Disease data prediction method, system and device and storage medium
By constructing multiple fit models and determining weights, combining time series models and fusion models, the problem of inaccurate prevalence prediction is solved, and higher prediction accuracy and stability are achieved.
Patent Information
- Application Number
- CN202510053322.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-14
- Publication Date
- 2025-08-29
AI Technical Summary
The existing prediction results of prevalence and number of patients are not very accurate, and there is a lack of effective multi-factor considerations.
Multiple fitting models are constructed, historical population index data are fitted through preset time series models, the degree of fit is verified using the initial prevalence data, model weights are determined, and a fusion model is constructed to improve prediction accuracy.
By fusing multiple fitting models, the accuracy of future disease data is improved, and multiple factors are fully considered, reducing the uncertainty and deviation of predictions.
Smart Images
Figure CN120565098A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a disease data prediction method, system, device and storage medium. Background Art
[0002] Data on future prevalence and number of patients in a population is a key factor in public health development and can also provide effective support for disease prevention and investment decisions. Existing predictions of prevalence and number of patients often rely on single models, resulting in inaccurate results. Summary of the Invention
[0003] The embodiments of the present application provide a disease data prediction method, system, device and storage medium to solve the problem of inaccurate disease data prediction results in the prior art.
[0004] In a first aspect, the present application provides a disease data prediction method, comprising:
[0005] Obtaining historical population index data and historical prevalence data of a target population, and constructing a plurality of fitting models according to the historical population index data and the historical prevalence data;
[0006] Fitting the historical population index data through a preset time series model to obtain predicted population index data;
[0007] Predicting and obtaining a plurality of initial morbidity data according to the predicted population index data and the plurality of fitting models;
[0008] Verifying the degree of fit of each fitting model using the initial prevalence data, and determining the weight of each fitting model according to the degree of fit;
[0009] A fusion model is constructed according to the plurality of fitting models and the weight corresponding to each fitting model, and the predicted population index data is input into the fusion model to obtain predicted morbidity data.
[0010] In one embodiment, obtaining historical population index data of a target population includes:
[0011] Obtain the historical per capita income, historical average years of education, and historical low-age fertility rate of the target population;
[0012] The historical per capita income, historical average years of education and historical low-age fertility rate are normalized to obtain the historical population index data.
[0013] In one embodiment, the normalized formula is:
[0014]
[0015] Among them, X i The historical per capita income, historical average years of education and historical low fertility rate of the target population are worst. i The best value is the historical highest value corresponding to the historical per capita income, historical average years of education and historical low-age fertility rate of the target population. i The historical lowest values corresponding to the historical per capita income, historical average years of education and historical low-age fertility rate of the target population;
[0016] The calculation formula for the historical population index data is:
[0017]
[0018] SDI is the historical population index data.
[0019] In one embodiment, the fitting model includes a first fitting model, and the first fitting model includes:
[0020] E1[logit(Y l,a,s,y )]=β1SDI l,y +α l,a,s ,
[0021] Among them, E1[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, sex s and year y, β1 is the fixed coefficient corresponding to the historical population index data of the target location-age-sex (l,a,s), α l,a,s is the intercept of target location-age-sex (l,a,s).
[0022] In one embodiment, the fitting model includes a second fitting model, and the second fitting model includes:
[0023] E2[logit(Y l,a,s,y )]=β2SDI l,a,y +α l,a,s
[0024] Among them, E2[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, sex s and year y, β2 is the fixed coefficient corresponding to the historical population index data of the target location-year (l,y), α l,a,s is the intercept of target location-age-sex (l,a,s).
[0025] In one embodiment, the fitting model includes a third fitting model, and the third fitting model includes:
[0026] E3[logit(Y l,a,s,y )]=β3SDI l,y +α l,a,s
[0027] Among them, E3[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence Y{l,a,s,y} for each location l, sex s, and year y, β3 is the fixed coefficient corresponding to the historical population index data of the target location-sex (l,s), α l,a,s is the intercept of target location-age-sex (l,a,s).
[0028] In one embodiment, determining the weight of each fitting model according to the verification result includes:
[0029] Obtaining the degree of fit between the actual disease data corresponding to each fitting model and the initial disease rate data, and calculating the posterior probability according to the degree of fit;
[0030] The posterior probabilities are weighted averaged to obtain the weight of each fitting model.
[0031] In a second aspect, the present application provides a disease data prediction system, comprising:
[0032] An acquisition module is used to acquire historical population index data and historical morbidity data of a target population, and to construct a plurality of fitting models according to the historical population index data and the historical morbidity data;
[0033] A fitting module, configured to fit the historical population index data using a preset time series model to obtain predicted population index data;
[0034] A prediction module, configured to predict and obtain a plurality of initial morbidity data based on the predicted population index data and the plurality of fitting models;
[0035] A verification module, configured to verify the degree of fit of each fitting model using the initial prevalence data, and determine the weight of each fitting model according to the degree of fit;
[0036] A fusion module is used to construct a fusion model based on the multiple fitting models and the weight corresponding to each fitting model, and input the predicted population index data into the fusion model to obtain predicted morbidity data.
[0037] In a third aspect, the present application provides a computer device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the disease data prediction method described in the first aspect.
[0038] In a fourth aspect, the present application provides a computer storage medium storing a computer program, which, when executed on a processor, implements the disease data prediction method according to the first aspect.
[0039] The embodiments of the present application have the following beneficial effects:
[0040] The disease data prediction method provided in the embodiment of the present application obtains a fusion model by fusing multiple fitting models, which fully considers various factors when predicting future disease data, thereby improving the accuracy of future disease data. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] To more clearly illustrate the technical solution of this application, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of this application and should not be considered as limiting the scope of protection of this application. Those skilled in the art can also derive other relevant drawings based on these drawings without inventive effort.
[0042] Figure 1 A schematic diagram of a process for predicting disease data provided by an embodiment of the present application is shown;
[0043] Figure 2 A schematic diagram of a process for obtaining historical population index data provided by an embodiment of the present application is shown;
[0044] Figure 3 A schematic diagram of the framework structure of a disease data prediction system provided in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0045] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments.
[0046] The components of the embodiments of the present application generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but rather merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.
[0047] Hereinafter, the terms "including", "having" and their cognates, which may be used in various embodiments of the present application, are intended only to indicate target features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or adding the possibility of one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items.
[0048] Furthermore, the terms “first,” “second,” “third,” etc., are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.
[0049] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by those skilled in the art to which the various embodiments of the present application belong. The terms (such as those defined in generally used dictionaries) will be interpreted as having the same meaning as in the context of the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning unless clearly defined in the various embodiments of the present application.
[0050] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.
[0051] Reference Figure 1 , Figure 1 This is a flow chart of a disease data prediction method provided in this embodiment, which includes:
[0052] S101. Obtain historical population index data and historical morbidity data of a target population, and construct a plurality of fitting models according to the historical population index data and the historical morbidity data.
[0053] The target population can be divided according to regions and areas. For example, if you want to predict the disease data of a certain area or region in the future, you can first obtain the historical population index data and historical disease rate data of the area or region, and then build a corresponding fitting model based on the correspondence between the historical population index data and the historical disease rate data.
[0054] For example, if we want to predict future disease rates in Region A from 2025 to 2050, we can first obtain historical population index data and historical disease prevalence data for Region A from 2010 to 2020. The prevalence rate can be for a specific disease type, such as Parkinson's disease. Then, based on the corresponding relationship between Region A's historical population index data and the historical Parkinson's disease prevalence data, we can construct a fitting model.
[0055] A fitting model is constructed based on the relationship between the prevalence data and the target population data. Each fitting model can consider different factors, so that each fitting model can predict a target prevalence data.
[0056] For example, for region A, you can use location, gender, and year as the main factors to build a model to predict the prevalence data. You can also use location, gender, age, and year as the main factors to build another model to predict the prevalence data, and so on, thus obtaining multiple fitting models.
[0057] S102. Fitting the historical population index data through a preset time series model to obtain predicted population index data.
[0058] The preset time series model can be an ARIMA model. The preset time series model ARIMA can automatically fit past trends and predict future values. Therefore, the historical population index data of region A from 2010 to 2020 can be integrated into the form of a time series, and the ARIMA model can be trained, and then used to predict the population data from 2025 to 2050.
[0059] S103. Predict and obtain a plurality of initial morbidity data based on the predicted population index data and the plurality of fitting models.
[0060] Since the fitting model mainly reflects the correspondence between population data and prevalence data, after obtaining the predicted population index data through the preset time series model, the predicted population index data can be input into each fitting model so that each fitting model outputs a corresponding initial prevalence data.
[0061] S104: Verify the degree of fit of each fitting model using the initial prevalence data, and determine the weight of each fitting model according to the degree of fit.
[0062] In order to make the model prediction more accurate, we can first verify the degree of fit of each fitting model and determine the weight of each fitting model based on the degree of fit.
[0063] For example, if you want to predict the prevalence data for 2025, you can first predict the prevalence data for 2024 by fitting the model, and then use the actual prevalence data for 2024 and the predicted prevalence data for 2024 to verify the degree of fit of each fitting model.
[0064] S105 , constructing a fusion model based on the multiple fitting models and the weight corresponding to each fitting model, inputting the predicted population index data into the fusion model to obtain predicted morbidity data.
[0065] A fusion model is constructed based on each fitting model and its corresponding weight to improve the stability and scientificity of the model, and then the fusion model is used to predict the predicted prevalence data.
[0066] This embodiment obtains a fusion model by fusing multiple fitting models, which fully considers various factors when predicting future disease data, thereby improving the accuracy of future disease data.
[0067] Reference Figure 2 , step S101 includes: steps S1011-S1012.
[0068] S1011. Obtain the historical per capita income, historical average years of education and historical low-age fertility rate of the target population.
[0069] S1012. Normalize the historical per capita income, historical average years of education, and historical low-age fertility rate to obtain the historical population index data.
[0070] Historical population index data can be reflected through the Socio-demographic Index (SDI). Compared with the traditional GDP, the SDI can more comprehensively reflect the level of progress in all aspects of society, including information on education, health, employment, environmental protection, and social security.
[0071] SDI is a summation of the following three key indicators:
[0072] 1. Per capita income (income level adjusted for lagged distribution): "Lag-distributed per capita income" (LDI) is usually used to represent the economic level.
[0073] 2. Average years of education: refers to the average years of education for people aged 15 and above, used to measure the level of education accessibility and quality.
[0074] 3. Low-age fertility rate: that is, the total fertility rate of people under 25 years old (TFR under age 25), which is used to reflect changes in population birth rate and fertility patterns.
[0075] Normalizing per capita income, average years of education, and low-age fertility rate yields SDI.
[0076] Among them, the normalized formula is:
[0077]
[0078] Among them, X iThe historical per capita income, historical average years of education and historical low fertility rate of the target population are worst. i The best value is the historical highest value corresponding to the historical per capita income, historical average years of education and historical low-age fertility rate of the target population. i The historical lowest values corresponding to the historical per capita income, historical average years of education and historical low-age fertility rate of the target population;
[0079] The calculation formula for the historical population index data is:
[0080]
[0081] SDI is the historical population index data.
[0082] This embodiment uses the social development index as an indicator of population data, which is more representative than other indicators, thereby making the predicted disease data more accurate.
[0083] In order to improve the prediction accuracy of the fusion model, this embodiment lists six fusion models, namely the first fitting model, the second fitting model, the third fitting model, the fourth fitting model, the fifth fitting model and the sixth fitting model. The six fusion models are used to predict the disease data from different perspectives.
[0084] In one embodiment, a first fitting model is used to predict disease data based on region, age, and gender factors, where the first fitting model includes:
[0085] E1[logit(Y l,a,s,y )]=β1SDI l,y +α l,a,s ,
[0086] Among them, E1[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, sex s and year y, β1 is the fixed coefficient corresponding to the historical population index data of the target location-age-sex (l,a,s), α l,a,s is the intercept of target location-age-sex (l,a,s).
[0087] In one embodiment, a second fitting model is used to predict disease data for regional and year factors, where the second fitting model includes:
[0088] E2[logit(Y l,a,s,y )]=β2SDI l,a,y +α l,a,s
[0089] Among them, E2[logit(Yl,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, sex s and year y, β2 is the fixed coefficient corresponding to the historical population index data of the target location-year (l,y), α l,a,s is the intercept of target location-age-sex (l,a,s).
[0090] In one embodiment, a third fitting model is used to predict disease data for regional and gender factors, and the third fitting model includes:
[0091] E3[logit(Y l,a,s,y )]=β3SDIl,y+α l,a,s
[0092] Among them, E3[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence Y{l,a,s,y} for each location l, sex s, and year y, β3 is the fixed coefficient corresponding to the historical population index data of the target location-sex (l,s), α l,a,s is the intercept of target location-age-sex (l,a,s).
[0093] In one embodiment, a fourth fitting model is used to predict disease data based on region, age, and gender factors. The fourth fitting model includes:
[0094]
[0095] Among them, E4[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, age a, sex s and year y, β4 is the fixed coefficient corresponding to the predicted SDI, α l,a,s is the site-age-sex specific random intercept.
[0096] In one embodiment, a fifth fitting model is used to predict disease data for regional and gender factors, the fifth fitting model comprising:
[0097]
[0098] E5[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, age a, sex s and year y, β5 is the fixed coefficient corresponding to the predicted SDI, and α l,a,sFinally, to account for trends that cannot be explained by the covariates, a random walk model (autoregressive integrated moving average model, ARIMA(0,1,0)) is applied to the residuals of the regression model, and the predictions of these residuals are incorporated into the final estimated forecasts.
[0099] In one embodiment, a sixth fitting model is used to predict disease data based on region, age, and gender factors. The sixth fitting model includes:
[0100] E6[logit(Y l,a,s,y )]=β6SDI l,a,y +α l,a,s
[0101] E6[logit(Y l,a,s,y )] is the logit-transformed prevalence estimate Y{l,a,s,y} for each location l, age a, sex s, and year y, β6 is the fixed coefficient corresponding to the SDI for a specific location-age-sex (l,y), and α l,a,s is the intercept for the specific location-age-sex (l, a, s). Finally, to account for trends that cannot be explained by the covariates, a random walk model (autoregressive integrated moving average model, ARIMA(0,1,0)) is applied to the residuals of the regression model, and the predictions of these residuals are incorporated into the final estimated forecasts.
[0102] In one embodiment, determining the weight of each fitting model according to the verification result includes:
[0103] Obtaining the degree of fit between the actual disease data corresponding to each fitting model and the initial disease rate data, and calculating the posterior probability according to the degree of fit;
[0104] The posterior probabilities are weighted averaged to obtain the weight of each fitting model.
[0105] In this embodiment, the prediction results of the six fitting models may be weighted and integrated by using Bayesian Model Averaging (BMA) to reduce the uncertainty of the model.
[0106] Specifically, the posterior probability is calculated for each fitted model based on its fit to the historical data and its predictive performance. The predictions from these six fitted models are then weighted averaged according to their respective posterior probabilities. The resulting forecast combines the advantages of each model while reducing the bias and overfitting issues that may arise from a single model. In this way, BMA can more comprehensively utilize available information, thereby improving the stability and reliability of forecasts.
[0107] In order to evaluate the performance of BMA, this application uses data from 1990 to 2010 to predict data from 2011 to 2021, and verifies the method of this application based on the comparison results. To quantify the prediction performance, the predicted values of BMA are compared with the retained data, and the root mean square error (RMSE) and bias are calculated. The bias is calculated by subtracting the median of the difference between all predicted values and the actual observed values for each age, gender, year and location. From the results, compared with the optimal model, the prediction accuracy of the integrated BMA is improved by about 20%.
[0108] Reference Figure 3 The present application also provides a disease data prediction system 300, comprising:
[0109] An acquisition module 301 is configured to acquire historical population index data and historical prevalence data of a target population, and to construct a plurality of fitting models based on the historical population index data and the historical prevalence data;
[0110] A fitting module 302 is configured to fit the historical population index data using a preset time series model to obtain predicted population index data;
[0111] A prediction module 303 is configured to predict and obtain a plurality of initial morbidity data based on the predicted population index data and the plurality of fitting models;
[0112] A verification module 304 is configured to verify the degree of fit of each fitting model using the initial prevalence data, and determine a weight of each fitting model according to the degree of fit;
[0113] The fusion module 305 is used to construct a fusion model based on the multiple fitting models and the weight corresponding to each fitting model, input the predicted population index data into the fusion model, and obtain predicted morbidity data.
[0114] It can be understood that the system of this embodiment corresponds to the disease data prediction method of the above embodiment, and the optional options in the above embodiment are also applicable to this embodiment, so they will not be repeated here.
[0115] The present application also provides a computer device. Exemplarily, the computer device includes a processor and a memory, wherein the memory stores a computer program, and the processor runs the computer program to enable the computer device to execute the functions of the various modules in the above-mentioned disease data prediction method or the above-mentioned disease data prediction system.
[0116] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including a central processing unit (CPU), a graphics processing unit (GPU) and a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or at least one of other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application.
[0117] The memory may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. The memory is used to store a computer program, and the processor may execute the computer program accordingly after receiving an execution instruction.
[0118] The present application also provides a computer storage medium for storing the computer program used in the above-mentioned computer device. The computer storage medium may be a readable storage medium, a non-volatile storage medium, or a volatile storage medium. For example, the computer storage medium may include, but is not limited to, various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0119] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and structure diagrams in the accompanying drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in an alternative implementation, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the structure diagram and / or flowchart, and the combination of boxes in the structure diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions.
[0120] In addition, the functional modules or units in the various embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.
[0121] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a smart phone, personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.
[0122] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.
Claims
1. A disease data prediction method, characterized in that: include: Obtaining historical population index data and historical prevalence data of a target population, and constructing a plurality of fitting models according to the historical population index data and the historical prevalence data; Fitting the historical population index data through a preset time series model to obtain predicted population index data; Predicting and obtaining a plurality of initial morbidity data according to the predicted population index data and the plurality of fitting models; Verifying the degree of fit of each fitting model using the initial prevalence data, and determining the weight of each fitting model according to the degree of fit; A fusion model is constructed according to the plurality of fitting models and the weight corresponding to each fitting model, and the predicted population index data is input into the fusion model to obtain predicted morbidity data.
2. The disease data prediction method according to claim 1, characterized in that: The acquisition of historical population index data of the target population includes: Obtain the historical per capita income, historical average years of education, and historical low-age fertility rate of the target population; The historical per capita income, historical average years of education and historical low-age fertility rate are normalized to obtain the historical population index data.
3. The disease data prediction method according to claim 2, characterized in that: The normalized formula is: Among them, X i The historical per capita income, historical average years of education and historical low fertility rate of the target population are worst. i The best value is the historical highest value corresponding to the historical per capita income, historical average years of education and historical low-age fertility rate of the target population. i The historical lowest values corresponding to the historical per capita income, historical average years of education and historical low-age fertility rate of the target population; The calculation formula for the historical population index data is: SDI is the historical population index data.
4. The disease data prediction method according to claim 1, characterized in that: The fitting model includes a first fitting model, and the first fitting model includes: E1[logit(Y l,a,s,y )]=β1SDI l,y +α l,a,s , Among them, E1[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence value Y{l,a,s,y} for each location l, sex s and year y, β1 is the fixed coefficient corresponding to the historical population index data of the target location-age-sex (l,a,s), α l,a,s is the intercept of target location-age-sex (l,a,s).
5. The disease data prediction method according to claim 1, characterized in that: The fitting model includes a second fitting model, and the second fitting model includes: E2[logit(Y l,a,s,y )]=β2SDI l,a,y +α l,a,s Among them, E2[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence Y{l,a,s,y} for each location l, sex s, and year y, β2 is the fixed coefficient corresponding to the historical population index data of the target location-year (l,y), and α l,a,s is the intercept of target location-age-sex (l,a,s).
6. The disease data prediction method according to claim 1, characterized in that: The fitting model includes a third fitting model, and the third fitting model includes: E3[logit(Y l,a,s,y )]=β3SDI l,y +α l,a,s Among them, E3[logit(Y l,a,s,y )] is the logit transformation of the estimated prevalence Y{l,a,s,y} for each location l, sex s, and year y, β3 is the fixed coefficient corresponding to the historical population index data of the target location-sex (l,s), α l,a,s is the intercept of target location-age-sex (l,a,s).
7. The disease data prediction method according to claim 1, characterized in that: Determining the weight of each fitting model according to the fitting degree includes: Obtaining the degree of fit between the actual disease data corresponding to each fitting model and the initial disease rate data, and calculating the posterior probability according to the degree of fit; The posterior probabilities are weighted averaged to obtain the weight of each fitting model.
8. A disease data prediction system, characterized in that: include: An acquisition module is used to acquire historical population index data and historical morbidity data of a target population, and to construct a plurality of fitting models according to the historical population index data and the historical morbidity data; A fitting module, configured to fit the historical population index data using a preset time series model to obtain predicted population index data; A prediction module, configured to predict and obtain a plurality of initial morbidity data based on the predicted population index data and the plurality of fitting models; A verification module, configured to verify the degree of fit of each fitting model using the initial prevalence data, and determine the weight of each fitting model according to the degree of fit; A fusion module is used to construct a fusion model based on the multiple fitting models and the weight corresponding to each fitting model, and input the predicted population index data into the fusion model to obtain predicted morbidity data.
9. A computer device, characterized in that: The computer device includes a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the disease data prediction method according to any one of claims 1 to 7.
10. A computer storage medium, characterized in that The device stores a computer program, which, when executed on a processor, implements the disease data prediction method according to any one of claims 1 to 7.