Epidemic situation development prediction method and device, computer equipment and storage medium

Abstract

The invention provides an epidemic situation development prediction method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining standardized epidemic situation data of a target area, wherein the standardized epidemic situation data comprises epidemic situation data of a plurality of channels; performing feature extraction on the epidemic situation data of each channel in the standardized epidemic situation data by using multi-channel adaptive Fourier decomposition to obtain epidemic situation features; calling a sensing disease transmission model based on the standardized epidemic situation data to start to simulate the basic transmission dynamics of the epidemic situation to obtain epidemic situation transmission dynamic information; and inputting the epidemic situation characteristics and the epidemic situation propagation dynamic information into a target epidemic situation development prediction model obtained based on recurrent neural network training to obtain an epidemic situation development prediction result. According to the invention, the accuracy and comprehensiveness of epidemic situation development prediction can be improved.

Description

Technical Field

[0001] The present application relates to the field of epidemic prediction technology, and in particular to a method, device, computer equipment and storage medium for predicting the development of an epidemic. Background Art

[0002] The SIR model is an infectious disease transmission model used to abstractly describe the transmission process of infectious diseases. It is a classic model among infectious disease models, where S stands for Susceptible, I stands for Infected, and R stands for Recovered.

[0003] There are related technologies that use the SIR model to predict the development trend of the epidemic. However, the inventors found during the implementation process that the SIR model assumes that the population is evenly mixed, which is a simplified assumption, and the model does not fully consider the heterogeneity of the population, the complex social network structure, and the variable transmission paths in reality. This will cause the SIR model to be difficult to capture the nonlinear and non-stationary characteristics of the propagation process when facing complex epidemic data, thereby limiting the accuracy of the final prediction results. In addition, epidemic data usually include multi-channel data (such as the number of infected people, the use of medical resources, social isolation measures, and other channels), and there are complex dynamic correlations between these data, but traditional prediction methods often cannot make full use of these multi-dimensional information, which makes the prediction model unable to accurately and comprehensively predict the development trend of the epidemic. Summary of the invention

[0004] The present application provides a method, apparatus, computer equipment and storage medium for predicting the development of an epidemic, aiming to solve the problem that the SIR model has difficulty in capturing the nonlinear and non-stationary characteristics of the propagation process when facing complex epidemic data, thereby limiting the accuracy of the final prediction results, as well as the problem of insufficient accuracy of the prediction results due to the inability to fully utilize the multi-dimensional information in the epidemic data.

[0005] In a first aspect, the present application provides a method for predicting the development of an epidemic. In one embodiment, the method includes:

[0006] Obtain standardized epidemic data of the target area, which includes epidemic data of multiple channels;

[0007] Use multi-channel adaptive Fourier decomposition to extract features from the epidemic data of each channel in the standardized epidemic data to obtain epidemic characteristics;

[0008] Based on the standardized epidemic data, the sensor disease transmission model is called to start simulating the basic transmission dynamics of the epidemic and obtain the dynamic information of the epidemic transmission;

[0009] The epidemic characteristics and epidemic spread dynamic information are input into the target epidemic development prediction model obtained by training the recurrent neural network to obtain the epidemic development prediction results.

[0010] In one embodiment, the infectious disease propagation model is a SIR model; based on the standardized epidemic data, the sensor disease propagation model is called to start simulating the basic propagation dynamics of the epidemic, and the epidemic propagation dynamics information is obtained, including:

[0011] Determine the initial number of susceptible, infected, and removed individuals based on standardized epidemic data;

[0012] The initial number of susceptible, infected, and removed persons is input into the SIR model, so that the SIR model simulates the basic transmission dynamics of the epidemic based on the preset infection rate and recovery rate, and obtains the number of susceptible, infected, and removed persons that changes over time;

[0013] The number of susceptible, infected, and removed persons changing over time is used as information on the dynamic spread of the epidemic.

[0014] In one embodiment, multi-channel adaptive Fourier decomposition is used to extract features of the epidemic data of each channel in the standardized epidemic data to obtain epidemic features, including:

[0015] Multi-channel adaptive Fourier decomposition is used to perform frequency decomposition on the epidemic data of each channel, extract the main frequency components and dynamic characteristics of the epidemic data of each channel, and obtain the epidemic characteristics.

[0016] In one embodiment, obtaining standardized epidemic data of a target area includes:

[0017] Obtain newly collected original epidemic data of the target area;

[0018] The original epidemic data is cleaned and standardized in turn to obtain standardized epidemic data.

[0019] In one embodiment, the target area is a designated region of a designated country; the standardized epidemic data include epidemic data at multiple stages of epidemic development in the designated region; the epidemic data at each stage of epidemic development include multiple items including the number of newly confirmed cases, number of hospitalizations, utilization rate of intensive care units, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the periodic changes of the epidemic and the dynamic characteristics reflecting the short-term fluctuations and trend changes of the epidemic; the epidemic development forecast results include the number of new cases and hospitalization rate in the designated region in the next week.

[0020] In one embodiment, the target area is multiple countries and regions around the world; the standardized epidemic data include epidemic data of multiple epidemic waves in various countries and regions; the epidemic data of each epidemic wave include multiple items including the daily number of newly confirmed cases, mortality rate, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the cyclical changes of the epidemic and the dynamic characteristics reflecting the short-term drastic changes in the epidemic; the epidemic development forecast results include the number of new cases and epidemic trends in each country and region in the next month.

[0021] In one embodiment, the target area is a designated country; the standardized epidemic data include epidemic data at multiple stages of epidemic development of the designated country; the epidemic data at each stage of epidemic development include multiple items including the daily number of newly confirmed cases, vaccination rate, vaccine types and their effectiveness, number of severe cases and deaths, implementation of public health intervention measures and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the cyclical changes of the epidemic, the dynamic characteristics reflecting the short-term drastic changes in the epidemic, and the vaccination impact characteristics reflecting the impact of vaccination on the spread of the epidemic; the epidemic development forecast results include the epidemic development in the designated country in the next two months.

[0022] In a second aspect, the present application provides a device for predicting the development of an epidemic. In one embodiment, the device includes:

[0023] An epidemic data acquisition module is used to acquire standardized epidemic data of a target area, wherein the standardized epidemic data includes epidemic data of multiple channels;

[0024] An epidemic feature extraction module is used to extract features of the epidemic data of each channel in the standardized epidemic data using multi-channel adaptive Fourier decomposition to obtain epidemic features;

[0025] The epidemic spread simulation module is used to call the sensor disease spread model based on standardized epidemic data to start simulating the basic spread dynamics of the epidemic and obtain the dynamic information of the epidemic spread;

[0026] The epidemic development prediction module is used to input the epidemic characteristics and epidemic spread dynamic information into the target epidemic development prediction model obtained by training the recurrent neural network to obtain the epidemic development prediction results.

[0027] In a third aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the method for predicting the development of an epidemic provided in the above embodiment is implemented.

[0028] In a fourth aspect, the present application provides a computer device comprising: one or more processors; a memory; and one or more computer programs, the processor and the memory being connected via a bus, wherein the one or more computer programs are stored in the memory and configured to be executed by the one or more processors, characterized in that when the processor executes the computer program, the epidemic development prediction method provided in the above-mentioned embodiment is implemented.

[0029] This application innovatively applies multi-channel adaptive Fourier decomposition to epidemic prediction, and combines the disease transmission mechanism of the infectious disease transmission model with the time series modeling capability of the recurrent neural network, so that the target epidemic development prediction model finally trained can more accurately capture the dynamic changes of epidemic transmission, and realize the organic integration of epidemiology and deep learning, so as to overcome the deficiencies or defects of related technologies. Specifically, this application can extract the main frequency components and dynamic characteristics of each channel used for epidemic prediction by adaptively decomposing multi-channel epidemic data. This method can better capture the nonlinear and non-stationary characteristics of the data. On this basis, the sensor disease transmission model is called based on the standardized epidemic data to start simulating the basic transmission dynamics of the epidemic, and the dynamic information of the epidemic transmission is obtained. Finally, the epidemic characteristics and the dynamic information of the epidemic transmission are input into the target epidemic development prediction model as input data, so as to obtain more accurate and comprehensive epidemic development prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flowchart of a method for predicting the development of an epidemic provided in one embodiment of the present application.

[0031] Figure 2 This is a functional module block diagram of an epidemic development prediction device provided in one embodiment of the present application.

[0032] Figure 3 It is a specific structural block diagram of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and beneficial effects of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0034] See also Figure 1 , is a flow chart of an epidemic development prediction method provided in an embodiment of the present application. This embodiment mainly uses the epidemic development prediction method applied to a computer device as an example for illustration. The epidemic development prediction method provided in an embodiment of the present application includes the following steps:

[0035] S101. Obtain standardized epidemic data of a target area, where the standardized epidemic data includes epidemic data of multiple channels.

[0036] The specific content of the target area and standardized epidemic data is related to the prediction task. The target area can be a region in a country, or part of the geographical area or all of the geographical areas of multiple countries. Standardized epidemic data includes epidemic data of multiple channels (types). Standardized epidemic data can include the number of newly confirmed cases, hospitalizations, intensive care unit utilization rate, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data and population mobility data in the target area every day over the past period of time.

[0037] Obtaining standardized epidemic data for the target area includes: obtaining newly collected original epidemic data for the target area; performing data cleaning and standardization on the original epidemic data in turn to obtain standardized epidemic data. Data cleaning includes removing outliers and noise data in the epidemic data to ensure data quality. Standardization is used to eliminate the magnitude differences of data from different channels.

[0038] S102. Use multi-channel adaptive Fourier decomposition to extract features of the epidemic data of each channel in the standardized epidemic data to obtain epidemic features.

[0039] Multi-channel adaptive Fourier decomposition is used to extract features of the epidemic data of each channel in the standardized epidemic data to obtain epidemic characteristics, including: using multi-channel adaptive Fourier decomposition to perform frequency decomposition on the epidemic data of each channel, extracting the main frequency components and dynamic characteristics in the epidemic data of each channel, and obtaining epidemic characteristics.

[0040] Multi-channel adaptive Fourier decomposition performs well in processing complex, nonlinear and non-stationary multi-channel data. Using it for feature extraction can adaptively extract the main frequency components and dynamic features in multi-channel epidemic data, and align the correlation between different channels, thereby improving the accuracy and effectiveness of epidemic feature extraction. It can provide high-quality input features for subsequent prediction models, enabling the prediction model to more accurately capture the dynamic changes in the spread of the epidemic, thereby improving the overall prediction accuracy.

[0041] Specifically, multi-channel adaptive Fourier decomposition can adaptively extract the main frequency components and dynamic features in each channel and align common basis functions between multiple channels. Through this decomposition, a low-dimensional feature set reflecting the key features of the epidemic data is extracted to provide high-quality features for subsequent model input. Among them, the main frequency components represent the periodic changes in the epidemic and represent the long-term spread trend of the epidemic, while the dynamic features (also called fluctuation features) can reflect the short-term fluctuations and trend changes of the epidemic.

[0042] Among them, multi-channel adaptive Fourier decomposition is to extend the adaptive Fourier decomposition (AFD) to the multi-channel case. The adaptive Fourier decomposition can decompose the main trends and periodic fluctuations in the epidemic data by adaptively selecting the appropriate Fourier basis function, and can gradually extract the Fourier components by minimizing the reconstruction error until the error reaches the preset threshold.

[0043] The decomposition steps include:

[0044] a. Initialize the parameters of the Fourier basis function, such as period and amplitude;

[0045] b. Use the least squares method to fit the epidemic data and obtain the initial Fourier component;

[0046] c. Dynamically adjust parameters and minimize reconstruction error through iterative optimization;

[0047] d. Extract the main Fourier components as input for subsequent recurrent neural network modeling.

[0048] Adaptive Fourier Decomposition Using Matching Pursuit Method to Identify Adaptive Basis Function Systems Each function B n It is called the modified Blaschke product. The modified Blaschke product of each basis function is defined as follows:

[0049]

[0050] Among them, a n is the complex coefficient in the unit disk D = {z∈C:|z|<1, and is an adaptive parameter chosen to maximize energy convergence at each decomposition level. For any sequence of a in the unit disk D n are orthogonal. The processed signal G(e jt ) can be expressed as the series expansion form of the adaptive basis function:

[0051]

[0052] Among them, A n For nth th These coefficients are obtained by combining the signal G and the basis function B n The inner product of <G,B n >Calculated.

[0053] In multi-channel adaptive Fourier decomposition (MAFD), all channels use the same number of single components and the same basis functions to maintain the alignment of common oscillations among channels. The total energy to be maximized in multi-channel adaptive Fourier decomposition is:

[0054]

[0055] Where C is the total number of channels, G c,n represents the processed signal of the cth channel in the nth decomposition level. This formula is used to determine the optimal parameter a that maximizes the energy of all channels n .

[0056] S103. Based on the standardized epidemic data, the sensor disease transmission model is called to start simulating the basic transmission dynamics of the epidemic and obtain the dynamic information of the epidemic transmission.

[0057] Among them, the infectious disease transmission model is the SIR model. Based on the standardized epidemic data, the sensor disease transmission model is called to start simulating the basic transmission dynamics of the epidemic, and the dynamic information of the epidemic transmission is obtained, including: determining the initial number of susceptible, infected and removed persons based on the standardized epidemic data; inputting the initial number of susceptible, infected and removed persons into the SIR model, so that the SIR model simulates the basic transmission dynamics of the epidemic based on the preset infection rate and recovery rate, and obtains the number of susceptible, infected and removed persons that changes with time; and using the number of susceptible, infected and removed persons that changes with time as the dynamic information of the epidemic transmission.

[0058] First, based on the SIR model, the standardized epidemic data is divided into three categories: susceptible, infected, and recovered. The initial number of susceptible, infected, and removed people is determined, and then these initial numbers are input into the SIR model to simulate the spread of the epidemic.

[0059] Among them, it is necessary to set appropriate infection rate and recovery rate parameters for the SIR model in advance (the infection rate determines the speed at which susceptible people become infected people, and the recovery rate determines the speed at which infected people recover to removed people), and then let SIR simulate the spread of the epidemic in different populations based on the infection rate and recovery rate. Finally, the dynamic information of the epidemic spread can be obtained, that is, the number of susceptible people, infected people and removed people changing over time.

[0060] S104. Input the epidemic characteristics and epidemic spread dynamic information into the target epidemic development prediction model obtained through recurrent neural network training to obtain the epidemic development prediction result.

[0061] The output of the SIR model is used together with the epidemic characteristics extracted by multi-channel adaptive Fourier decomposition as the input of the recurrent neural network. The recurrent neural network can further model the complex dependencies of time series data through its cyclic structure and achieve dynamic prediction.

[0062] This embodiment integrates multi-channel adaptive Fourier decomposition, SIR model and recurrent neural network in a unified framework, which can simplify the design and tuning process of complex epidemic prediction models. Users can achieve efficient prediction without complex feature engineering, thereby improving the ease of use of the model.

[0063] On this basis, this embodiment also deeply integrates the classic SIR model with the recurrent neural network to form a hybrid prediction model that combines epidemiological principles and deep learning capabilities. The SIR model provides preliminary predictions based on the transmission mechanism of infectious diseases, while the recurrent neural network further optimizes the prediction results by learning complex dependencies in time series data. This deep integration method effectively combines the disease transmission mechanism of the SIR model with the powerful modeling capabilities of the recurrent neural network, so that the target epidemic development prediction model can not only capture short-term epidemic changes, but also better reflect long-term trends and dynamic changes, thereby improving the comprehensiveness and accuracy of the prediction.

[0064] In one embodiment, the training process of the target epidemic development prediction model is as follows:

[0065] Build a SIR-RNN model; RNN refers to a recurrent neural network;

[0066] The SIR-RNN model can be trained using historical epidemic data (original epidemic data collected in the past), and the model parameters can be optimized by supervised learning. During the training process, the back propagation algorithm (such as BPTT, Backpropagation Through Time) and the gradient descent algorithm can be used to update the model parameters to ensure the convergence and stability of the model. At the same time, the hyperparameters of the model can be optimized through methods such as cross-validation to avoid overfitting.

[0067] After training, the SIR-RNN model can predict new epidemic data. In practical applications, as epidemic data changes, the model can continuously extract features and make adaptive adjustments through multi-channel adaptive Fourier decomposition, dynamically update the parameters of the SIR model and the prediction model built based on the recurrent neural network, to ensure high-precision predictions in different time periods.

[0068] Among them, it can be assumed that the output of the SIR model is:

[0069] X SIR (t)=[S(t),I(t),R(t)]

[0070] Assume that the epidemic characteristics extracted by multi-channel adaptive Fourier decomposition are:

[0071] X MAFD (t)

[0072] Then at time t, the input of the SIR-RNN model can be expressed as:

[0073] X(t)=[X SIR (t),X MAFD (t)]

[0074] The hidden state update formula of RNN is:

[0075] h(t)=σ(W h ·X(t)+U h ·h(t-1)+b h )

[0076] in:

[0077] h(t) is the hidden state at time t.

[0078] W h and U h are the input weight matrix and the hidden state weight matrix respectively.

[0079] b h is the bias term.

[0080] σ is the activation function (such as tanh or ReLU)

[0081] The following describes in detail the epidemic development prediction method provided in the above embodiment by taking a plurality of different prediction tasks as examples.

[0082] In one embodiment, the prediction task is a regional COVID-19 (COVID-19 is only an example, and the present application can also be used to predict the development trend of other epidemics) epidemic prediction, such as predicting the number of new cases and hospitalization rate in a certain area in the next week. At this time, the target area is specifically a region in a designated country (the region can be any region of the country designated in advance); the standardized epidemic data includes epidemic data at multiple stages of epidemic development in the designated area; the epidemic data at each stage of epidemic development includes the number of newly confirmed cases per day, the number of hospitalizations, the utilization rate of intensive care units, the vaccination rate, the number of tests, the implementation of public health intervention measures, meteorological data, and multiple population mobility data; the epidemic characteristics specifically include the main frequency component characteristics representing the periodic changes of the epidemic and the dynamic characteristics reflecting the short-term fluctuations and trend changes of the epidemic. At this time, the epidemic development prediction results output by the target epidemic development prediction model include the number of new cases and hospitalization rate in the region in the next week.

[0083] In one embodiment, the prediction task is global epidemic monitoring and prediction, such as predicting the number of new cases and epidemic trends in the world in the next month. At this time, the target area specifically refers to multiple countries and regions in the world; the standardized epidemic data specifically includes the epidemic data of multiple epidemic waves in various countries and regions; the epidemic data of each epidemic wave includes the number of newly confirmed cases per day, mortality rate, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the periodic changes of the epidemic and the dynamic characteristics reflecting the short-term drastic changes of the epidemic. At this time, the epidemic development prediction results output by the target epidemic development prediction model include the number of new cases and epidemic trends in various countries and regions in the next month.

[0084] In the above two embodiments, combining the epidemic characteristics extracted by multi-channel adaptive Fourier decomposition and the epidemic propagation dynamic information output by the SIR model as input data of the recurrent neural network can enable the recurrent neural network to more accurately capture the epidemic dynamics and improve the accuracy and stability of the overall prediction, so that regardless of whether the prediction task is regional or global, accurate prediction of the epidemic development trend can be achieved.

[0085] In one embodiment, the prediction task is to predict the epidemic situation under the influence of vaccination, such as predicting the development of the epidemic situation in a certain country in the next two months. At this time, the target area is specifically a country (the country can be any country in the world); the standardized epidemic data includes the epidemic data of multiple epidemic development stages of the epidemic in the specified country; the epidemic data at each epidemic development stage include the number of newly confirmed cases per day, vaccination rate, vaccine type and its effectiveness, number of severe cases and deaths, implementation of public health intervention measures and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the cyclical changes of the epidemic, the dynamic characteristics reflecting the short-term drastic changes of the epidemic, and the vaccination impact characteristics reflecting the impact of vaccination on the spread of the epidemic. At this time, the epidemic development prediction results output by the target epidemic development prediction model include the development of the epidemic in the next two months in the specified country.

[0086] In this embodiment, compared with the other two embodiments, the standardized epidemic data also includes vaccination rates, vaccine types and their effectiveness, etc., which can reflect the impact of epidemic vaccination on the spread of the epidemic. Accordingly, the epidemic characteristics extracted from the standardized epidemic data through multi-channel adaptive Fourier decomposition include not only the main frequency components and fluctuation characteristics in the epidemic data of each channel, but also vaccination impact characteristics that reflect the impact of vaccination on the spread of the epidemic. The SIR model can also simulate the impact of vaccination on the spread of the epidemic, and the recurrent neural network can also learn the correlation between vaccination-related characteristics and epidemic spread, ultimately enabling the recurrent neural network to accurately predict the development trend of the epidemic under the influence of vaccination.

[0087] The present application also provides a device for predicting the development of an epidemic. In some embodiments, see Figure 2 , the device comprises:

[0088] An epidemic data acquisition module 10 is used to acquire standardized epidemic data of a target area, wherein the standardized epidemic data includes epidemic data of multiple channels;

[0089] The epidemic feature extraction module 20 is used to extract features of the epidemic data of each channel in the standardized epidemic data using multi-channel adaptive Fourier decomposition to obtain epidemic features;

[0090] The epidemic spread simulation module 30 is used to call the sensor disease spread model based on the standardized epidemic data to start simulating the basic spread dynamics of the epidemic and obtain the epidemic spread dynamic information;

[0091] The epidemic development prediction module 40 is used to input the epidemic characteristics and epidemic spread dynamic information into the target epidemic development prediction model obtained by training the recurrent neural network to obtain the epidemic development prediction result.

[0092] In some embodiments, the infectious disease transmission model is a SIR model; the epidemic transmission simulation module 30 includes:

[0093] An initial number determination unit, used to determine the initial number of susceptible persons, infected persons, and removed persons based on standardized epidemic data;

[0094] The epidemic spread simulation unit inputs the initial number of susceptible, infected and removed persons into the SIR model, so that the SIR model simulates the basic spread dynamics of the epidemic based on the preset infection rate and recovery rate, and obtains the number of susceptible, infected and removed persons that changes with time, and uses the number of susceptible, infected and removed persons that changes with time as the dynamic information of the epidemic spread.

[0095] In some embodiments, the epidemic feature extraction module 20 includes:

[0096] The decomposition unit is used to perform frequency decomposition on the epidemic data of each channel using multi-channel adaptive Fourier decomposition, extract the main frequency components and dynamic characteristics in the epidemic data of each channel, and obtain the epidemic characteristics.

[0097] In some embodiments, the epidemic data acquisition module 10 includes:

[0098] A raw data acquisition unit, used to acquire newly collected raw epidemic data of the target area;

[0099] The data processing unit is used to perform data cleaning and standardization on the original epidemic data in sequence to obtain standardized epidemic data.

[0100] In some embodiments, the target area is a designated region of a designated country; the standardized epidemic data include epidemic data at multiple stages of epidemic development in the designated region; the epidemic data at each stage of epidemic development include multiple items including the daily number of newly confirmed cases, number of hospitalizations, utilization rate of intensive care units, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the cyclical changes of the epidemic and the dynamic characteristics reflecting the short-term fluctuations and trend changes of the epidemic; the epidemic development forecast results include the number of new cases and hospitalization rate in the designated region in the next week.

[0101] In some embodiments, the target area is multiple countries and regions around the world; the standardized epidemic data includes epidemic data of multiple epidemic waves in various countries and regions; the epidemic data of each epidemic wave include multiple items including the daily number of newly confirmed cases, mortality rate, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the cyclical changes of the epidemic and the dynamic characteristics reflecting the short-term drastic changes in the epidemic; the epidemic development forecast results include the number of new cases and epidemic trends in each country and region in the next month.

[0102] In some embodiments, the target area is a designated country; the standardized epidemic data include epidemic data at multiple stages of epidemic development of the designated country; the epidemic data at each stage of epidemic development include multiple items including the daily number of newly confirmed cases, vaccination rate, vaccine types and their effectiveness, number of severe cases and deaths, implementation of public health intervention measures and population mobility data; the epidemic characteristics include the main frequency component characteristics representing the cyclical changes of the epidemic, the dynamic characteristics reflecting the short-term drastic changes in the epidemic, and the vaccination impact characteristics reflecting the impact of vaccination on the spread of the epidemic; the epidemic development forecast results include the epidemic development in the designated country in the next two months.

[0103] The epidemic development prediction device provided in one embodiment of the present application and the epidemic development prediction method provided in the present application belong to the same concept. The specific implementation process is detailed in the full text of the specification and will not be repeated here.

[0104] An embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the epidemic development prediction method provided in an embodiment of the present application.

[0105] Figure 3 A specific structural block diagram of a computer device provided in an embodiment of the present application is shown, and the computer device 100 includes: one or more processors 101, a memory 102, and one or more computer programs, wherein the processor 101 and the memory 102 are connected via a bus, the one or more computer programs are stored in the memory 102, and are configured to be executed by the one or more processors 101, and when the processor 101 executes the computer program, the epidemic development prediction method provided in an embodiment of the present application is implemented.

[0106] It should be understood that each step in each embodiment of the present application is not necessarily performed in sequence according to the order indicated by the step number. Unless there is clear explanation in this article, the execution of these steps does not have strict order restriction, and these steps can be performed in other orders. Moreover, at least a part of the steps in each embodiment can include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0107] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink), DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0108] The technical features of the above embodiments may 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.

[0109] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for predicting the development of an epidemic, characterized in that: The method comprises: Acquire standardized epidemic data of a target area, wherein the standardized epidemic data includes epidemic data of multiple channels; Using multi-channel adaptive Fourier decomposition to extract features of the epidemic data of each channel in the standardized epidemic data to obtain epidemic features; Based on the standardized epidemic data, the sensor disease transmission model is called to start simulating the basic transmission dynamics of the epidemic, and the epidemic transmission dynamics information is obtained; The epidemic characteristics and the epidemic spread dynamic information are input into the target epidemic development prediction model obtained by training the recurrent neural network to obtain the epidemic development prediction result.

2. The method according to claim 1, characterized in that The infectious disease transmission model is a SIR model; based on the standardized epidemic data, the sensor disease transmission model is called to start simulating the basic transmission dynamics of the epidemic, and the epidemic transmission dynamics information is obtained, including: Determining the initial number of susceptible, infected, and removed persons based on the standardized epidemic data; Inputting the initial number of susceptible persons, infected persons, and removed persons into the SIR model, so that the SIR model simulates the basic transmission dynamics of the epidemic based on the preset infection rate and recovery rate, and obtains the number of susceptible persons, infected persons, and removed persons that changes over time; The number of susceptible persons, infected persons and removed persons that changes with time is used as the dynamic information of the epidemic spread.

3. The method according to claim 1, characterized in that The epidemic data of each channel in the standardized epidemic data is subjected to feature extraction using multi-channel adaptive Fourier decomposition to obtain epidemic features, including: Multi-channel adaptive Fourier decomposition is used to perform frequency decomposition on the epidemic data of each channel, and the main frequency components and dynamic characteristics in the epidemic data of each channel are extracted to obtain the epidemic characteristics.

4. The method according to claim 1, characterized in that Obtain standardized epidemic data for the target area, including: Acquire newly collected original epidemic data of the target area; The original epidemic data is sequentially cleaned and standardized to obtain the standardized epidemic data.

5. The method according to any one of claims 1 to 4, characterized in that The target area is a designated region of a designated country; The standardized epidemic data include epidemic data at multiple epidemic development stages in the designated area; the epidemic data at each epidemic development stage include multiple items of the daily number of newly confirmed cases, number of hospitalizations, utilization rate of intensive care units, vaccination rate, number of tests, implementation of public health intervention measures, meteorological data, and population mobility data; The epidemic characteristics include the main frequency component characteristics indicating the periodic changes of the epidemic and the dynamic characteristics reflecting the short-term fluctuations and trend changes of the epidemic; The forecast results of the epidemic development include the number of new cases and hospitalization rate in the designated area in the next week.

6. The method according to any one of claims 1 to 4, characterized in that The target areas are multiple countries and regions around the world; The standardized epidemic data include epidemic data of multiple epidemic waves in various countries and regions; the epidemic data of each epidemic wave include multiple items of the number of newly confirmed cases per day, mortality rate, vaccination rate, number of people tested, implementation of public health intervention measures, meteorological data and population mobility data; The epidemic characteristics include the main frequency component characteristics representing the periodic changes of the epidemic and the dynamic characteristics reflecting the short-term drastic changes of the epidemic; The epidemic development forecast results include the number of new cases and epidemic trends in each country and region in the next month.

7. The method according to any one of claims 1 to 4, characterized in that The target area is a designated country; The standardized epidemic data include epidemic data at multiple epidemic development stages of the epidemic in the designated country; the epidemic data at each epidemic development stage include multiple items including the number of newly confirmed cases per day, vaccination rate, vaccine types and their effectiveness, number of severe cases and deaths, implementation of public health intervention measures and population mobility data; The epidemic characteristics include main frequency component characteristics indicating periodic changes in the epidemic, dynamic characteristics reflecting short-term dramatic changes in the epidemic, and vaccination impact characteristics reflecting the impact of vaccination on the spread of the epidemic; The epidemic development forecast results include the epidemic development in the designated country in the next two months.

8. A device for predicting the development of an epidemic, characterized in that: The device comprises: An epidemic data acquisition module, used to acquire standardized epidemic data of a target area, wherein the standardized epidemic data includes epidemic data of multiple channels; An epidemic feature extraction module, used to extract features of the epidemic data of each channel in the standardized epidemic data using multi-channel adaptive Fourier decomposition to obtain epidemic features; An epidemic spread simulation module is used to call the sensor disease spread model based on the standardized epidemic data to start simulating the basic spread dynamics of the epidemic and obtain the epidemic spread dynamic information; The epidemic development prediction module is used to input the epidemic characteristics and the epidemic spread dynamic information into the target epidemic development prediction model obtained by training the recurrent neural network to obtain the epidemic development prediction result.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the epidemic development prediction method as described in any one of claims 1 to 7 are implemented.

10. A computer device, characterized in that: include: one or more processors; Memory; as well as One or more computer programs, the processor and the memory are connected via a bus, wherein the one or more computer programs are stored in the memory and are configured to be executed by the one or more processors, wherein the processor implements the steps of the epidemic development prediction method as described in any one of claims 1 to 7 when executing the computer program.

Images