A method and system for predicting epidemic trends of infectious diseases

Abstract

The present invention discloses a method and system for predicting epidemic trends of infectious diseases. The present invention combines two models, namely a statistical distribution model variant and a transmission dynamics model. In the statistical distribution model variant, the influence of temperature on each parameter of the model is taken into account, so that the model takes into account the influence of multiple factors in calculation, which is convenient for the accuracy of later result prediction. At the same time, on the basis of preliminarily obtaining a prediction curve of the epidemic trend of infectious diseases, a weight allocation mechanism is established, and different weights are constructed according to different models to obtain prediction results. The method improves the consistency between the prediction results and the actual situation, reduces the dependence of the model on parameters, and improves the accuracy of the prediction results.

Description

Technical Field

[0001] The present invention belongs to the field of digital medical technology and relates to a method and system for predicting epidemic trends of infectious diseases. Background Art

[0002] The monitoring and early warning of infectious diseases is one of the key tasks in infectious disease prevention and control, and the prediction of epidemic trends of infectious diseases is the top priority of monitoring and early warning work.

[0003] At present, the prediction of short-term epidemic trends of infectious diseases is generally based on transmission dynamics models, such as the classic SEIR model. Transmission dynamics models have many limitations in actual combat and are often inaccurate. The reasons are that, first, the model is heavily dependent on parameter settings, and sometimes the uncertainty of parameters is large or the parameter changes cannot be expressed by functional relationships; second, the model cannot take into account all influencing factors, especially when there are hidden variables in reality, it is difficult for the model to take them into account, which reduces the accuracy of the model prediction. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the models in the prior art are seriously dependent on parameters when predicting the epidemic trends of infectious diseases. When encountering large parameter changes or parameters whose relationship cannot be expressed by functions, the prediction accuracy of the model will be affected. Secondly, when affected by hidden variables, the model cannot take their influencing factors into account, resulting in limitations in model prediction and reducing the accuracy of model prediction. A method and system for predicting the epidemic trends of infectious diseases are provided.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for predicting epidemic trends of infectious diseases, comprising the following steps:

[0007] Obtain the number of infectious diseases and establish infectious disease time series data based on the number of infectious diseases;

[0008] Establish a statistical distribution model variant and a transmission dynamics model, use the infectious disease time series data as input for the statistical distribution model variant and the transmission dynamics model, and obtain the prediction curve of the epidemic trend of the infectious disease;

[0009] The distribution ratio of the model is constructed, combined with the predicted curve of the epidemic trend of infectious diseases, and the final prediction result is obtained after normalization.

[0010] A further improvement of the present invention is:

[0011] The statistical distribution model variants include extreme value distribution model variants and gamma distribution model variants.

[0012] The extreme value distribution model variant is expressed by the following formula:

[0013]

[0014] in, y is the number of patients at time x, y0, A (T) 、p (T) 、w (T) is the undetermined parameter, where A (T) 、p (T) 、w (T) is a function of the average temperature T during the x period of time;

[0015] The gamma distribution model variant is expressed by the following formula:

[0016]

[0017] In the formula, y is the number of patients at time x, y0, A (T) , α, β (T) To be determined parameter.

[0018] The transmission dynamics model is a SEIR model, in which S represents susceptible persons, E represents latent persons, I represents exposed persons, and R represents recovered persons.

[0019] The allocation weight of the model building includes the following steps:

[0020] When the virus type is influenza virus, the ratios among the extreme value distribution model variant, the transmission dynamics model, and the gamma distribution model variant are: 0.42: 0.42: 0.16;

[0021] When the virus type is the new coronavirus, the ratio between the extreme value distribution model variant, the transmission dynamics model, and the gamma distribution model variant is: 0.45:0.40:0.15.

[0022] The final prediction result is obtained by:

[0023] Using Bayesian theorem to predict real-time propagation index R t The value includes the following steps:

[0024] Establish the distribution likelihood function of the average infection λ;

[0025] Establishing the real-time transmission index R based on the distribution likelihood function of the average infection λ t The distribution likelihood function of ;

[0026] Based on the real-time propagation index R t The distribution likelihood function is used to calculate the real-time propagation index R in a certain period of time. t The probability distribution of

[0027] Based on the real-time propagation index R in a certain period of time t The probability distribution of R in all time periods is obtained through iterative calculation. t The probability distribution of .

[0028] The real-time propagation index R t The probability distribution of is:

[0029]

[0030] In the formula, K represents the number of confirmed cases;

[0031] The above-mentioned iterative calculation obtains R in all time periods. t The probability distribution of , including:

[0032]

[0033] In the formula, t represents a certain period of time.

[0034] An infectious disease epidemic trend prediction system, comprising:

[0035] A data acquisition module is used to obtain the number of infectious diseases and establish infectious disease time series data based on the number of infectious diseases;

[0036] The model fitting module is used to establish a statistical distribution model variant and a transmission dynamics model, and use the infectious disease time series data as inputs of the statistical distribution model variant and the transmission dynamics model, respectively, to obtain a prediction curve of the epidemic trend of the infectious disease;

[0037] The result acquisition module is used to construct the distribution ratio of the model, combined with the prediction curve of the epidemic trend of infectious diseases, and obtain the final prediction result after normalization.

[0038] A terminal device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any method described in the present invention when executing the computer program.

[0039] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any method described in the present invention.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention discloses a method for predicting epidemic trends of infectious diseases. The present invention combines two models, namely a statistical distribution model variant and a transmission dynamics model. In the statistical distribution model variant, the influence of temperature on each parameter of the model is taken into account, so that the model takes into account the influence of multiple factors in calculation, which is convenient for the accuracy of subsequent result prediction. At the same time, on the basis of preliminarily obtaining a prediction curve of the epidemic trend of infectious diseases, a weight allocation mechanism is established, and different weights are constructed according to different models to obtain prediction results. The method improves the consistency between the prediction results and the actual situation, reduces the dependence of the model on parameters, and improves the accuracy of the prediction results.

[0042] Furthermore, in the present invention, the statistical distribution model variant is combined with the extreme value distribution model variant and the gamma distribution model variant to establish a parameter equation to calculate the influence of temperature on each parameter of the model, so as to facilitate accurate prediction of the results. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 is a flow chart of the method of the present invention;

[0045] Figure 2 It is a schematic diagram of the software system flow of the present invention;

[0046] Figure 3 It is a PINN network structure diagram of the present invention;

[0047] Figure 4a Display diagram for inputting disease information;

[0048] Figure 4b The interface diagram for single strain prediction and result display;

[0049] Figure 4c The interface diagram for double strain prediction and result display;

[0050] Figure 4d For R t Value evolution curve diagram;

[0051] Figure 5 This is a diagram of the implementation process and result display interface of the popular trend prediction algorithm of Example 1 of the present invention;

[0052] Figure 6 is R of Example 1 of the present invention tPrediction algorithm implementation process and result display interface diagram;

[0053] Figure 7 This is a diagram of the implementation process and result display interface of the popular trend prediction algorithm of Example 2 of the present invention;

[0054] Figure 8 is R of Example 2 of the present invention t Diagram of the prediction algorithm implementation process and result display interface. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0056] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0058] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0059] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0060] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0062] See also Figures 1 to 8 The embodiment of the present invention discloses a method for predicting the epidemic trend of infectious diseases, which solves the problem of inaccuracy of traditional infectious disease prediction methods and builds a software system for use in infectious disease monitoring and early warning work.

[0063] See also Figure 1 to Figure 2 Specifically, the method disclosed in the present invention comprises the following steps:

[0064] Step 1: Obtain the number of infectious diseases and establish infectious disease time series data based on the number of infectious diseases;

[0065] In this step, the number of infectious diseases can be obtained from the infectious disease monitoring network.

[0066] It should be noted that in this step, data acquisition is not limited to the infectious disease monitoring network, but can also be obtained from relevant records such as hospitals or disease research institutes.

[0067] Step 2: Establish a statistical distribution model variant and a transmission dynamics model, use the infectious disease time series data as input to the statistical distribution model variant and the transmission dynamics model, and obtain the prediction curve of the epidemic trend of the infectious disease;

[0068] Specifically, this step includes the following steps:

[0069] In this embodiment, the statistical distribution model variants include extreme value distribution model variants and gamma distribution model variants, specifically:

[0070] The extreme value distribution model variant is:

[0071]

[0072] in,

[0073] Specifically, in the formula, y is the number of patients at time x, y0, A (T) 、p (T) 、w (T) is the undetermined parameter, where A(T) 、p (T) 、w (T) is a function of the average temperature T over a period of time x.

[0074] The gamma distribution model variants are:

[0075]

[0076] In the formula, y is the number of patients at time x, y0, A (T) , α, β (T) To be determined parameter.

[0077] Furthermore, in the extreme value distribution variant A (T) With the gamma distribution variant A (T) equal.

[0078] Furthermore, in the extreme value distribution variant A (T) 、p (T) 、w (T) Substitute them into the following parametric equations to express them:

[0079]

[0080] p (T) =p0+m(l|T-T1|)

[0081]

[0082] Among them, T∈[-10,35], A0, k, p0, m, w0, n are all fitting parameters;

[0083] In the formula, A (T) The interpretation is:

[0084] When the disease is influenza A, Y1 = 0.64, Y2 = 0.36, T1 = 0.64, T2 = 28, σ1 = 12.5, σ2 = 3.92;

[0085] When the disease is influenza B, Y1=1, Y2=0, T1=5, σ1=20.48;

[0086] When the disease is COVID-19, Y1=1, Y2=0, T1=9, σ1=115.52.

[0087] p (T) The interpretation of the parameters is:

[0088] When the disease is influenza A, l = 1, T1 = 4, when T∈[-10,21]; l = 2.43, T1 = 28, when T∈[-21,35]

[0089] When the disease is influenza B: l = 1, T1 = 5;

[0090] When the disease is COVID-19, l=1 and T1=9.

[0091] w (T) The interpretation of the parameters is:

[0092] When the disease is influenza A, ε1 = 0.81, ε2 = 0.19, T1 = 4, T2 = 28, q1 = 22.45, q2 = 7.22;

[0093] When the disease is influenza B: ε1=1, ε2=0, T1=5, T2=52.02;

[0094] When the disease is COVID-19: ε1=1, ε2=0, T1=9, T2=605.5.

[0095] Furthermore, in the gamma distribution variant, β (T) Substitute the following parametric equations:

[0096]

[0097] Among them, T∈[-10,35], β0, p are the functions to be fitted;

[0098] In the formula, for β (T) The interpretation of the parameters is:

[0099] When the disease is influenza A, δ=1, T1=4, when T∈[-10,21]; δ=1.34, T1=28, when T∈[-21,35];

[0100] When the disease is influenza B: δ=1, T1=5;

[0101] When the disease is COVID-19: δ=1, T1=9;

[0102] It should be noted that the above parameters are only parameters of one embodiment of the present invention. Specifically, in actual infectious disease prediction, the parameters vary according to the type of infectious disease to be predicted and are not limited to the specific values ​​in this formula.

[0103] Furthermore, in this step, the transmission dynamics model is the SEIR model, where S represents susceptible persons, E represents latent persons, I represents exposed persons, and R represents recovered persons. The specific function structure is:

[0104] Susceptible person(s):

[0105]

[0106] Lurker(E):

[0107]

[0108] Patients (I):

[0109]

[0110] Recovered (R):

[0111]

[0112] Furthermore, the boundary conditions of the model are:

[0113] S (0,a) +E (0,a) +I (0,a) +R (0,a) =1

[0114] Specifically, the transmission dynamics model divides the population into four compartments, namely susceptible, latent, exposed, and recovered. The proportion of people in each compartment to the total population is affected by two states: time and age. The boundary condition is that S+E+I+R=1 when t=0. In the mathematical structure, γ (t,a) is the contact infection rate, k (t,a) For vaccine effectiveness, u (t,a) is the mortality rate, b (t) is the birth rate, σ (t,a) The ratio of recovered patients to susceptible patients, a (t,a) is the probability of latent transformation into a patient, λ (t,a) is the probability of a patient turning into a recovered patient, d (t,a) is the case fatality rate. The mathematical structure consists of four partial differential equations (PDEs) that describe the entire process of the epidemic of infectious diseases. However, this set of equations involves a large number of tensor calculations and high computing power requirements. In practical applications, it is often simplified and used in various variants to solve specific problems. When infectious diseases are susceptible to all age groups and the infection rates do not differ much, such as influenza and the new crown, age factors can be ignored, and the above PDE equations are simplified to four ordinary differential equations (ODE). If the disease has a low mortality rate and a short epidemic period, factors such as mortality rate, birth rate, and reinfection rate can be ignored to further reduce the amount of calculation. If there is no effective vaccine for the disease, the protective effect of the vaccine can be ignored.

[0115] Furthermore, the above model is used to fit the epidemic curve of infectious diseases, including the following steps:

[0116] See also Figure 3 , using physical information neural network (PINN) for processing. PINN has good fitting ability for mixed mathematical structures, good high-dimensional operations, and relatively low hardware cost for calculation. x1 and x2 are input layers, and the middle is m hidden layers with n neurons. is the output of the neural network.

[0117] Furthermore, in the fitting of the parameters of each prediction model of the present invention, the number of PINN neurons is selected in the range of 2 to 100, and the number of layers does not exceed 3. The loss function is the sum of the structural loss, boundary loss and initial value loss of the PDE, ODE or ordinary equation. The network is trained using the gradient descent algorithm to minimize the loss function. In the result display stage, the root mean square error (MSE) is used to display the degree of deviation between the prediction result and the real data.

[0118] Step 3: Construct the distribution ratio of the model, combine it with the prediction curve of the epidemic trend of infectious diseases, and obtain the final prediction result after normalization

[0119] Specifically, the present invention comprises the following steps:

[0120] In this step, the weights of different infectious diseases are allocated in different ways. This embodiment discloses an embodiment of weight allocation for predictions of two different infectious diseases, specifically including weight allocation for predictions of influenza virus and new coronavirus.

[0121] Among them, when the virus to be predicted is influenza virus, the weight distribution method of each model is:

[0122] The ratios between the extreme value distribution model variant, the propagation dynamics model, and the gamma distribution model variant are: 0.42:0.42:0.16;

[0123] When the virus to be predicted is the new coronavirus, the weight distribution of each model is as follows:

[0124] The ratios between the extreme value distribution model variants, the propagation dynamics model, and the gamma distribution model variants are: 0.45:0.40:0.15.

[0125] Furthermore, according to the prediction results of each model, the final prediction result is obtained after normalization according to the set weights. Based on the final prediction result, the real-time propagation index R is predicted using the Bayesian theorem. t value, including the following steps:

[0126] Establish the distribution likelihood function of the average infection probability λ;

[0127] The average infection rate of infectious diseases λ basically conforms to the Poisson distribution. Combining λ and R t The relationship derives R when the confirmed case is k t Distribution likelihood function, γ is the inverse of the mean disease duration;

[0128] Assuming that the number of confirmed cases in a certain time period t is only related to the number of confirmed cases in the t-1 time period, R t The number of confirmed cases in a certain period of time and the R t-1 Related, using Bayes' theorem, derive Rt Probability distribution.

[0129] Assume R t-1 Subject to R t is the normal distribution of the mean, and through iteration, R of all time periods is calculated t The probability distribution of , a is the proportional coefficient.

[0130] Specifically, the distribution likelihood function of the average infection probability λ is:

[0131]

[0132] Furthermore, λ and R t The relationship is expressed as:

[0133]

[0134] Furthermore, by substituting formula (2) into formula (1), we can obtain that when the number of confirmed cases is K, R t The distribution likelihood function of is:

[0135] L(R t |k) (2)

[0136] Furthermore, when the number of confirmed cases is K, R t The probability distribution of is:

[0137]

[0138] Further, iterate until t is 0:

[0139]

[0140] Based on the above disclosed method, the present invention also discloses two specific infectious disease development trend prediction processes. The programming software used in the embodiment is Python 3.10.9. Specifically, see FIG4 , which is an interface display diagram of the present invention. Figures 4a to 4c In , Peak, Gamma, SEIR and Combine are the extreme value distribution, gamma distribution, propagation dynamics model and the integration of the above three prediction results. Figure 4d In the above example, Critical time refers to R t The time point at which the value first becomes less than 1.

[0141] Example 1

[0142] See also Figures 5 and 6 , prediction of dual-strain co-epidemic influenza.

[0143] This embodiment uses the mixed epidemic of Victoria and influenza A H3N2 in a certain area from week 44 of 2019 to week 11 of 2020 as an example to illustrate the implementation process and demonstrate the software system. When the positive rates of Victoria and influenza A H3N2 in the infectious disease monitoring network are both close to 15%, the prediction system is started, and 9 weeks of data are available at this time. The corresponding weekly average temperature is queried through the meteorological website, and the average temperature data of each week in the future is obtained by averaging the temperature data of the past five years.

[0144] Convert the data into two columns, one for the week number and the other for the number of cases detected by the infectious disease surveillance network that week.

[0145] On the initial interface, select influenza as the disease type, 2 strains, and 35 weeks for prediction. Click Next to enter the prediction interface. Import data in the prediction interface and click Fit. The program uses the PINN algorithm in the background to calculate the extreme value distribution, gamma distribution, and SEIR model fitting parameters based on the existing 9 weeks of data. The parameters of the week where the prediction peak is located are given and displayed on the interface. According to the preset weights assigned to each algorithm, the final prediction result is obtained, as shown below: Figure 5 As shown, click to enter R t Value prediction interface; click Prediction to get R t Value prediction result, when R t When it is less than 1, it means that the infectious disease is effectively controlled. Figure 6 The parameters of the week in which each model predicts the peak, and the root mean square error (MSE) between the predicted number of cases and the actual number of cases are shown in Figure 5 .

[0146] The algorithm predicts that the total number of cases, Victoria strain, and influenza A (H3N2) strain epidemic peak times will be in the third, fourth, and first weeks of 2020, respectively. Subsequent monitoring data from the infectious disease monitoring network showed that the total number of cases, Victoria strain, and influenza A (H3N2) strain epidemic peak times for this round of influenza were all in the second to third weeks of 2020. t The forecast value shows that the epidemic level of the disease began to decline from the fourth week of 2020. The forecast results are consistent with the actual results.

[0147] Example 2

[0148] See also Figures 7 and 8 , single strain COVID-19 epidemic prediction.

[0149] The implementation example uses the COVID-19 epidemic in a certain region from April to July 2023 as an example to illustrate the implementation process and demonstrate the software system. This prediction system is started when the positive rate of the infectious disease monitoring network is close to 15%, and there are 7 weeks of data at this time. The corresponding weekly average temperature is queried through the meteorological website, and the average temperature data of each week in the future is obtained by averaging the temperature data of the past five years.

[0150] Convert the data into two columns, one for the week number and the other for the number of cases detected by the infectious disease surveillance network that week.

[0151] On the initial interface, select the predicted disease type as COVID-19, the number of strains as 1, and the number of predicted weeks as 40. Click Next to enter the prediction interface. Import data in the prediction interface and click Fit. The program uses the PINN algorithm in the background to calculate the extreme value distribution, gamma distribution, and SEIR model fitting related parameters based on the existing 7 weeks of data. The parameters of the week where the predicted peak is located are given and displayed on the interface. According to the preset weights assigned to each algorithm, the final prediction result is obtained, as shown below: Figure 7 As shown, click to enter R t Value prediction interface; click Prediction to get R t Value prediction result, when R t When it is less than 1, it means that the infectious disease is effectively controlled. Figure 8 The parameters of the week in which each model predicts the peak, and the root mean square error (MSE) between the predicted number of cases and the actual number of cases are shown in Figure 7 .

[0152] The algorithm predicts that the peak will be in weeks 21 to 23 of 2023. Subsequent monitoring data from the infectious disease monitoring network showed that the peak platform of the new crown is in weeks 20 to 22 of 2023. t The forecast value shows that the prevalence of the disease has been on a downward trend since week 23 of year 23. The forecast results are consistent with the actual results.

[0153] This embodiment also discloses a system for predicting epidemic trends of infectious diseases, including:

[0154] A data acquisition module is used to obtain the number of infectious diseases and establish infectious disease time series data based on the number of infectious diseases;

[0155] The model fitting module is used to establish a statistical distribution model variant and a transmission dynamics model, and use the infectious disease time series data as inputs of the statistical distribution model variant and the transmission dynamics model, respectively, to obtain a prediction curve of the epidemic trend of the infectious disease;

[0156] The result acquisition module is used to construct the distribution ratio of the model, combine the prediction curve of the epidemic trend of infectious diseases, and obtain the final prediction result after normalization.

[0157] A schematic diagram of a terminal device provided in an embodiment of the present invention. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0158] The computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to accomplish the present invention.

[0159] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0160] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0161] The memory may be used to store the computer programs and / or modules, and the processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory.

[0162] If the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0163] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting epidemic trends of infectious diseases, characterized in that: The following steps are involved: Obtain the number of infectious diseases and establish infectious disease time series data based on the number of infectious diseases; Establish a statistical distribution model variant and a transmission dynamics model, use the infectious disease time series data as input for the statistical distribution model variant and the transmission dynamics model, and obtain the prediction curve of the epidemic trend of the infectious disease; The statistical distribution model variants include extreme value distribution model variants and gamma distribution model variants; The extreme value distribution model variant is expressed by the following formula: in, y is the number of patients at time x, y0, A (T) 、p (T) 、w (T) is the undetermined parameter, where A (T) 、p (T) 、w (T) is a function of the average temperature T during the x period of time; The gamma distribution model variant is expressed by the following formula: In the formula, y is the number of patients at time x, y0, A (T) , α, β (T) is a parameter to be determined; The distribution ratio of the model is constructed, combined with the predicted curve of the epidemic trend of infectious diseases, and the final prediction result is obtained after normalization.

2. A method for predicting epidemic trends of infectious diseases according to claim 1, characterized in that: The transmission dynamics model is a SEIR model, in which S represents susceptible persons, E represents latent persons, I represents exposed persons, and R represents recovered persons.

3. A method for predicting epidemic trends of infectious diseases according to claim 2, characterized in that: The allocation weight of the model building includes the following steps: When the virus type is influenza virus, the ratios among the extreme value distribution model variant, the transmission dynamics model, and the gamma distribution model variant are: 0.42: 0.42: 0.16; When the virus type is the new coronavirus, the ratio between the extreme value distribution model variant, the transmission dynamics model, and the gamma distribution model variant is: 0.45:0.40:0.

15.

4. A method for predicting epidemic trends of infectious diseases according to claim 3, characterized in that: The final prediction result is obtained by: Using Bayesian theorem to predict real-time propagation index R t The value includes the following steps: Establish the distribution likelihood function of the average infection λ; Establishing the real-time transmission index R based on the distribution likelihood function of the average infection λ t The distribution likelihood function of ; Based on the real-time propagation index R t The distribution likelihood function is used to calculate the real-time propagation index R in a certain period of time. t The probability distribution of Based on the real-time propagation index R in a certain period of time t The probability distribution of R in all time periods is obtained through iterative calculation. t The probability distribution of .

5. A method for predicting epidemic trends of infectious diseases according to claim 4, characterized in that: The real-time propagation index R t The probability distribution of is: In the formula, K represents the number of confirmed cases; The above-mentioned iterative calculation obtains R in all time periods. t The probability distribution of , including: In the formula, t represents a certain period of time.

6. A system for predicting epidemic trends of infectious diseases, characterized in that: include: A data acquisition module is used to obtain the number of infectious diseases and establish infectious disease time series data based on the number of infectious diseases; The model fitting module is used to establish a statistical distribution model variant and a transmission dynamics model, and use the infectious disease time series data as inputs of the statistical distribution model variant and the transmission dynamics model, respectively, to obtain a prediction curve of the epidemic trend of the infectious disease; The result acquisition module is used to construct the distribution ratio of the model, combined with the prediction curve of the epidemic trend of infectious diseases, and obtain the final prediction result after normalization.

7. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. 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 method according to any one of claims 1 to 5 are implemented.

Images