Sewage infectious disease monitoring and early warning method and system, electronic equipment and medium

Abstract

The invention discloses a sewage infectious disease monitoring and early warning method and system, electronic equipment and a medium, and the method comprises the steps: obtaining first external monitoring data and first clinical case time series data corresponding to to-be-predicted sewage, and obtaining a target prediction model comprising a first target prediction branch and a second target prediction branch; decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component; inputting the first low-frequency component and the first external monitoring data into a first target prediction branch in a target prediction model to obtain a predicted first number of infected people; inputting the first high-frequency component and the first external monitoring data into a second target prediction branch in the target prediction model to obtain a predicted second number of infected people; determining a prediction curve slope based on the first infected person and the second infected person; and performing sewage infectious disease monitoring and early warning according to the first infected person number, the second infected person number and the prediction curve slope. The sewage infectious disease monitoring accuracy can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of infectious disease monitoring, in particular to a sewage infectious disease monitoring and early warning method and system, an electronic device and a medium. BACKGROUND

[0002] In the existing infectious disease monitoring, in addition to the reporting of relevant cases in the clinic, the monitoring of virus concentration in sewage is also generally carried out in various places, in order to more comprehensively assess the situation of infectious diseases, and to discover potential infectious disease risks as early as possible through various monitoring and early warning methods. However, most of the current infectious disease prediction models are mainly established for the development of a single infectious disease, and lack early warning models for infectious diseases in the context of routine monitoring, making it difficult to adapt to the needs of continuous monitoring. Sewage epidemiology is particularly suitable for establishing early warning models because it can reflect the health status of the population and the trend of virus transmission. However, at the present stage, there are still relatively few virus infectious disease early warning models based on sewage epidemiology theory.

[0003] In addition, there are still many challenges in the practice of sewage infectious disease monitoring that have not been solved, which affects the effectiveness of the prediction model. For example, different preprocessing methods have a huge impact on the quantitative detection of virus RNA concentration; meteorological factors can affect the detected virus RNA concentration; and changes in population density also have an important impact on virus concentration. Moreover, due to the very complex mechanism of changes in the data of the normalized monitoring of infectious diseases, a single branch model is difficult to learn both long-term trends and short-term fluctuations, resulting in poor prediction accuracy of the model, especially in some important evaluation indicators such as mean absolute percentage error (MAPE), which lack effective evaluation.

[0004] Therefore, the existing related methods have low accuracy in sewage infectious disease monitoring. SUMMARY

[0005] The present application aims to provide a sewage infectious disease monitoring and early warning method, system, electronic device and medium, which can improve the accuracy of sewage infectious disease monitoring.

[0006] In a first aspect, the embodiments of the present application provide a sewage infectious disease monitoring and early warning method, which comprises:

[0007] obtaining first external monitoring data and first clinical case time series data corresponding to the sewage to be predicted, and obtaining a target prediction model comprising a first target prediction branch and a second target prediction branch;

[0008] decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component;

[0009] inputting the first low-frequency component and the first external monitoring data into a first target prediction branch in the target prediction model to obtain a predicted first infected population;

[0010] inputting the first high-frequency component and the first external monitoring data into a second target prediction branch in the target prediction model to obtain a predicted second infected population;

[0011] determining a prediction curve slope based on the first infected population and the second infected population;

[0012] performing sewage infectious disease monitoring and early warning according to the first infected population, the second infected population, and the prediction curve slope.

[0013] Compared with the prior art, the first aspect of the present application has the following beneficial effects:

[0014] The method obtains first external monitoring data and first clinical case time series data corresponding to a to-be-predicted sewage, and obtains a target prediction model including a first target prediction branch and a second target prediction branch. The first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component. The first low-frequency component and the first external monitoring data are input into the first target prediction branch in the target prediction model to obtain a predicted first infected population. The first high-frequency component and the first external monitoring data are input into the second target prediction branch in the target prediction model to obtain a predicted second infected population. A prediction curve slope is determined based on the first infected population and the second infected population. Sewage infectious disease monitoring and early warning are performed according to the first infected population, the second infected population, and the prediction curve slope. In this way, the double-branch structure model is used to predict the infected population of the low-frequency component and the high-frequency component, which can improve the accuracy of the prediction result. The infected population and the prediction curve slope are used together for real-time hierarchical early warning, which can improve the accuracy and response speed of sewage infectious disease monitoring and early warning.

[0015] In some embodiments, the target prediction model is obtained by training in the following manner:

[0016] obtaining second external monitoring data and second clinical case time series data;

[0017] decomposing the second clinical case time series data into a second low-frequency component and a second high-frequency component;

[0018] training the prediction model using the second external monitoring data, the second low-frequency component, and the second high-frequency component to obtain a target prediction model, wherein the first prediction branch is trained using the second low-frequency component and the second external monitoring data to obtain a first target prediction branch, and the second prediction branch is trained using the second high-frequency component and the second external monitoring data to obtain a second target prediction branch.

[0019] In some embodiments, the training the prediction model based on the second external monitoring data, the second low-frequency component and the second high-frequency component to obtain a target prediction model comprises:

[0020] constructing a target loss function of the prediction model;

[0021] training the prediction model based on the target loss function and the second external monitoring data, the second low-frequency component and the second high-frequency component to obtain a target prediction model.

[0022] In some embodiments, the constructing a target loss function of the prediction model comprises:

[0023]

[0024] wherein, denotes the target loss function, N denotes the total number of sewage monitoring sites, w i denotes the time weight of the dynamic sewage monitoring site, λ1 denotes the importance proportional factor of controlling the mean square error of the first prediction branch, denotes the prediction result of the first prediction branch at the t time step, y i denotes the actual number of infected people at the i-th sewage monitoring site at the t time step, λ2 denotes the importance proportional factor of controlling the mean square error of the second prediction branch, denotes the prediction result of the second prediction branch at the t time step, p i denotes the penalty function at the t time step.

[0025] In some embodiments, the decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component comprises:

[0026] decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component by using fast Fourier transform.

[0027] In some embodiments, the determining a prediction curve slope based on the first number of infected people and the second number of infected people comprises:

[0028] dynamically weighting and fusing the first number of infected people and the second number of infected people to obtain a total predicted number of infected people;

[0029] determining a prediction curve slope based on the total predicted number of infected people and the time corresponding to the sewage to be predicted.

[0030] In some embodiments, the sewage-borne infectious disease monitoring and early warning according to the first infected person number, the second infected person number and the prediction curve slope comprises:

[0031] a first threshold value and a second threshold value are preset;

[0032] the first infected person number and the second infected person number are dynamically weighted and fused to obtain a total predicted infected person number;

[0033] the total predicted infected person number and the first threshold value are compared to obtain a first comparison result;

[0034] the prediction curve slope and the second threshold value are compared to obtain a second comparison result;

[0035] the sewage-borne infectious disease monitoring and early warning is performed according to the first comparison result and the second comparison result.

[0036] In a second aspect, the embodiments of the present application further provide a sewage-borne infectious disease monitoring and early warning system, which comprises:

[0037] a data acquisition unit, configured to acquire first external monitoring data and first clinical case time series data corresponding to sewage to be predicted, and acquire a target prediction model comprising a first target prediction branch and a second target prediction branch;

[0038] a data decomposition unit, configured to decompose the first clinical case time series data into a first low-frequency component and a first high-frequency component;

[0039] a first prediction unit, configured to input the first low-frequency component and the first external monitoring data into the first target prediction branch in the target prediction model to obtain a predicted first infected person number;

[0040] a second prediction unit, configured to input the first high-frequency component and the first external monitoring data into the second target prediction branch in the target prediction model to obtain a predicted second infected person number;

[0041] a curve slope determination unit, configured to determine a prediction curve slope based on the first infected person number and the second infected person number;

[0042] a monitoring and early warning unit, configured to perform sewage-borne infectious disease monitoring and early warning according to the first infected person number, the second infected person number and the prediction curve slope.

[0043] In a third aspect, the embodiments of the present application further provide an electronic device, comprising at least one control processor and a memory connected with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to perform the sewage infectious disease monitoring and early warning method.

[0044] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores computer executable instructions for causing a computer to perform the sewage infectious disease monitoring and early warning method.

[0045] It can be understood that the beneficial effects of the above-mentioned second aspect to fourth aspect compared with the related art are the same as the beneficial effects of the above-mentioned first aspect compared with the related art, which can be referred to the related description in the first aspect and will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0046] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the appended drawings, wherein:

[0047] Figure 1 is a flowchart of an embodiment of the sewage infectious disease monitoring and early warning method provided by the present application;

[0048] Figure 2 is a schematic diagram of the overall process of the method in the best embodiment of the sewage infectious disease monitoring and early warning method provided by the present application;

[0049] Figure 3 is a schematic diagram of the function structure of a single machine heterogeneous server in the best embodiment of the sewage infectious disease monitoring and early warning method provided by the present application;

[0050] Figure 4 is a schematic diagram of the structure of an embodiment of the sewage infectious disease monitoring and early warning system provided by the present application;

[0051] Figure 5 is a schematic diagram of the structure of an embodiment of the electronic device provided by the present application. DETAILED DESCRIPTION

[0052] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the drawings, wherein the same or similar notations represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present application, and cannot be understood as a limitation of the present application.

[0053] In the description of the present application, if there is a description to the first, second, etc. is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features or the order of the indicated technical features.

[0054] In the description of the present application, it should be understood that the orientation description, such as the orientation or position relationship indicated by up, down, etc. is based on the orientation or position relationship shown in the drawings, only for the purpose of facilitating the description of the present application and simplifying the description, and is not intended to indicate or imply that the device or element indicated must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application.

[0055] In the description of the present application, it should be noted that, unless otherwise explicitly limited, the words such as setting, installing, connecting, etc. should be broadly understood, and the person skilled in the art can reasonably determine the specific meaning of the above words in the present application in combination with the specific content of the technical solution.

[0056] Firstly, the several terms involved in the present application are analyzed:

[0057] Fast Fourier Transform (FFT): It is a mathematical method for converting signals from time domain to frequency domain. Fast Fourier Transform (FFT) is a fast algorithm for Discrete Fourier Transform (DFT), which can significantly improve the calculation efficiency. FFT is commonly used in signal processing, image processing and polynomial multiplication, etc.

[0058] Bidirectional Long Short-Term Memory Network (BiLSTM): It is a deep learning model that has shown excellent performance in natural language processing (NLP), speech recognition and image processing, etc. BiLSTM is a combination of two long short-term memory networks (LSTM networks), one of which processes the forward input sequence and the other processes the reverse input sequence to capture the bidirectional dependencies in sequence data.

[0059] Residual Network (ResNet): It is a deep learning model designed to solve the problem of gradient vanishing and degradation in deep neural networks. By introducing skip connections, residual network allows information to flow more effectively in the network, enabling the network to train deeper layers, even more than 100 layers. Its core idea is to regard the mapping between input and output as an identity mapping, simplifying the training process of deep network. The success of residual network has made significant progress in image recognition and other fields of deep learning.

[0060] In the existing infectious disease monitoring, in addition to the clinical reporting of related cases, sewage virus concentration monitoring is also generally carried out in various places, in order to more comprehensively evaluate the situation of infectious diseases, and to discover potential infectious disease risks as early as possible through various monitoring and early warning methods. However, most of the current infectious disease prediction models are mainly established for the development of a single infectious disease, and lack early warning models for infectious diseases in the context of routine monitoring, making it difficult to meet the needs of continuous monitoring. Sewage epidemiology is particularly suitable for establishing early warning models because it can reflect the health status of the population and the trend of virus transmission. However, at the present stage, there are still relatively few virus infectious disease early warning models based on sewage epidemiology theory.

[0061] In addition, there are still many challenges in the practice of sewage infectious disease monitoring that have not been solved, which affects the effectiveness of the prediction model. For example, different preprocessing methods have a huge impact on the quantitative detection of viral RNA concentration; meteorological factors can affect the detected viral RNA concentration; and changes in population density also have an important impact on viral concentration. Moreover, due to the complex mechanism of changes in data from the normalized monitoring of infectious diseases, a single branch model is difficult to learn both long-term trends and short-term fluctuations, resulting in poor prediction accuracy of the model, especially in terms of some important evaluation indicators such as mean absolute percentage error (MAPE).

[0062] To solve the problem of low accuracy of existing related methods for sewage infectious disease monitoring, the present application provides a sewage infectious disease monitoring and early warning method, system, electronic device and medium.

[0063] With reference to Figure 1 The sewage infectious disease monitoring and early warning method provided by the embodiments of the present application. The sewage infectious disease monitoring and early warning method is applied to an electronic device, which can be a server or a mobile terminal, etc. As Figure 1 shown, the sewage infectious disease monitoring and early warning method can include the following steps:

[0064] Step S100, acquiring first external monitoring data and first clinical case time series data corresponding to the sewage to be predicted, and acquiring a target prediction model comprising a first target prediction branch and a second target prediction branch;

[0065] Step S200, decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component;

[0066] Step S300, inputting the first low-frequency component and the first external monitoring data into the first target prediction branch in the target prediction model to obtain a predicted first number of infections;

[0067] Step S400, input the first high-frequency component and the first external monitoring data into a second target prediction branch in the target prediction model to obtain a predicted second infected population;

[0068] Step S500, determine a prediction curve slope based on the first infected population and the second infected population;

[0069] Step S600, conduct sewage infectious disease monitoring and early warning according to the first infected population, the second infected population, and the prediction curve slope.

[0070] In the embodiment, the first external monitoring data and the first clinical case time series data corresponding to the sewage to be predicted are obtained, and a target prediction model containing a first target prediction branch and a second target prediction branch is obtained. The first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component. The first low-frequency component and the first external monitoring data are input into the first target prediction branch in the target prediction model to obtain a predicted first infected population. The first high-frequency component and the first external monitoring data are input into the second target prediction branch in the target prediction model to obtain a predicted second infected population. A prediction curve slope is determined based on the first infected population and the second infected population. Sewage infectious disease monitoring and early warning are conducted according to the first infected population, the second infected population, and the prediction curve slope. In this way, the double-branch structure model is used to predict the infected population of the low-frequency component and the high-frequency component respectively, which can improve the accuracy of the prediction result. The real-time hierarchical early warning is conducted by using the infected population and the prediction curve slope together, which can improve the accuracy and response speed of the sewage infectious disease monitoring and early warning.

[0071] The above-mentioned target prediction model containing a first target prediction branch and a second target prediction branch can be constructed by using a bidirectional long short-term memory network (BiLSTM) to construct the first prediction branch. The first prediction branch is trained to obtain the first target prediction branch. A residual network (ResNet) can be used to construct the second prediction branch. The second prediction branch is trained to obtain the second target prediction branch. The first target prediction branch and the second target prediction branch are constructed into the target prediction model.

[0072] The above-mentioned determination of the prediction curve slope based on the first infected population and the second infected population can be the calculation of the prediction curve slope according to the infected population. Since the prediction curve slope reflects the rate of change of the case number (i.e. the infected population) with time, the prediction curve slope can be calculated after the infected population is known.

[0073] The above-mentioned first external monitoring data can include sewage inflow information data (such as ammonia nitrogen, pH value, water temperature, etc. collected by online physicochemical sensors), virus RNA concentration monitoring data of sewage infectious diseases, and meteorological data (such as air temperature, precipitation, and relative humidity at the same timestamp).

[0074] In some embodiments, the target prediction model is trained by the following way:

[0075] obtaining second external monitoring data and second clinical case time series data;

[0076] decomposing the second clinical case time series data into a second low-frequency component and a second high-frequency component;

[0077] training the prediction model by using the second external monitoring data, the second low-frequency component and the second high-frequency component to obtain a target prediction model, wherein the first prediction branch is trained by using the second low-frequency component and the second external monitoring data to obtain a first target prediction branch, and the second prediction branch is trained by using the second high-frequency component and the second external monitoring data to obtain a second target prediction branch.

[0078] In the present embodiment, the target prediction model is obtained by obtaining second external monitoring data and second clinical case time series data; decomposing the second clinical case time series data into a second low-frequency component and a second high-frequency component; and training the prediction model by using the second external monitoring data, the second low-frequency component and the second high-frequency component to obtain a target prediction model, wherein the first prediction branch is trained by using the second low-frequency component and the second external monitoring data to obtain a first target prediction branch, and the second prediction branch is trained by using the second high-frequency component and the second external monitoring data to obtain a second target prediction branch. In this way, since the change of the clinical case time series data has a very complex mechanism, it is difficult for a single branch model to learn long-term trends and short-term fluctuations at the same time. Therefore, by decomposing the second clinical case time series data into a second low-frequency component and a second high-frequency component, and then training two prediction branches respectively, the target prediction model obtained after training can be more accurate.

[0079] The above-mentioned second external monitoring data can include sewage inflow information data (such as ammonia nitrogen, pH value, water temperature, etc. collected by online physicochemical sensors) for training the prediction model, sewage infectious disease virus RNA concentration monitoring data, and meteorological data (such as air temperature, precipitation, and relative humidity at the same timestamp).

[0080] The above-mentioned training of the prediction model by using the second external monitoring data, the second low-frequency component and the second high-frequency component to obtain a target prediction model can be dividing the overall data set containing the second external monitoring data, the second low-frequency component and the second high-frequency component into a training data set, a validation data set and a test data set, then training the prediction model by using the training data set, verifying the trained prediction model by using the validation data set, and finally testing the verified prediction model by using the test data set to obtain the final trained prediction model, i.e. the target prediction model.

[0081] In some embodiments, the second external monitoring data, the second low-frequency component and the second high-frequency component are used to train the prediction model to obtain a target prediction model, including:

[0082] constructing a target loss function of the prediction model;

[0083] Based on the target loss function, the second external monitoring data, the second low-frequency component and the second high-frequency component are used to train the prediction model to obtain a target prediction model.

[0084] In the present embodiment, based on the constructed target loss function, the second external monitoring data, the second low-frequency component and the second high-frequency component are used to train the prediction model to obtain a target prediction model, which can make the target prediction model have more accurate prediction results.

[0085] In some embodiments, the target loss function of the prediction model includes:

[0086]

[0087] wherein, denotes the target loss function, N denotes the total number of sewage monitoring sites, w i (t) denotes the time weight of the dynamic sewage monitoring site, λ1 denotes the importance proportional factor of controlling the mean square error of the first prediction branch, denotes the prediction result of the first prediction branch at time step t, y i (t) denotes the actual number of infected people at the i-th sewage monitoring site at time step t, λ2 denotes the importance proportional factor of controlling the mean square error of the second prediction branch, denotes the prediction result of the second prediction branch at time step t, p i (t) denotes the penalty function at time step t.

[0088] In the present embodiment, considering that the present embodiment adopts multi-site real-time monitoring and a double-branch frequency domain decomposition architecture, and introduces auxiliary external features and a hierarchical early warning mechanism, in order to fully exploit the collaborative optimization potential between multi-modal and multi-task, a multi-dimensional adaptive weighted target loss function is designed. Based on the target loss function, the prediction model is trained, which can improve the accuracy of the prediction model.

[0089] In some embodiments, the first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component, including:

[0090] The first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component by using fast Fourier transform.

[0091] In the embodiment, the first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component by using fast Fourier transform, thereby laying a good data foundation for more accurate model prediction in the later stage.

[0092] In some embodiments, the prediction curve slope is determined based on the first infected person number and the second infected person number, including:

[0093] The first infected person number and the second infected person number are dynamically weighted and fused to obtain a total predicted infected person number.

[0094] The prediction curve slope is determined based on the total predicted infected person number and the time corresponding to the sewage to be predicted.

[0095] In the embodiment, the first infected person number and the second infected person number are dynamically weighted and fused to obtain a total predicted infected person number, and the prediction curve slope is determined based on the total predicted infected person number and the time corresponding to the sewage to be predicted. In this way, the infected person numbers predicted by the first target prediction branch and the second target prediction branch are dynamically weighted and fused to obtain more accurate prediction results, and then the correct prediction curve slope is determined by comparing the correct infected person number results, thereby laying a good data foundation for improving the accuracy of sewage infectious disease monitoring and early warning in the later stage.

[0096] In some embodiments, the sewage infectious disease monitoring and early warning is performed according to the first infected person number, the second infected person number, and the prediction curve slope, including:

[0097] The first threshold and the second threshold are preset.

[0098] The first infected person number and the second infected person number are dynamically weighted and fused to obtain a total predicted infected person number.

[0099] The total predicted infected person number and the first threshold are compared to obtain a first comparison result.

[0100] The prediction curve slope and the second threshold are compared to obtain a second comparison result.

[0101] The sewage infectious disease monitoring and early warning is performed according to the first comparison result and the second comparison result.

[0102] In the embodiment, the first threshold and the second threshold are preset, the first infected person number and the second infected person number are dynamically weighted and fused to obtain a total predicted infected person number, the total predicted infected person number and the first threshold are compared to obtain a first comparison result, the prediction curve slope and the second threshold are compared to obtain a second comparison result, and the sewage infectious disease monitoring and early warning is performed according to the first comparison result and the second comparison result. In this way, the infected person number and the prediction curve slope are combined to perform real-time hierarchical early warning, thereby improving the accuracy of sewage infectious disease monitoring and early warning.

[0103] To facilitate the understanding of those skilled in the art, a set of best embodiments is provided below:

[0104] In the existing infectious disease monitoring, in addition to the clinical reporting of related cases, sewage virus concentration monitoring is also generally carried out in various places, in order to more comprehensively evaluate the situation of infectious diseases, and to discover potential infectious disease risks as early as possible through various monitoring and early warning methods. However, most of the current infectious disease prediction models are mainly established for the development of a single infectious disease, and lack of early warning models for infectious diseases in the context of daily monitoring, making it difficult to meet the needs of continuous monitoring. Sewage epidemiology is particularly suitable for establishing early warning models because it can reflect the health status of the population and the trend of virus transmission. However, at the present stage, there are still few virus infectious disease early warning models based on sewage epidemiology theory, which need to be developed urgently.

[0105] In addition, there are still many challenges in the practice of sewage infectious disease monitoring that have not been solved, which affects the effectiveness of the prediction model. For example, different preprocessing methods have a huge impact on the quantitative detection of virus RNA concentration; meteorological factors can affect the detected virus RNA concentration; and changes in population density also have an important impact on virus concentration. Moreover, due to the complex mechanism of changes in the data of the normalized monitoring of infectious diseases, a single branch model is difficult to learn both long-term trends and short-term fluctuations, resulting in poor prediction accuracy of the model, especially in some important evaluation indicators such as mean absolute percentage error (MAPE), which lack effective evaluation.

[0106] Based on the above considerations, the present embodiment develops a double-branch deep prediction model based on multi-site data based on the theory of sewage epidemiology, and conducts daily monitoring and early warning of infectious diseases through classification tasks and regression tasks. This method not only considers various influencing factors comprehensively, but also improves the accuracy and response speed of monitoring, providing scientific support for public health management. In addition, this method will provide practical examples for the widespread application of sewage epidemiology in infectious disease monitoring and promote in-depth related research.

[0107] The multi-site infectious disease monitoring and early warning method based on the double-branch deep network and the theory of sewage epidemiology proposed in the present embodiment includes the following contents:

[0108] 1. The overall process of the method of the present embodiment.

[0109] The data set used for infectious disease monitoring and early warning in the present embodiment is derived from multiple scenarios:

[0110] (1) Sewage inflow information data set;

[0111] (2) Sewage infectious disease virus RNA concentration monitoring data set;

[0112] (3) weather data set;

[0113] (4) clinical case monitoring data set of infectious diseases.

[0114] Firstly, the system deploys online physicochemical sensors and automatic sampling machines at the influent inlet of multiple wastewater treatment plant monitoring sites. The influent wastewater information such as ammonia nitrogen, pH value and water temperature is collected by online physicochemical sensors, and composite water samples are collected by automatic sampling machines at a frequency of 100 mL per hour. In the laboratory, pretreatment of wastewater samples and quantitative detection of viral RNA concentration of infectious diseases are completed. At the same time, the external characteristics such as air temperature, precipitation, relative humidity and clinical case monitoring data of infectious diseases at the same time stamp are also written into the database synchronously.

[0115] The embodiment first performs one-dimensional fast Fourier transform on the clinical case monitoring time series (i.e. clinical case time series data). The method of the embodiment divides the obtained frequency spectrum coefficients into low frequency (LFI) and high frequency (HFI) parts according to the energy proportion, with a sliding window of 4-16 weeks, the former retains the trend information, and the latter mainly contains mutations and noises. Subsequently, the dual-branch deep prediction stage is entered, the method of the embodiment concatenates the LFI and the aligned external characteristics and sends them into the BiLSTM for mean prediction; at the same time, the HFI enters the lightweight ResNet network for residual prediction. The final prediction sequence is obtained after the combination of the two. The prediction value is immediately sent to the decision module, the system first establishes the infection number threshold θ1 (i.e. the first threshold) or the slope threshold θ2 (i.e. the second threshold) of the prediction curve using the historical infection number. When one of the predicted infection number or the slope of the prediction curve exceeds the respective threshold, a yellow warning is triggered, and when both the predicted infection number and the slope of the prediction curve exceed the respective threshold, a red warning is triggered, and is pushed to the relevant departments of the CDC at the same time. In order to resist data distribution drift, the system has built-in online iterative training and rolling deployment mechanism, and the model parameters are fine-tuned according to the latest monitoring data.

[0116] Up to now, the "data sampling - frequency domain decomposition - dual-branch prediction - threshold determination - model iteration" forms a closed loop, realizing the multi-site, high-precision infectious disease monitoring and grading warning process. The overall flowchart of the method of the embodiment is shown in Figure 2 .

[0117] 2. Data preprocessing module.

[0118] In order to increase the generalization ability of the model, a series of preprocessing operations are performed on the input data in the embodiment. In the embodiment, the data preprocessing process is completed in five steps of "collection - cleaning - alignment - normalization - decomposition" in sequence, which specifically includes:

[0119] (1) First, the original virus RNA concentration is captured, and the sewage inflow information, weather information, and clinical infectious disease monitoring data of the same period are synchronously called;

[0120] (2) Then, the collected data is denoised;

[0121] (3) Resample and align the multi-source sequence according to the unified UTC timestamp, and generate a 3-week moving average and a 1-week lag feature, and generate the corresponding label, i.e. the number of infected people Y1 and the curve slope Y2, which can be obtained by counting the number of infected people Y1 in the period, and then calculating the curve slope Y2 in a period according to the number of infected people;

[0122] (4) Perform Min-Max normalization on all numerical features processed in step (3), and persist the normalization parameters for recovery in the inference stage;

[0123] (5) Perform FFT on the clinical infectious disease monitoring data sequence, and decompose it into low-frequency components and high-frequency residuals according to the set threshold.

[0124] After the above operations, the data is split into a 4-12 week review time window, and split into training data set, validation data set and test data set according to 7:2:1.

[0125] 3, Sewage infectious disease monitoring and early warning model architecture.

[0126] The sewage infectious disease monitoring and early warning model architecture proposed in this embodiment adopts multi-source data fusion and a double-branch deep learning structure (i.e. a prediction model), which specifically includes:

[0127] (1) Time series signal decomposition module based on fast Fourier transform (FFT). The clinical case monitoring sequence (i.e. clinical case time series data) is decomposed into LFI components (i.e. low-frequency components) reflecting the trend of the epidemic and HFI residuals (i.e. high-frequency components) representing sudden fluctuations. Here, an adjustable frequency band division parameter (i.e. a threshold) is set to adapt to different infectious disease epidemic situations in different regions and times;

[0128] (2) Double-branch heterogeneous network architecture. The LFI branch (i.e. the first prediction branch) is configured with a multi-channel BiLSTM (hidden unit 128-256) to process multi-site data and capture long-term propagation rules; the HFI branch (i.e. the second prediction branch) extracts mutation features through a 20-layer residual network to depict discrete surge signals. Here, the two are dynamically fused through a learnable weight.

[0129] (3) Fusion output prediction. The prediction values of BiLSTM and ResNet branches are fused with dynamic weights, and the final output of the prediction model is obtained, and the prediction values of the number of infected people Y1 and the curve slope Y2 are obtained respectively;

[0130] (4) Single machine multi-source infectious disease epidemic monitoring supernode (i.e. Figure 3 The single machine supernode server refers to a high-performance node server deployed on a single physical server, which is usually used in blockchain, distributed network or high-performance computing scenarios. Four-way data from online sewage sensors, weather stations, clinical reports and laboratory RT-qPCR are written into the ring buffer in GPU memory. GPU uses cuFFT to complete FFT and sliding window alignment in batches, and runs BiLSTM low-frequency branch and ResNet high-frequency branch in parallel on 2 CUDA Streams, and then fuses the output in the memory. CPU is only responsible for threshold judgment and alarm publishing.

[0131] With the help of "zero-copy data injection + GPU preprocessing + double Stream pipeline" triple optimization, the average end-to-end delay is compressed, and the GPU utilization is improved. The closed-loop processing of data access, real-time prediction and automatic alarm is realized in a single server, reducing the risk of data leakage. The function structure diagram of single machine heterogeneous server is shown in Figure 3 .

[0132] 4. Training strategy.

[0133] The data is divided into training data set, validation data set and test data set according to the ratio of 7:2:1. This embodiment uses BiLSTM network to capture low-frequency signal trend, and introduces ResNet to process high-frequency mutation. Feature extraction is performed through FFT. In the prediction model training process, the final loss function (i.e. target loss function) is used, and the EarlyStopping technology is used to monitor the validation data set loss to avoid overfitting. The batch size is set to 64 to balance the calculation resources and the training speed, and the learning rate is set to 0.001. The number of training iterations is 100, and the Dropout layer is introduced to reduce the risk of overfitting. According to the performance indicators MAPE, MSE, R2 and MAE, continuous optimization is carried out. Finally, the saved double-branch heterogeneous network is quantized (i.e. target prediction model) and exported to FPGA for fast deployment for real-time prediction. High-frequency residual processing is still performed on GPU to ensure inference speed.

[0134] 5. Construction of target loss function.

[0135] Considering that the embodiment adopts a multi-site real-time monitoring and double-branch frequency domain decomposition architecture, and introduces auxiliary external features and a hierarchical early warning mechanism, in order to fully tap the collaborative optimization potential between multi-modal and multi-task, a multi-dimensional adaptive weighted target loss function is designed as follows:

[0136] Suppose there are N sewage monitoring sites, and the time series prediction value adopts a double-branch fusion model output:

[0137]

[0138] Among them, is the low-frequency trend prediction result at time step t (i.e. the prediction result of the first prediction branch), is the high-frequency residual prediction result at time step t (i.e. the prediction result of the second prediction branch), i = 1, 2, …, N, t is the time step.

[0139] The mean square errors of the low-frequency branch and the high-frequency branch are defined respectively:

[0140]

[0141] Among them, w i (t) is a dynamic site time weight, defined as follows.

[0142]

[0143] Among them, α i is the site static weight (which can be assigned by experts, reflecting the importance of key sites), β adjusts the influence of auxiliary features on the weight, is the mean auxiliary feature of all sites, x i (t) represents the auxiliary feature of the i-th sewage monitoring site, x j (t) represents the auxiliary feature of the j-th sewage monitoring site, y i (t) represents the future actual number of infections of the i-th sewage monitoring site.

[0144] The design of this dynamic site time weight can ensure that when the external environmental characteristics of a sewage monitoring site deviate from the average, the error response of the loss function to this site is enhanced, and the adaptability of the prediction model to abnormal complex situations is strengthened.

[0145] The embodiment also combines a warning threshold τ. When the prediction result deviates from the warning limit, the penalty is increased. The penalty function is defined as follows:

[0146]

[0147] Among them, γ is the penalty coefficient. This design encourages the model to reduce false negatives or false positives.

[0148] In summary, the final loss function (i.e., the target loss function) is defined by combining the weighted multi-branch error and the hierarchical early warning penalty as follows:

[0149]

[0150] where λ1 and λ2 control the importance proportion of the low-frequency branch error and the high-frequency branch error, respectively.

[0151] 6. Sewage infectious disease prediction and early warning.

[0152] Based on the final loss function, after training the BiLSTM branch and the ResNet branch through the training data set, the validation data set, and the test data set, the trained BiLSTM branch and the ResNet branch are used to predict the sewage data set to be predicted (which is processed by the data preprocessing module). The BiLSTM branch and the ResNet branch each predict the number of infected people, then dynamically weight and fuse the number of infected people predicted by the BiLSTM branch and the ResNet branch to obtain the total predicted number of infected people, and calculate the prediction curve slope according to the total predicted number of infected people. According to the total predicted number of infected people and the infected person threshold θ1, or comparing the prediction curve slope with the slope threshold θ2 of the prediction curve. When one of the total predicted number of infected people or the prediction curve slope exceeds the corresponding threshold, a yellow early warning is triggered, and when both the total predicted number of infected people and the prediction curve slope exceed the corresponding threshold, a red early warning is triggered, and is pushed to the relevant departments at the same time.

[0153] Compared with the prior art, the method of the embodiment has the following advantages:

[0154] The method of the embodiment adopts a closed loop of "sampling - frequency domain decomposition - double-branch prediction - threshold determination - model iteration" to realize a multi-site, high-precision infectious disease monitoring and hierarchical early warning process. Not only can various influencing factors be comprehensively considered, but also the accuracy and response speed of sewage infectious disease monitoring can be improved, providing scientific support for public health management.

[0155] In the evaluation indicators of the model of the embodiment, the mean square error (MSE) and the mean absolute percentage error (MAPE) are at a leading level compared with the prior art, the mean absolute error (MAE) and the determination coefficient (R 2) are at a high level. The method of the embodiment is also at a leading level in terms of early warning accuracy, with a sensitivity of 0.900 (95% CI: 0.667-1.000), a specificity of 0.833 (95% CI: 0.500-1.000), an accuracy of 0.875 (95% CI: 0.688-1.000), an AUC of 0.937 (95% CI: 0.845-1.000), and an F1 index of 0.900 (95% CI: 0.737-1.000). Among them, CI represents the confidence interval, and AUC (Area Under Curve) is defined as the area surrounded by the ROC curve and the coordinate axis.

[0156] With reference to Figure 4 The application also provides a sewage infectious disease monitoring and early warning system, which comprises a data acquisition unit 100, a data decomposition unit 200, a first prediction unit 300, a second prediction unit 400, a curve slope determination unit 500, and a monitoring and early warning unit 600.

[0157] The data acquisition unit 100 is configured to acquire first external monitoring data and first clinical case time series data corresponding to sewage to be predicted, and acquire a target prediction model comprising a first target prediction branch and a second target prediction branch.

[0158] The data decomposition unit 200 is configured to decompose the first clinical case time series data into a first low-frequency component and a first high-frequency component.

[0159] The first prediction unit 300 is configured to input the first low-frequency component and the first external monitoring data into the first target prediction branch of the target prediction model to obtain a predicted first number of infections.

[0160] The second prediction unit 400 is configured to input the first high-frequency component and the first external monitoring data into the second target prediction branch of the target prediction model to obtain a predicted second number of infections.

[0161] The curve slope determination unit 500 is configured to determine a predicted curve slope based on the first number of infections and the second number of infections.

[0162] The monitoring and early warning unit 600 is configured to perform sewage infectious disease monitoring and early warning based on the total predicted number of infections and the predicted curve slope.

[0163] In some embodiments, the data acquisition unit 100 can be specifically configured to:

[0164] acquire second external monitoring data and second clinical case time series data;

[0165] decompose the second clinical case time series data into a second low-frequency component and a second high-frequency component.

[0166] The second external monitoring data, the second low-frequency component and the second high-frequency component are used to train the prediction model to obtain a target prediction model, wherein the second low-frequency component and the second external monitoring data are used to train the first prediction branch to obtain a first target prediction branch, and the second high-frequency component and the second external monitoring data are used to train the second prediction branch to obtain a second target prediction branch.

[0167] In some embodiments, the data acquisition unit 100 can be specifically used for:

[0168] constructing a target loss function of the prediction model;

[0169] Based on the target loss function, the second external monitoring data, the second low-frequency component and the second high-frequency component are used to train the prediction model to obtain a target prediction model.

[0170] In some embodiments, the data acquisition unit 100 can be specifically used for:

[0171]

[0172] wherein, represents the target loss function, N represents the total number of sewage monitoring sites, w i (t) represents the time weight of the dynamic sewage monitoring site, λ1 represents the importance proportional factor of controlling the mean square error of the first prediction branch, represents the prediction result of the first prediction branch at the t time step, y i (t) represents the actual number of infected people at the i-th sewage monitoring site at the t time step, λ2 represents the importance proportional factor of controlling the mean square error of the second prediction branch, represents the prediction result of the second prediction branch at the t time step, p i (t) represents the penalty function at the t time step.

[0173] In some embodiments, the data decomposition unit 200 can be specifically used for:

[0174] The first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component by using fast Fourier transform.

[0175] In some embodiments, the curve slope determination unit 500 can be specifically used for:

[0176] The first number of infected people and the second number of infected people are dynamically weighted and fused to obtain a total predicted number of infected people;

[0177] Based on the total predicted number of infected people and the time corresponding to the sewage to be predicted, a prediction curve slope is determined.

[0178] In some embodiments, the monitoring and early warning unit 600 can be specifically used for:

[0179] presetting the first threshold value and the second threshold value;

[0180] dynamically weighting and fusing the first number of infected people and the second number of infected people to obtain a total predicted number of infected people;

[0181] comparing the total predicted number of infected people with the first threshold value to obtain a first comparison result;

[0182] comparing the slope of the prediction curve with the second threshold value to obtain a second comparison result;

[0183] monitoring and early warning of the sewage infectious disease according to the first comparison result and the second comparison result.

[0184] It should be noted that, since the sewage infectious disease monitoring and early warning system in the embodiment and the sewage infectious disease monitoring and early warning method described above are based on the same inventive concept, the corresponding contents in the method embodiment are also applicable to the system embodiment, which will not be described in detail here.

[0185] Referring to Figure 5 , the present embodiment also provides an electronic device, and the electronic device comprises:

[0186] at least one memory;

[0187] at least one processor;

[0188] at least one program;

[0189] The program is stored in the memory, and the processor executes the at least one program to implement the sewage infectious disease monitoring and early warning method described above.

[0190] The electronic device can be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), a vehicle-mounted computer, etc.

[0191] The electronic device of the present embodiment will be described in detail below.

[0192] The processor 1600 can be implemented in the form of a general central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the present embodiment.

[0193] The memory 1700 can be implemented in the form of a Read Only Memory (ROM), a static storage device, a dynamic storage device, or a Random Access Memory (RAM), etc. The memory 1700 can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present disclosure are implemented by software or firmware, the related program codes are stored in the memory 1700 and are called and executed by the processor 1600 to implement the sewage infectious disease monitoring and early warning method of the embodiments of the present disclosure.

[0194] The input / output interface 1800 is configured to realize information input and output.

[0195] The communication interface 1900 is configured to realize the communication interaction between the device and other devices. The communication can be realized by a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.).

[0196] The bus 2000 is configured to transmit information between various components (for example, the processor 1600, the memory 1700, the input / output interface 1800, and the communication interface 1900) of the device.

[0197] The processor 1600, the memory 1700, the input / output interface 1800, and the communication interface 1900 are connected to each other through the bus 2000 to realize the communication connection between the device.

[0198] The present disclosure further provides a storage medium, which is a computer readable storage medium, and stores computer executable instructions for causing a computer to execute the sewage infectious disease monitoring and early warning method.

[0199] The memory is a non-transitory computer readable storage medium, which can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0200] The embodiments described in the present disclosure are to more clearly illustrate the technical solutions of the present disclosure, and do not constitute a limitation on the technical solutions provided by the present disclosure. Those skilled in the art can know that, as technology evolves and new application scenarios appear, the technical solutions provided by the present disclosure are also applicable to similar technical problems.

[0201] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the present disclosure, and can include more or fewer steps than the figures shown, or combine certain steps, or different steps.

[0202] The device embodiments described above are merely illustrative, and units described as separate components can or can not be physically separated, that is, can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment.

[0203] Those skilled in the art can understand that all or some steps in the above disclosed method, the functions of the modules / units in the system and the device can be implemented as software, firmware, hardware and their appropriate combinations.

[0204] The terms "first", "second", "third", "fourth" and the like (if any) in the specification of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0205] It should be understood that, in the application, "at least one" refers to one or more, and "multiple" refers to two or more. "And / or" is used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, "A and / or B" can represent three cases of only A, only B, and A and B existing at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after it. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0206] It should be noted that in the present application, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or other elements inherent to such a process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to the order of performing the functions shown or discussed, but can also include performing the functions in a substantially simultaneous manner or in reverse order, for example, the described method can be performed in an order different from that described, and various steps can also be added, omitted, or combined. In addition, the features described with reference to some examples can be combined in other examples.

[0207] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed objects can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0208] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the units may be selected according to actual needs to achieve the purpose of the embodiment.

[0209] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present alone, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0210] If the integrated unit is realized 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 technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes multiple instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media. The embodiments of the present application are described in detail above in combination with the drawings, but the present application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the purpose of the present application, and all forms within the scope of the claims are within the protection of the present application.

[0211] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above specific embodiments. The above specific embodiments are only illustrative and not limiting. Those skilled in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, and all forms are within the protection of the present application.

Claims

1. A sewage infectious disease monitoring and early warning method, characterized in that: The method comprises: Acquire first external monitoring data and first clinical case time series data corresponding to the sewage to be predicted, and acquire a target prediction model including a first target prediction branch and a second target prediction branch; Decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component; Inputting the first low-frequency component and the first external monitoring data into a first target prediction branch in the target prediction model to obtain a predicted first number of infected persons; Inputting the first high-frequency component and the first external monitoring data into a second target prediction branch in the target prediction model to obtain a predicted second number of infected persons; Determining a predicted curve slope based on the first number of infected persons and the second number of infected persons; Sewage infectious disease monitoring and early warning are carried out based on the first number of infected people, the second number of infected people and the slope of the predicted curve.

2. The sewage infectious disease monitoring and early warning method according to claim 1 is characterized in that: The target prediction model is trained in the following way: acquiring second external monitoring data and second clinical case time series data; Decomposing the second clinical case time series data into a second low-frequency component and a second high-frequency component; The second external monitoring data, the second low-frequency component and the second high-frequency component are used to train the prediction model to obtain a target prediction model, wherein the second low-frequency component and the second external monitoring data are used to train the first prediction branch to obtain the first target prediction branch, and the second high-frequency component and the second external monitoring data are used to train the second prediction branch to obtain the second target prediction branch.

3. The sewage infectious disease monitoring and early warning method according to claim 2 is characterized in that: The method of training a prediction model using the second external monitoring data, the second low-frequency component, and the second high-frequency component to obtain a target prediction model includes: Constructing a target loss function of the prediction model; Based on the target loss function, the prediction model is trained using the second external monitoring data, the second low-frequency component, and the second high-frequency component to obtain a target prediction model.

4. The sewage infectious disease monitoring and early warning method according to claim 3 is characterized in that: The objective loss function of the prediction model is constructed, including: in, represents the target loss function, N represents the total number of sewage monitoring stations, w i (t) represents the time weight of the dynamic sewage monitoring station, λ1 represents the importance proportional factor of controlling the mean square error of the first prediction branch, represents the prediction result of the first prediction branch at time step t, y i (t) represents the actual number of infected people at the i-th sewage monitoring station at time step t, λ2 represents the importance proportional factor for controlling the mean square error of the second prediction branch, represents the prediction result of the second prediction branch at time step t, p i (t) represents the penalty function at time step t.

5. The sewage infectious disease monitoring and early warning method according to claim 1 is characterized in that: Decomposing the first clinical case time series data into a first low-frequency component and a first high-frequency component includes: The first clinical case time series data is decomposed into a first low-frequency component and a first high-frequency component using fast Fourier transform.

6. The sewage infectious disease monitoring and early warning method according to claim 1 is characterized in that: The determining of the slope of the predicted curve based on the first number of infected persons and the second number of infected persons includes: Dynamically weighting and fusing the first number of infected people and the second number of infected people to obtain the total predicted number of infected people; The slope of the prediction curve is determined based on the total predicted number of infected people and the time corresponding to the sewage to be predicted.

7. The sewage infectious disease monitoring and early warning method according to claim 1 is characterized in that: The sewage infectious disease monitoring and early warning according to the first number of infected persons, the second number of infected persons, and the slope of the prediction curve includes: Presetting a first threshold and a second threshold; Dynamically weighting and fusing the first number of infected people and the second number of infected people to obtain the total predicted number of infected people; Comparing the total predicted number of infected persons with the first threshold to obtain a first comparison result; Comparing the slope of the prediction curve with the second threshold to obtain a second comparison result; Sewage infectious disease monitoring and early warning are performed based on the first comparison result and the second comparison result.

8. A sewage infectious disease monitoring and early warning system, characterized in that: The system comprises: A data acquisition unit, configured to acquire first external monitoring data and first clinical case time series data corresponding to the sewage to be predicted, and to acquire a target prediction model comprising a first target prediction branch and a second target prediction branch; a data decomposition unit, configured to decompose the first clinical case time series data into a first low-frequency component and a first high-frequency component; a first prediction unit, configured to input the first low-frequency component and the first external monitoring data into a first target prediction branch in the target prediction model to obtain a predicted first number of infected persons; a second prediction unit, configured to input the first high-frequency component and the first external monitoring data into a second target prediction branch in the target prediction model to obtain a predicted second number of infected persons; A curve slope determining unit, configured to determine a predicted curve slope based on the first number of infected persons and the second number of infected persons; A monitoring and early warning unit is used to perform sewage infectious disease monitoring and early warning based on the first number of infected people, the second number of infected people and the slope of the prediction curve.

9. An electronic device, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the sewage infectious disease monitoring and early warning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the sewage infectious disease monitoring and early warning method as described in any one of claims 1 to 7.

Images