El Nino index-based avian influenza occurrence risk prediction model construction method and system

Abstract

The invention discloses an El Nino index-based avian influenza occurrence risk prediction model construction method and system, and belongs to the technical field of risk prediction, and the method comprises the steps: respectively constructing an H5 subtype avian influenza event data set, a multivariate El Nino index data set and a climate-avian influenza data set; constructing a regression model of the avian influenza event number to the climate index based on the climate-avian influenza data set; constructing a plurality of deep learning models based on the multivariate El Nino index data set, and selecting an optimal prediction model; and predicting the number of future avian influenza events based on the regression model and the optimal prediction model. According to the method, MEI is used as a core representative variable of an El Nino climatic factor, historical epidemiological characteristics and time factors of avian influenza viruses are combined, and the influence of climatic changes on transmission of highly pathogenic avian influenza is quantitatively analyzed, so that the occurrence risk of avian influenza under future climatic conditions is dynamically predicted, and the risk of occurrence of avian influenza is predicted. And the high-risk space-time distribution of the H5 subtype strain can be accurately predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of risk prediction, and more particularly to a method and system for constructing an avian influenza risk prediction model based on the El Nino index. Background Art

[0002] Avian influenza is an animal infectious disease that is dangerous to the poultry industry, the ecological environment and human health. Among them, the H5 subtype avian influenza virus is one of the most concerned subtypes in the world. This subtype covers a number of highly pathogenic and low pathogenic strains. Its prevalence not only threatens the healthy development of animal husbandry, but also may pose a major challenge to global public health security. In recent years, the epidemic pattern of the H5 subtype avian influenza virus has shown a high degree of complexity, especially in certain regions and seasons, its transmission trajectory and outbreak frequency are closely related to climate conditions.

[0003] Global climate change, especially the climate anomaly caused by El Niño, is considered to be one of the important driving forces affecting the spread of avian influenza virus and host behavior. As a global climate anomaly event, the multi-faceted impacts of El Niño on the climate system and ecological environment have been widely studied. There is increasing evidence that El Niño may play an important role in the outbreak and spread of avian influenza by affecting climate conditions (such as temperature, humidity and precipitation). Therefore, it is necessary to select an indicator that can fully reflect the intensity and duration of El Niño to quantify its impact.

[0004] The El Niño phenomenon will significantly change the global meteorological conditions such as temperature, precipitation, and humidity. These climate changes may not only affect the survival and transmission ability of the H5 subtype avian influenza virus in the environment, but also drive the virus transmission dynamics by changing indirect factors such as migratory bird migration routes, poultry breeding environments, and ecosystem balance. In particular, the epidemic pattern of the H5 highly pathogenic subtype strain is more susceptible to the climate anomalies caused by the El Niño phenomenon due to its high pathogenicity, wide host spectrum, and sensitivity to environmental conditions. However, there is still a gap in the research on the risk prediction of the H5 subtype avian influenza based on the MEI index, and there is an urgent need to develop more accurate prediction tools.

[0005] Therefore, how to provide a method and system for constructing an avian influenza risk prediction model that can accurately predict the H5 subtype strain is an urgent problem that technicians in this field need to solve. Summary of the invention

[0006] In view of this, the present invention provides a method and system for constructing an avian influenza risk prediction model based on the El Niño index, taking H5 subtype avian influenza events as the research object, using MEI as the core representative variable of the El Niño climate factor, combining the historical epidemiological characteristics and time factors of avian influenza viruses, and quantitatively analyzing the impact of climate change on the spread of highly pathogenic avian influenza, to achieve dynamic prediction of the risk of avian influenza under future climate conditions, and to accurately predict the high-risk spatiotemporal distribution of H5 subtype strains.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A method for constructing an avian influenza risk prediction model based on the El Nino index comprises the following steps:

[0009] Respectively constructing an H5 subtype avian influenza event data set and a multivariate El Niño index data set, and performing data processing on the H5 subtype avian influenza event data set and the multivariate El Niño index data set to obtain a climate-avian influenza data set;

[0010] Constructing a regression model of the number of avian influenza events versus the climate index based on the climate-avian influenza dataset;

[0011] Building multiple deep learning models based on the multivariate El Niño index data set, evaluating the multiple deep learning models, and obtaining an optimal prediction model for future climate change indices;

[0012] The number of future avian influenza events is predicted based on the regression model and the optimal prediction model of the future climate change index.

[0013] Preferably, a regression model of the number of avian influenza events versus the climate index is constructed based on the climate-avian influenza data set, comprising:

[0014] determining a plurality of influencing factors affecting the number of avian influenza events based on the climate-avian influenza data set;

[0015] The multiple influencing factors are randomly combined as independent variables, and the number of avian influenza events is used as the dependent variable to construct multiple generalized additive models;

[0016] Calculate the AIC values ​​of the multiple generalized additive models according to the AIC criterion, and select the generalized additive model with the smallest AIC value as the optimal model;

[0017] The parameters of the optimal model were optimized by using smoothing method and approximate method, and a regression model of the number of avian influenza events and climate index was obtained.

[0018] Preferably, the multiple deep learning models constructed based on the multivariate El Niño index data set include: multi-layer perceptron, convolutional neural network, recurrent neural network, long short-term memory network, convolutional neural network-long short-term memory network deep learning model.

[0019] Preferably, the evaluation indicators for evaluating the multiple deep learning models include: mean absolute error, mean absolute percentage error and root mean square error.

[0020] Preferably, predicting the number of future avian influenza events based on the regression model and the optimal prediction model of the future climate change index includes:

[0021] Predicting climate change in a future period based on the optimal prediction model for the future climate change index;

[0022] The climate change in the future time period is added as an independent variable into the regression model to obtain a predicted value of the number of avian influenza events in the future time period.

[0023] On the other hand, the present invention provides a system for constructing an avian influenza risk prediction model based on the El Nino index, comprising:

[0024] A data set construction module is used to construct an H5 subtype avian influenza event data set and a multivariate El Niño index data set respectively, and perform data processing on the H5 subtype avian influenza event data set and the multivariate El Niño index data set to obtain a climate-avian influenza data set;

[0025] A regression model building module, used for building a regression model of the number of avian influenza events versus the climate index based on the climate-avian influenza data set;

[0026] A prediction model building module, used to build multiple deep learning models based on the multivariate El Niño index data set, evaluate the multiple deep learning models, and obtain the optimal prediction model for future climate change index;

[0027] A prediction module is used to predict the number of future avian influenza events based on the regression model and the optimal prediction model of the future climate change index.

[0028] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method and system for constructing an avian influenza risk prediction model based on the El Nino index. By integrating the H5 subtype avian influenza event data set and the multivariate El Nino index data set, the present invention can comprehensively consider the impact of climate change on avian influenza events, thereby constructing a more accurate prediction model. In particular, by constructing multiple deep learning models and evaluating them, the optimal model is selected to predict the future climate change index, further improving the prediction accuracy of the number of future avian influenza events. When constructing a regression model of the number of avian influenza events to the climate index, the present invention adopts a generalized additive model, and by randomly combining influencing factors and selecting the optimal model, the robustness and adaptability of the model are enhanced. This method can cope with complex and changeable climatic conditions and improve the prediction performance of the model under different environments. By introducing deep learning technologies, such as multi-layer perceptrons, convolutional neural networks, recurrent neural networks, etc., the present invention can efficiently process and analyze a large amount of climate data and avian influenza event data, thereby quickly constructing a prediction model. This greatly improves the efficiency of prediction, enables the model to respond to climate change more quickly, and provides timely and effective support for avian influenza prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0030] Figure 1 A flow chart of a model building method is provided for the present invention;

[0031] Figure 2 modeling workflows for statistics;

[0032] Figure 3 The RNN model predicts the MEI value for the next three months;

[0033] Figure 4 The statistical regression model is used to predict the number of highly pathogenic avian influenza events in the next three months;

[0034] Figure 5 A framework diagram of a model building system is provided for the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0036] The embodiment of the present invention discloses a method for constructing an avian influenza risk prediction model based on the El Nino index, such as Figure 1 As shown, the following steps are included:

[0037] The H5 subtype avian influenza event dataset and the multivariate El Niño index dataset were constructed respectively, and the H5 subtype avian influenza event dataset and the multivariate El Niño index dataset were processed to obtain the climate-avian influenza dataset.

[0038] Specifically, the constructed global H5 subtype avian influenza dataset contains the location (continent, country, city) and discovery date (accurate to year, month, and day) of each avian influenza event. Then, an El Niño dataset represented by the multivariate El Niño index (MEI) is constructed, including each specific value corresponding to the average value of the past two months. R language version 4.3.2 is used to clean and filter invalid data, and the dates of the two datasets are aligned and unified.

[0039] The specific operations for constructing the H5 subtype avian influenza event dataset are as follows: filter the original data frame, leaving the "Animal.type", "Region", and "Observation.date..dd.mm.yyyy." columns, and only leave the Domestic and Wild classes in the "Animal.type" column, and then change the specific date in the "Observation.date..dd.mm.yyyy." column to a date format with months as the unit, and delete blank data.

[0040] The specific operations for constructing the MEI dataset are as follows: Use the pivot_longer function in the tidyr package (version 1.3.1) to convert the original data frame from wide format to long format, and then modify the date name at the same time, changing the column names such as DJ (the average of December and January), JF, FM...ON, ND, DJ, etc. to the full name of the month represented by the previous letter.

[0041] The month and year are combined to be consistent with the date column format of the H5 subtype avian influenza event dataset. Finally, the dates of the two datasets are aligned and combined into a new dataset for backup.

[0042] A regression model of the number of avian influenza events on climate index was constructed based on the climate-avian influenza dataset;

[0043] Based on the multivariate El Niño index dataset, multiple deep learning models are constructed and evaluated to obtain the optimal prediction model for future climate change indices;

[0044] Predict the number of future avian influenza events based on regression model and optimal prediction model of future climate change index.

[0045] Furthermore, according to the distribution of the number of avian influenza events, by comparing the effects of various statistical models (such as Poisson regression and pseudo-Poisson regression models, negative binomial generalized linear models and generalized additive models, etc.), the generalized additive model (Generalized Additive Models, GAM) was finally selected as the main model framework (expressed as Y = β 0 +f 1 (X 1 )+f 2 (X 2 )+...+f p (X p )+∈), a regression model of the number of avian influenza events on the climate index is constructed based on the climate-avian influenza dataset, including:

[0046] Identify multiple factors that influence the number of avian influenza events based on the climate-avian influenza dataset;

[0047] Multiple influencing factors were randomly combined as independent variables, and the number of avian influenza events was used as the dependent variable to construct multiple generalized additive models;

[0048] According to the AIC criterion, the AIC values ​​of multiple generalized additive models are calculated respectively, and the generalized additive model with the smallest AIC value is selected as the optimal model;

[0049] Use smoothing method and approximate method to optimize the parameters of the optimal model (refer to Figure 2 The method comparison part) was used to improve the prediction accuracy and generalization of the model and obtain a regression model of the number of avian influenza events on the climate index.

[0050] Furthermore, multiple deep learning models built based on the multivariate El Niño index dataset include: multi-layer perceptron, convolutional neural network, recurrent neural network, long short-term memory network, convolutional neural network-long short-term memory network deep learning model. Based on the TensorFlow framework in Python 3.10.14, multi-layer perceptron, convolutional neural network, recurrent neural network, long short-term memory network, convolutional neural network-long short-term memory network deep learning model is implemented to predict the MEI index.

[0051] First, the cleaned time series data is converted into a supervised learning format so that it can be used as training data for input features (X) and output labels (Y). By constructing the historical and future data of the time series, the model input features and output labels are effectively separated, supporting single-variable, multi-variable, and single-step or multi-step prediction tasks, automatically handling time step misalignment and missing value problems, ensuring data integrity, and being suitable for common deep learning models to provide standardized data input for time series analysis models. Then define the sliding window; separate the input features and output labels to facilitate model training; divide the training set and validation set in a ratio of 8:2 to ensure the generalization ability of the model; normalize the data to enhance the stability of model training.

[0052] The deep learning models MLP, CNN, RNN, LSTM, and CNN-LSTM were constructed respectively. Each model was trained 200 times, and the prediction time step nout was set to 3.

[0053] MLP uses a fully connected layer (Dense) as the main component. The number of neurons in the first layer is 100, and the ReLU activation function is used to introduce nonlinearity. The input dimension is the product of the time step and the number of features. The number of neurons in the second layer is 60, and the activation function is not set. The default is linear activation. The number of neurons in the output layer is 3.

[0054] The CNN model first needs to adjust the shape of the input data to adapt it to the input requirements; the one-dimensional convolution layer is used to extract local patterns in the time series, using 64 convolution kernels with a convolution kernel size of 2, and using the ReLU activation function to increase the nonlinear expression ability of the model; the maximum pooling layer is used for dimensionality reduction and key feature extraction, and the pooling window size is set to 2; the flattening layer converts the pooled two-dimensional feature map into a one-dimensional vector for input to the output layer; the first layer of the output layer is set to 50 neurons, and the second layer is set to 3 neurons.

[0055] The input layer of RNN is set to 50 neurons and the output layer is set to 3 neurons.

[0056] There are 50 LSTM units in the first and second LSTM layers, and 3 LSTM units in the output layer.

[0057] The CNN-LSTM model first reshapes the data into a four-dimensional array, divides the time series data into two subsequences, and processes each subsequence independently; the local pattern features of the subsequences are extracted through the convolution layer of CNN, and 64 convolution kernels of size 1 are set; the features are then reduced in dimension through the pooling layer and flattened into a one-dimensional vector; the flattened subsequence features are input into the LSTM layer containing 50 units, and 3 neurons are set in the output layer.

[0058] By comparing the model evaluation index loss function (loss) of the validation set (the loss function in the present invention is set to mean square error Mean Square Error, MSE), the optimal model is RNN. Then, a RNN model for time series prediction was constructed. First, the training set, validation set and test set were divided into a ratio of 8:1:1, and a layer of RNN layer containing 50 units was added to the model, and then an output layer was added, and the number of output neurons was set to 3, which generated the final prediction results. MSE was selected as the loss function during model compilation, and the Adam optimizer was used to adjust the model parameters, and the mean absolute error (MAE), mean absolute percentage error (MAPE) and root mean square error (RMSE) were used to evaluate the performance of the model. During the training process, the model received input data strictly in chronological order, and the data was not shuffled to maintain the integrity of the time series. The training parameters were set to 200 iterations, and 32 data were processed each time. By monitoring the performance of the model in real time on the validation set, it was ensured that the model could fully capture the feature changes in the time series and achieve a reasonable prediction of the MEI value for the next 3 months.

[0059] like Figure 3 As shown in the figure, the deep learning model RNN predicts the MEI values ​​for the next three months with good results; the predicted values ​​for the next three months are: 0.9 in December 2023, 0.9 in January 2024, and 0.8 in February 2024.

[0060] Going further, the evaluation metrics used to evaluate multiple deep learning models include: mean absolute error, mean absolute percentage error, and root mean square error.

[0061] Furthermore, the optimal prediction model based on the regression model and future climate change index predicts the number of future avian influenza events, including:

[0062] Predict climate change in future time periods based on the optimal prediction model of future climate change index;

[0063] Climate change in the future time period is added as an independent variable into the regression model to obtain the predicted value of the number of avian influenza events in the future time period.

[0064] like Figure 4 As shown, the three-month MEI forecast values ​​were put into the constructed and optimized statistical regression model to predict the number of highly pathogenic avian influenza events in the next three months, and the corresponding three-month forecast values ​​were: 14.520938, 12.017535, and 9.857997.

[0065] On the other hand, the present invention provides a system for constructing an avian influenza risk prediction model based on the El Nino index, such as Figure 5 As shown, including:

[0066] The data set construction module is used to construct the H5 subtype avian influenza event data set and the multivariate El Niño index data set respectively, and perform data processing on the H5 subtype avian influenza event data set and the multivariate El Niño index data set to obtain the climate-avian influenza data set;

[0067] A regression model building module is used to build a regression model of the number of avian influenza events against the climate index based on the climate-avian influenza dataset;

[0068] The prediction model building module is used to build multiple deep learning models based on the multivariate El Niño index dataset, evaluate multiple deep learning models, and obtain the optimal prediction model for future climate change indices;

[0069] The prediction module is used to predict the number of future avian influenza events based on the regression model and the optimal prediction model of the future climate change index.

[0070] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0071] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing an avian influenza risk prediction model based on the El Nino index, characterized in that: The following steps are involved: Respectively constructing an H5 subtype avian influenza event data set and a multivariate El Niño index data set, and performing data processing on the H5 subtype avian influenza event data set and the multivariate El Niño index data set to obtain a climate-avian influenza data set; Constructing a regression model of the number of avian influenza events versus the climate index based on the climate-avian influenza dataset; Building multiple deep learning models based on the multivariate El Niño index data set, evaluating the multiple deep learning models, and obtaining an optimal prediction model for future climate change indices; The number of future avian influenza events is predicted based on the regression model and the optimal prediction model of the future climate change index.

2. The method for constructing an avian influenza risk prediction model based on the El Nino index according to claim 1, characterized in that: A regression model of the number of avian influenza events versus the climate index is constructed based on the climate-avian influenza dataset, including: determining a plurality of influencing factors affecting the number of avian influenza events based on the climate-avian influenza data set; The multiple influencing factors are randomly combined as independent variables, and the number of avian influenza events is used as the dependent variable to construct multiple generalized additive models; Calculate the AIC values ​​of the multiple generalized additive models according to the AIC criterion, and select the generalized additive model with the smallest AIC value as the optimal model; The parameters of the optimal model were optimized by using smoothing method and approximate method, and a regression model of the number of avian influenza events and climate index was obtained.

3. The method for constructing an avian influenza risk prediction model based on the El Nino index according to claim 1, characterized in that: The multiple deep learning models constructed based on the multivariate El Niño index data set include: multi-layer perceptron, convolutional neural network, recurrent neural network, long short-term memory network, convolutional neural network-long short-term memory network deep learning model.

4. The method for constructing an avian influenza risk prediction model based on the El Nino index according to claim 3, characterized in that: The evaluation indicators for evaluating the multiple deep learning models include: mean absolute error, mean absolute percentage error and root mean square error.

5. The method for constructing an avian influenza risk prediction model based on the El Nino index according to claim 1, characterized in that: Predicting the number of future avian influenza events based on the regression model and the future climate change index optimal prediction model includes: Predicting climate change in a future period based on the optimal prediction model for the future climate change index; The climate change in the future time period is added as an independent variable into the regression model to obtain a predicted value of the number of avian influenza events in the future time period.

6. A system for constructing an avian influenza risk prediction model based on the El Nino index, characterized in that: include: A data set construction module is used to construct an H5 subtype avian influenza event data set and a multivariate El Niño index data set respectively, and perform data processing on the H5 subtype avian influenza event data set and the multivariate El Niño index data set to obtain a climate-avian influenza data set; A regression model building module, used for building a regression model of the number of avian influenza events versus the climate index based on the climate-avian influenza data set; A prediction model building module, used to build multiple deep learning models based on the multivariate El Niño index data set, evaluate the multiple deep learning models, and obtain the optimal prediction model for future climate change index; A prediction module is used to predict the number of future avian influenza events based on the regression model and the optimal prediction model of the future climate change index.

Images