshold, generating multiple infectious disease report cards;

[0124] Extract the address information of infectious disease cases from each infectious disease report card, and use the modeling method to perform data cleaning operations on the address information of infectious disease cases to generate multiple address information;

[0125] Multiple address information is determined as infectious disease reporting card data.

[0126] In one embodiment, the processor performs a data cleansing operation on the address information of infectious disease cases using a modeling method to generate multiple address information by specifically performing the following operations:

[0127] Extract attribute values ​​from multiple address information;

[0128] Based on the extracted attribute values, a data verification model is constructed in combination with the decision tree algorithm;

[0129] The address information of infectious disease cases is verified one by one according to the data verification model, and the erroneous parameters found are corrected to obtain multiple address information.

[0130] In one embodiment, when the processor performs matrix conversion on infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned, the following operations are specifically performed:

[0131] Divide the geographical space to be warned into grids according to preset grid parameters and longitude and latitude to obtain a grid matrix;

[0132] Determining a target position of each address information in the grid matrix among the plurality of address information to obtain the target position of each address information;

[0133] Associating each address information with a grid matrix according to a target position of each address information to obtain a target grid matrix;

[0134] The number of people associated with each grid in the target grid matrix is ​​counted, and the number of people associated with each grid is filled into the corresponding grid to obtain a two-dimensional matrix of the geographic space to be warned.

[0135] In one embodiment, when the processor inputs the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial warning model and outputs the risk probability of each area in the geographic space to be warned, the processor specifically performs the following operations:

[0136] The two-dimensional matrix of the geographic space to be warned is processed sequentially through the convolution layer, pooling layer, and fully connected layer to obtain a multi-dimensional feature vector;

[0137] Based on the multidimensional feature vector and combined with the normalized exponential function, normalization calculation is performed to obtain and output the risk probability of each area in the geographical space to be warned.

[0138] In one embodiment, when the processor performs an infectious disease risk warning based on the risk probability of each area, the processor specifically performs the following operations:

[0139] Arrange the risk probabilities of each area in descending order to obtain the sorted risk probabilities;

[0140] Taking the starting point of the ranked risk probabilities as the starting point, sequentially intercepting the risk probability values ​​of preset percentages to obtain multiple target risk probabilities;

[0141] Identify areas corresponding to multiple target risk probabilities as high-risk areas;

[0142] or,

[0143] Compare the risk probability of each area with a preset value to determine a risk probability greater than the preset value, and determine the area corresponding to the risk probability greater than the preset value as a high-risk area;

[0144] Conduct infectious disease risk warnings for high-risk areas.

[0145] In one embodiment, before the processor collects and pre-processes infectious disease report cards for the geographic space to be warned and generates infectious disease report card data, it further performs the following operations:

[0146] Constructing a spatial early warning model for infectious diseases using convolutional neural networks;

[0147] Obtain historical infectious disease report cards and historical early warning data for the infectious disease to be predicted within a preset period in the target area;

[0148] Construct a two-dimensional matrix of the target area;

[0149] Extract the location information of infected persons from historical infectious disease report cards, and determine the number of reported cases in each grid in the two-dimensional matrix based on the mapping relationship between the location information of infected persons and the grids in the two-dimensional matrix;

[0150] Fill the number of reported cases in each grid into the two-dimensional matrix to obtain the first matrix;

[0151] Construct the dimension vector of the first matrix according to the historical warning data;

[0152] Associate the dimension vector as a label to the first matrix to obtain a training sample;

[0153] Input the training samples into the infectious disease spatial early warning model and output the loss value;

[0154] When the loss value reaches the preset threshold, a pre-trained infectious disease spatial early warning model is generated.

[0155] In an embodiment of the present application, the infectious disease risk warning system first collects and pre-processes the infectious disease report cards of the geographic space to be warned, generates infectious disease report card data, and then performs matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned. Secondly, the two-dimensional matrix of the geographic space to be warned is input into a pre-trained infectious disease space warning model, and the risk probability of each area in the geographic space to be warned is output. Finally, an infectious disease risk warning is performed based on the risk probability of each area. Since the present application performs matrix conversion on the infectious disease report card data, characterizes the data features of the infectious disease report card in the form of a two-dimensional matrix, and combines the deep learning model for warning, it makes it possible to determine the abnormal growth of complex space through deep learning network technology on the basis of big data processing, thereby improving the warning performance of infectious diseases in the geographic space.

[0156] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable medium, and when executed, the program can include the processes in the above-described method embodiments. The aforementioned medium can be a non-volatile medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0157] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] The above embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for early warning of infectious disease risk, characterized in that: The method comprises: Collect and pre-process infectious disease report cards in the geographic space to be warned, and generate infectious disease report card data; The collecting and pre-processing of infectious disease report cards in the geographic space to be warned to generate infectious disease report card data includes: Collect infectious disease report cards within a preset period in the geographical space to be warned; When the number of collected infectious disease report cards equals a preset threshold, generating multiple infectious disease report cards; Extracting the address information of the infectious disease case from each infectious disease report card, and performing a data cleaning operation on the address information of the infectious disease case using a modeling method to generate multiple address information; Determining the plurality of address information as infectious disease reporting card data; Performing matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned; The matrix conversion of the infectious disease report card data to generate a two-dimensional matrix of the geographic space to be warned includes: Divide the geographical space to be warned into grids according to preset grid parameters and longitude and latitude to obtain a grid matrix; Determine a target position of each piece of address information in the grid matrix to obtain the target position of each piece of address information; Associating each address information with the grid matrix according to the target position of each address information to obtain a target grid matrix; Counting the number of people associated with each grid in the target grid matrix, and filling the counted number of people associated with each grid into the corresponding grid to obtain a two-dimensional matrix of the geographic space to be warned; Inputting the two-dimensional matrix of the geographic space to be warned into a pre-trained infectious disease spatial warning model, and outputting the risk probability of each area in the geographic space to be warned; Conduct infectious disease risk warnings based on the risk probabilities of each area; The following steps are used to generate a pre-trained infectious disease spatial early warning model, including: Constructing a spatial early warning model for infectious diseases using convolutional neural networks; Obtain historical infectious disease report cards and historical early warning data for the infectious disease to be predicted within a preset period in the target area; constructing a two-dimensional matrix of the target area; Extracting the location information of infected persons from the historical infectious disease report card, and determining the number of reported cases in each grid in the two-dimensional matrix based on a mapping relationship between the location information of infected persons and the grids in the two-dimensional matrix; Filling the number of reported cases in each grid into the two-dimensional matrix to obtain a first matrix; Constructing a dimension vector of the first matrix according to the historical warning data; Associating the dimension vector as a label to the first matrix to obtain a training sample; Inputting the training samples into the infectious disease spatial early warning model and outputting a loss value; When the loss value reaches a preset threshold, a pre-trained infectious disease spatial early warning model is generated.

2. The method according to claim 1, characterized in that The modeling method is used to perform data cleaning operations on the address information of the infectious disease case to generate multiple address information, including: extracting attribute values ​​from the plurality of address information; Based on the extracted attribute values, a data verification model is constructed in combination with the decision tree algorithm; The address information of the infectious disease cases is verified one by one according to the data verification model, and the erroneous parameters found are corrected to obtain multiple address information.

3. The method according to claim 1, characterized in that The pre-trained infectious disease spatial early warning model includes a convolutional layer, a pooling layer, a fully connected layer, and a normalized exponential function; Input the two-dimensional matrix of the geographic space to be warned into the pre-trained infectious disease spatial warning model, and output the risk probability of each area in the geographic space to be warned, including: The two-dimensional matrix of the geographic space to be warned is processed sequentially through the convolution layer, the pooling layer, and the fully connected layer to obtain a multi-dimensional feature vector; According to the multidimensional feature vector, normalization calculation is performed in combination with the normalized exponential function to obtain and output the risk probability of each area in the geographic space to be warned.

4. The method according to claim 1, wherein The infectious disease risk warning based on the risk probability of each area includes: Arrange the risk probabilities of the various regions in descending order to obtain sorted risk probabilities; Taking the starting point of the ranked risk probabilities as the starting point, sequentially intercepting the risk probability values ​​of preset percentages to obtain multiple target risk probabilities; Determining areas corresponding to the multiple target risk probabilities as high-risk areas; or, Comparing the risk probability of each area with a preset value to determine a risk probability greater than the preset value, and determining the area corresponding to the risk probability greater than the preset value as a high-risk area; Issue infectious disease risk warnings for the high-risk areas.

5. An infectious disease risk early warning system implemented using the method according to any one of claims 1 to 4, characterized in that: The system comprises: The infectious disease report card data generation module is used to collect and pre-process the infectious disease report cards in the geographical space to be warned and generate infectious disease report card data; A two-dimensional matrix generation module is used to perform matrix conversion on the infectious disease report card data to generate a two-dimensional matrix of the geographical space to be warned; A risk probability output module is used to input the two-dimensional matrix of the geographic space to be warned into a pre-trained infectious disease spatial warning model and output the risk probability of each area in the geographic space to be warned; The infectious disease risk warning module is used to provide infectious disease risk warning based on the risk probability of each area.

6. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the infectious disease risk warning method as described in any one of claims 1 to 4.

7. A medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the steps of infectious disease risk warning as described in any one of claims 1 to 4.

Images