An influenza monitoring and early warning method and system based on a mobile epidemic interval method
Patent Information
- Application Number
- CN202410633866.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-21
- Publication Date
- 2026-10-27
AI Technical Summary
[0005]但是，在流感监测预警领域中，目前的绝大多数流感监测研究都是着眼于某地区的所有人员的流感流行强度和流行阈值，根据现有的流感监测研究无法提供更有针对性的流感防控方案
[0047]1、通过上述技术方案，本申请的方法将某地区的人群按照行业领域进行划分，然后利用移动流行区间法模型分别对每一领域的人群进行流感流行强度和流行阈值的分析，从而就可以获得更贴合与每一领域的人群的实际流感发展情况(包括流感流行强度和流行阈值)，进而就可以便于相关部门能够对每一领域人员分别进行更有针对性的流感分析，以制定与实际情况更加贴合地、流感预测效果更佳的流感防控方案。
Smart Images
Figure CN122991667A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of regional influenza surveillance technology, specifically to an influenza surveillance and early warning method based on the mobile epidemic interval method. Background Technology
[0002] Influenza, commonly known as the flu, is an acute respiratory infectious disease caused by the influenza virus, with fever and cough as the main symptoms. The influenza virus is highly contagious and primarily spreads through respiratory droplets. Seasonal influenza causes 290,000 to 650,000 deaths globally each year; therefore, real-time monitoring of influenza activity and prevalence levels is crucial for early warning and the implementation of appropriate prevention and control measures.
[0003] In the prevention and control of influenza, early warning and assessment of epidemic intensity are crucial. Currently, there is no unified standard for defining influenza epidemic thresholds, either domestically or internationally. Commonly used early warning models for influenza, such as cumulative sum control charts, exponentially weighted moving averages, Serfling regression models, and differential autoregressive moving average models, can be used for early epidemic warning, but they cannot accurately classify the level of epidemic intensity.
[0004] The Moving Epidemic Method (MEM) model was first proposed in 2004 by Spanish scholars Vega et al. to monitor the intensity of seasonal influenza epidemics. Due to its reliable results, it has been widely used in European countries. This method uses historical monitoring data to determine epidemic and non-epidemic periods and evaluates the sensitivity, specificity, positive predictive value, and negative predictive value of the established epidemic threshold.
[0005] However, in the field of influenza surveillance and early warning, most current influenza surveillance studies focus on the intensity and threshold of influenza prevalence among all people in a certain region, and cannot provide more targeted influenza prevention and control solutions based on existing influenza surveillance studies. Summary of the Invention
[0006] To address the technical problems in related technologies, this application provides an influenza surveillance and early warning method and system based on the mobile epidemic interval method. The influenza surveillance and early warning method based on the mobile epidemic interval method of this application can more accurately monitor the influenza epidemic intensity and epidemic threshold for each population group, thereby providing more targeted influenza prevention and control solutions.
[0007] To achieve the above objectives, the technical solution adopted in this application includes:
[0008] According to a first aspect of this application, an influenza surveillance and early warning method based on the mobile epidemic interval method is provided, comprising the following steps:
[0009] Based on industry sectors, all people in a certain region are divided into several groups;
[0010] The peak flu season time is divided for each population group separately;
[0011] Using the MEM model for analysis, the start and end of the monitoring season are divided by determining the minimum number of consecutive weeks r when the cumulative monitoring indicators reach their maximum percentage based on the peak time of influenza epidemic, and a MAP curve function calculation formula is constructed.
[0012] Then, select the largest monitoring indicator in the pre-epidemic period of each epidemic season, and calculate the upper limit of the one-sided 95% confidence interval of its arithmetic mean to obtain multiple epidemic thresholds. Then, combine them and calculate the upper limits of the one-sided 40%, 90% and 95% confidence intervals of their geometric means to obtain the first epidemic intensity threshold, the second epidemic intensity threshold and the third epidemic intensity threshold.
[0013] Cross-validation was used to evaluate the effectiveness of the MEM model. For different populations, each monitoring season was used as the target season for evaluation, and the remaining seasons were used as historical reference seasons. The pre- and post-epidemic thresholds established in the historical seasons were used as the reference standards for judgment. The number of positive cases monitored during the epidemic and non-epidemic periods of the target monitoring season was compared with the reference standards to obtain evaluation data for the diagnostic four-fold table.
[0014] Alternatively, the formula for calculating the MAP curve function can be constructed using the following equation:
[0015] in,
[0016] In the formula, This is the MAP function, which represents the percentage of the maximum cumulative sum of monitored indicators over r consecutive weeks in the j-th epidemic season to the total cumulative sum, t i,j Let be the i-th monitoring indicator for the j-th epidemic season, S be the total number of weeks in each epidemic season, and k be the starting week of r. The maximum cumulative sum of monitoring indicators for the j-th consecutive weeks of the epidemic season. This represents the total cumulative sum of monitoring indicators for the j-th epidemic season;
[0017] right Perform smooth regression to obtain the function. And calculate its weekly increment. When the weekly increase first falls below the preset critical threshold δ, the corresponding value of r is the optimal epidemic period length for the j-th epidemic season.
[0018] Determining the optimal length of the epidemic period Afterwards, the popular phases were divided by k * The decision was made that the epidemic period would begin from the kth...* Starting in the week, on the The week ended, and the pre-epidemic period lasted from week 1 to week k. * -1 week, the later stage of the epidemic is the 1st week Week S, k * Satisfy the following formula:
[0019] Optionally, the specific value of the critical threshold δ can be selected as follows:
[0020] The functional relationship for setting the critical threshold δ is as follows:
[0021] In the formula, δ i This refers to the critical threshold corresponding to each group of people. These are weighting coefficients corresponding to the start and end weeks of the peak influenza epidemic for each population group, where 'a' is the week number corresponding to the start week and 'b' is the week number corresponding to the end week. It is one in a thousand of the influenza positivity rate in the corresponding population;
[0022] Furthermore, the value range of δ is set to 2.0 to 4.0. If the δ calculated by the above formula exceeds the value range, the endpoint value of the value range closest to the calculation result is selected as the value of δ.
[0023] Optionally, the calculation of the popularity threshold specifically includes:
[0024] Select the n largest monitoring indicators in the pre-epidemic phase of each epidemic season, where n = 30 / N, and N is the number of historical seasons included.
[0025] Optionally, before dividing all people in a region into several groups based on industry sectors, data preprocessing is also included:
[0026] For data from a given year that exhibits two distinct peaks, a moving average method is used to perform a moving average over several weeks to smooth the data.
[0027] Optionally, the aforementioned groups include: cadres and employees, workers, housewives and unemployed, teachers, retirees, farmers, scattered children, commercial service workers, students, and children in childcare facilities.
[0028] Optionally, the peak influenza season can be divided for each population group, including:
[0029] The peak period for influenza among cadres and employees is from week 39 of the current year to week 23 of the following year;
[0030] The peak period for influenza among the working population is from week 41 of the current year to week 20 of the following year;
[0031] The peak period for influenza among housewives and the unemployed is from week 43 of the current year to week 16 of the following year.
[0032] The peak period for influenza among teachers is from week 45 of the current year to week 16 of the following year;
[0033] The peak season for influenza among retirees is from week 41 of the current year to week 23 of the following year;
[0034] The peak period for influenza among farmers is from week 41 of the current year to week 18 of the following year;
[0035] The peak period for influenza among scattered children is from week 39 of the current year to week 22 of the following year;
[0036] The peak period for influenza among commercial service workers is from week 44 of the current year to week 18 of the following year.
[0037] The peak season for influenza among students is from week 39 of the current year to week 24 of the following year;
[0038] The peak period for influenza among young children in daycare is from week 39 of the current year to week 25 of the following year.
[0039] According to a second aspect of this application, an influenza surveillance and early warning system based on the mobile epidemic zone method is also provided, characterized in that it is applied to the influenza surveillance and early warning method based on the mobile epidemic zone method described in any of the technical solutions of the first aspect of this application, and the system includes:
[0040] The data acquisition module is used to acquire epidemiological information on influenza-like illnesses of all people in a certain region, including their occupations;
[0041] The data segmentation module is used to segment the influenza-like illness epidemiological information of all people in a certain region according to industry sectors and their peak influenza epidemic time.
[0042] The MEM model is used to divide the start and end of the monitoring season by determining the minimum number of consecutive weeks r when the cumulative and maximum percentage of monitoring indicators are determined according to the peak time of the influenza epidemic, and to construct the MAP curve function calculation formula. It is used to select the largest monitoring indicator in the pre-epidemic period of each epidemic season, and then calculate the upper limit of the one-sided 95% confidence interval of its arithmetic mean to obtain multiple epidemic thresholds. After merging, the upper limits of the one-sided 40%, 90%, and 95% confidence intervals of their geometric means are calculated to obtain the first epidemic intensity threshold, the second epidemic intensity threshold, and the third epidemic intensity threshold.
[0043] The cross-validation module is used to evaluate the performance of the MEM model. The evaluation metrics used in this cross-validation module include sensitivity, specificity, positive predictive value, negative predictive value, and Youden index.
[0044] According to a third aspect of this application, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, is able to implement the steps of the influenza monitoring and early warning method based on the mobile epidemic zone method described in any of the technical solutions of the first aspect of this application.
[0045] According to a fourth aspect of this application, a computer-readable storage medium is also provided, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, it is able to implement the steps of the influenza monitoring and early warning method based on the mobile epidemic interval method described in any of the technical solutions of the first aspect of this application.
[0046] Beneficial effects:
[0047] 1. Through the above technical solution, the method of this application divides the population of a certain region according to industry sectors, and then uses the mobile epidemic interval method model to analyze the influenza epidemic intensity and epidemic threshold of the population in each sector. This allows for obtaining more accurate data on the actual influenza development (including influenza epidemic intensity and epidemic threshold) for each sector, which in turn enables relevant departments to conduct more targeted influenza analysis on the population in each sector, so as to formulate influenza prevention and control plans that are more in line with the actual situation and have better influenza prediction results.
[0048] 2. Other beneficial effects or advantages of this application will be described in detail in the specific embodiments. Attached Figure Description
[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0050] in:
[0051] Figure 1 This is a flowchart illustrating the steps of an influenza surveillance and early warning method based on the mobile epidemic zone method provided in an exemplary embodiment of this application. Detailed Implementation
[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of this application, but not all embodiments.
[0053] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0054] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. It should also be noted that in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be construed as preferred or advantageous over other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.
[0055] To facilitate a clearer and more accurate understanding of the technical solution of this application by those skilled in the art, the technical concept of this application will be explained below:
[0056] First, existing influenza surveillance studies primarily focus on the prevalence and threshold of influenza across the entire population of a given region. This ensures that influenza prevention and control plans developed based on these studies are generally applicable within that region. However, when considering specific population groups—such as those who commute daily to public, high-traffic areas (e.g., workers, commercial service personnel), or those who confine themselves to fixed, smaller spaces with low traffic (e.g., housewives, unemployed individuals, childcare workers)—it becomes clear that influenza prevention and control plans based on the influenza situation of the entire population are not effectively tailored to these specific groups. In other words, existing influenza surveillance methods do not provide targeted influenza prevention and control strategies.
[0057] In view of this, this application provides a novel solution: the influenza surveillance and early warning method based on the Moving Epidemic Method (MEM). This method divides the population into industry sectors (people in the same industry sector have similar activity ranges or patterns) and then uses the MEM model for influenza surveillance. This allows for obtaining more accurate data on the actual influenza situation (including influenza epidemic intensity and threshold) for people in the corresponding sectors, thus facilitating relevant departments to formulate more targeted influenza prevention and control plans based on the actual influenza situation in the corresponding sectors.
[0058] It should be noted that the influenza data for this applicant population can be sourced from epidemiological information on influenza-like illness (ILI) cases, the percentage of ILI cases, and the positive rate of nucleic acid testing, collected from multiple sentinel hospitals in a specific region, and summarized weekly. ILI cases are defined as patients with a body temperature higher than 38°C accompanied by cough or sore throat, but lacking other laboratory diagnoses. A period of high fever should occur during the acute fever episode. All collected ILI cases must undergo nucleic acid testing using a quantitative real-time method, and the results must be entered into the influenza surveillance network.
[0059] The technical solution of this application will be described in detail below with reference to the accompanying drawings.
[0060] Example 1
[0061] like Figure 1 As shown, according to a first aspect of this application, this embodiment provides an influenza surveillance and early warning method based on the mobile epidemic interval method, comprising the following steps:
[0062] Based on industry sectors, all people in a certain region can be divided into several categories. In this implementation, all people can be divided according to the actual situation of the region and the industry distribution in the region. For example, taking Xinjiang as an example, it can be divided into cadres and employees, workers, housewives and unemployed people, teachers, retirees, farmers, scattered children, commercial service people, students, and children in childcare.
[0063] The peak influenza season can be divided for each population group. Since the peak influenza season varies for different industries, the peak influenza season can be divided based on the collected data.
[0064] Using the MEM model for analysis, the start and end of the monitoring season are first determined based on the peak influenza epidemic time to calculate the minimum consecutive weeks r required for the cumulative and maximum percentage of monitoring indicators. This means that the length, start, and end times of the epidemic period for each industry sector are determined, dividing the entire epidemic season into three time periods: pre-epidemic, epidemic period, and post-epidemic. Then, the MAP curve function calculation formula can be constructed.
[0065] Then, select the largest monitoring indicator in the pre-epidemic period of each epidemic season, and calculate the upper limit of the one-sided 95% confidence interval of its arithmetic mean to obtain multiple epidemic thresholds. Then, combine them and calculate the upper limits of the one-sided 40%, 90% and 95% confidence intervals of their geometric means to obtain the first epidemic intensity threshold, the second epidemic intensity threshold and the third epidemic intensity threshold.
[0066] In this embodiment, the present application uses data from the pre-epidemic and post-epidemic periods to calculate the upper limit of the one-sided 95% confidence interval of the arithmetic mean of multiple maximum monitoring indicators as the pre-epidemic threshold and the post-epidemic threshold, respectively. At the same time, it uses data from the epidemic period to calculate the upper limits of the one-sided 40%, 90%, and 95% confidence intervals of the geometric mean of multiple maximum monitoring indicators as the first epidemic intensity threshold (i.e., the medium-intensity epidemic intensity threshold), the second epidemic intensity threshold (i.e., the high-intensity epidemic intensity threshold), and the third epidemic intensity threshold (i.e., the extremely high-intensity epidemic intensity threshold).
[0067] Cross-validation was used to evaluate the effectiveness of the MEM model. For different populations, each monitoring season was used as the target season for evaluation, and the remaining seasons were used as historical reference seasons. The pre- and post-epidemic thresholds established in the historical seasons were used as the reference standards for judgment. The number of positive cases monitored during the epidemic and non-epidemic periods of the target monitoring season was compared with the reference standards to obtain evaluation data for the diagnostic four-fold table.
[0068] In this implementation, the evaluation metrics used in cross-validation may include sensitivity, specificity, positive predictive value, negative predictive value, and Youden's index. Then, a database can be established to organize the data, and statistical analysis and modeling of the influenza virus detection positivity rate can be performed using the mem and memapp packages based on R software. This allows for more targeted influenza prevention and control solutions for different population groups.
[0069] Through the above technical solution, the method of this application divides the population of a certain region according to industry sectors, and then uses the mobile epidemic interval method model to analyze the influenza epidemic intensity and epidemic threshold of the population in each sector. This allows for obtaining more accurate data on the actual influenza development (including influenza epidemic intensity and epidemic threshold) for each sector, which in turn enables relevant departments to conduct more targeted influenza analysis on the population in each sector, so as to formulate influenza prevention and control plans that are more in line with the actual situation and have better influenza prediction results.
[0070] Taking the influenza surveillance conducted in Xinjiang from 2013 to 2023 by sector as an example, the method proposed in this application could be:
[0071] In accordance with the requirements of the China Influenza Surveillance Network, epidemiological information (including age, gender, occupation, and address), percentage of influenza-like cases, and nucleic acid test positivity rate of influenza-like cases monitored by 36 influenza sentinel hospitals in Xinjiang from January 2013 to December 2023 were collected and summarized for analysis on a weekly basis.
[0072] Xinjiang is located in a temperate arid zone, typically experiencing only one influenza peak each year. Based on this, the population of Xinjiang can be categorized into several groups: cadres and employees, workers, housewives and unemployed individuals, teachers, retirees, farmers, scattered children, commercial service workers, students, and children in childcare facilities. Furthermore, based on the prevalence of positive nucleic acid test rates, the peak influenza season can be determined for each of these groups. For example, the peak period of influenza epidemic for the entire population can be defined as week 38 of the current year to week 26 of the following year; the peak period for influenza epidemic for cadres and employees is week 39 of the current year to week 23 of the following year; the peak period for influenza epidemic for workers is week 41 of the current year to week 20 of the following year; the peak period for influenza epidemic for housewives and the unemployed is week 43 of the current year to week 16 of the following year; the peak period for influenza epidemic for teachers is week 45 of the current year to week 16 of the following year; the peak period for influenza epidemic for retirees is week 41 of the current year to week 23 of the following year; the peak period for influenza epidemic for farmers is week 41 of the current year to week 18 of the following year; the peak period for influenza epidemic for scattered children is week 39 of the current year to week 22 of the following year; the peak period for influenza epidemic for commercial service workers is week 44 of the current year to week 18 of the following year; the peak period for influenza epidemic for students is week 39 of the current year to week 24 of the following year; and the peak period for influenza epidemic for children in kindergartens and nurseries is week 39 of the current year to week 25 of the following year.
[0073] Subsequently, historical monitoring data can be used to build a model using the MEM model. For each epidemic season, the start and end of the monitoring season are defined by the minimum number of consecutive weeks *r* when the maximum accumulated rates percentage (MAP) of the defined monitoring indicators is reached. Furthermore, using data from the pre-epidemic and post-epidemic periods, the one-sided 95% CI upper limit of the arithmetic mean of the 30 largest monitoring indicators is calculated as the pre-epidemic and post-epidemic thresholds, respectively. Simultaneously, using data from the epidemic period, the one-sided 40%, 90%, and 95% CI upper limits of the geometric mean of the 30 largest monitoring indicators are calculated as the epidemic warning thresholds for three levels: medium, high, and extremely high intensity epidemics.
[0074] Finally, for different population groups, each monitoring season was used as the target season for evaluation, while other seasons were used as historical reference seasons. Pre- and post-epidemic thresholds established from historical seasons were used as the reference standards. The number of positive cases during the epidemic and non-epidemic periods of the target monitoring season was compared with the reference standards to obtain evaluation data for the diagnostic four-fold table. The evaluation indicators used in cross-validation could include sensitivity, specificity, positive predictive value, negative predictive value, and Youden's index. Then, a database could be established using Excel, and the data could be organized. Statistical analysis and modeling of the influenza virus detection positivity rate were performed using the mem and memapp packages based on R software. This provides more targeted influenza prevention and control solutions for different population groups.
[0075] Specifically, using influenza virus positivity data from the 2013-2023 flu seasons in Xinjiang as a baseline, a MEM early warning model was established to calculate the start and end thresholds of the influenza epidemic cycle. Then, the 30 largest monitoring data points within the epidemic cycle were taken, and the upper limits of the one-sided 40%, 90%, and 95% CI of the geometric mean were used as the three influenza epidemic early warning thresholds for medium, high, and very high levels. Based on this, the start time and intensity of the epidemic in different occupational groups in Xinjiang during the 2023 / 2024 flu season were estimated.
[0076] The results showed that the threshold for the start of the influenza season in the entire population of Xinjiang during the 2023 / 2024 influenza surveillance year was 184.45‰, with the threshold for the start of the season ranging from 2.17‰ to 69.04‰ for different occupational groups; the threshold for the end of the season was 170.22‰, with the threshold for the end of the season ranging from 1.57‰ to 81.85‰ for different occupational groups; the influenza virus positivity rate in the entire population was 184.45‰ to 331.72‰, indicating a low epidemic level; 331.72‰ to 415.62‰, indicating a medium epidemic level; 415.62‰ to 459.18‰, indicating a high epidemic level; and ≥459.18‰, indicating an extremely high epidemic level. The 48th week of 2023 was the start week of the epidemic, and the 9th week of 2024 was the end week, with the epidemic lasting 13 weeks. Specific details regarding the thresholds for the start of the season, the end of the season, and the medium, high, and extremely high epidemic levels for different occupational groups can be found in the table below.
[0077]
[0078]
[0079] In this way, by using the MEM method to establish influenza epidemic thresholds for different occupational groups in Xinjiang, a mid-latitude arid region, and to conduct early warning classifications for the incidence of influenza in the next monitoring cycle, this application can effectively establish more reliable and accurate influenza early warnings for different industries and groups. This makes it easier to use the above results to provide more suitable influenza prevention and control programs for different occupational groups in the region.
[0080] In one embodiment of this application, the MAP curve function calculation formula can be constructed using the following formula:
[0081] in,
[0082] In the formula, This is the MAP function, which represents the percentage of the maximum cumulative sum of monitored indicators over r consecutive weeks in the j-th epidemic season to the total cumulative sum, t i,j Let be the i-th monitoring indicator for the j-th epidemic season, S be the total number of weeks in each epidemic season, and k be the starting week of r. The maximum cumulative sum of monitoring indicators for the j-th consecutive weeks of the epidemic season. This represents the total cumulative sum of monitoring indicators for the j-th epidemic season;
[0083] right Perform smooth regression to obtain the function. And calculate its weekly increment. When the weekly increase first falls below the preset critical threshold δ, the corresponding value of r is the optimal epidemic period length for the j-th epidemic season.
[0084] Determining the optimal length of the epidemic period Afterwards, the popular phases were divided by k * The decision was made that the epidemic period would begin from the kth... * Starting in the week, on the The week ended, and the pre-epidemic period lasted from week 1 to week k. * -1 week, the later stage of the epidemic is the 1st week Week S, k * Satisfy the following formula:
[0085] In one embodiment of this application, the specific value of the critical threshold δ selected in this application can be as follows:
[0086] The functional relationship for setting the critical threshold δ is as follows:
[0087] In the formula, δ i This refers to the critical threshold corresponding to each group of people. These are weighting coefficients corresponding to the start and end weeks of the peak influenza epidemic for each population group, where 'a' is the week number corresponding to the start week and 'b' is the week number corresponding to the end week. It is one in a thousand of the influenza positivity rate in the corresponding population;
[0088] Furthermore, the value range of δ is set to 2.0 to 4.0. If the δ calculated by the above formula exceeds the value range, the endpoint value of the value range closest to the calculation result is selected as the value of δ.
[0089] In this way, by using the functional relationship of the critical threshold δ set in this way, a critical threshold that fits more closely to each group of people can be obtained, thereby enabling the MEM model to achieve optimal performance. That is, the MEM model established by the critical threshold method given by the functional relationship of the critical threshold δ in this application has better sensitivity and specificity than the conventional moving interval method (see the table below).
[0090] The critical threshold δ of the conventional mobile epidemic interval method is usually selected by direct assignment, which does not closely match the actual situation of each type of population (or rather, its closeness depends entirely on experience and luck).
[0091] For example, please refer to the table below, which shows the results of the model generated by the critical threshold δ assigned by the conventional mobile popularity interval method and the results of the model generated by the critical threshold δ assigned by the assignment method of this application.
[0092]
[0093] As shown in the table above, relatively speaking, the method in this application shows better fitting results for workers, retirees, farmers, scattered children, and children in childcare facilities, with sensitivity reaching over 75% and specificity reaching over 85%. Although the sensitivity of other occupational groups is lower, the specificity can still reach over 90%. Therefore, it is entirely feasible to establish corresponding early warning thresholds for these groups using the method in this application, and more targeted influenza prevention and control plans can be implemented based on the corresponding early warning thresholds.
[0094] For example, in one exemplary influenza prevention and control scheme of this application, the establishment of the epidemic threshold and the determination of the epidemic start week can provide a more reliable basis for the optimal time and time planning of influenza vaccination. The threshold intensity and start time of influenza epidemic are different for different groups. For example, the epidemic start time for students, scattered children, and children in childcare is earlier, so the early supply of influenza vaccine can be prioritized for these groups. The epidemic start time for retirees, farmers, commercial service workers, teachers, and other groups is slightly later, so the time for influenza vaccination can be postponed. In other words, according to the different influenza epidemic times of different groups, staggered vaccination can avoid the strain on medical resources and help ensure the smooth progress of universal vaccination and the control of influenza epidemic.
[0095] In one embodiment of this application, the calculation of the epidemic threshold may specifically include: selecting the n largest monitoring indicators in the pre-epidemic phase of each epidemic season, where n = 30 / N, and N is the number of historical seasons included.
[0096] In one embodiment of this application, before dividing all people in a region into several groups based on industry sectors, data preprocessing is also included: for cases where there are two epidemic peaks in the data of a certain year, a moving average method is used to perform a moving average over several weeks to smooth the data.
[0097] This is because the MEM model fits single-peak epidemic diseases better than double-peak ones during the epidemic season. In most years, the influenza epidemic pattern in Xinjiang is single-peak, but in some years, there is a small epidemic peak after the epidemic peak. In order to better fit the model and make predictions, this implementation method can be used. That is, for a certain year's data with double epidemic peaks, the moving average method is used to perform a moving average for several weeks (e.g., 5 weeks) to smooth out outliers and fluctuations, so as to improve the fitting effect and improve the sensitivity and specificity of early warning.
[0098] Understandably, if the data quality for a particular year is low and the model fit is poor, a targeted elimination approach can be adopted to retain the years with better data quality for model optimization.
[0099] In one embodiment of this application, the various population categories may include: cadres and employees, workers, housewives and unemployed, teachers, retirees, farmers, scattered children, commercial service workers, students, and children in childcare facilities.
[0100] It is understandable that the distribution of occupations may vary in different regions and can be adjusted according to the actual situation. This application does not impose specific restrictions on this.
[0101] In one embodiment of this application, the peak influenza season for each population group is divided as follows:
[0102] The peak period for influenza among cadres and employees is from week 39 of the current year to week 23 of the following year;
[0103] The peak period for influenza among the working population is from week 41 of the current year to week 20 of the following year;
[0104] The peak period for influenza among housewives and the unemployed is from week 43 of the current year to week 16 of the following year.
[0105] The peak period for influenza among teachers is from week 45 of the current year to week 16 of the following year;
[0106] The peak season for influenza among retirees is from week 41 of the current year to week 23 of the following year;
[0107] The peak period for influenza among farmers is from week 41 of the current year to week 18 of the following year;
[0108] The peak period for influenza among scattered children is from week 39 of the current year to week 22 of the following year;
[0109] The peak period for influenza among commercial service workers is from week 44 of the current year to week 18 of the following year.
[0110] The peak season for influenza among students is from week 39 of the current year to week 24 of the following year;
[0111] The peak period for influenza among young children in daycare is from week 39 of the current year to week 25 of the following year.
[0112] According to a second aspect of this application, an influenza surveillance system based on the mobile epidemic zone method is also provided, characterized in that it is applied to the influenza surveillance and early warning method based on the mobile epidemic zone method in any of the technical solutions of the first aspect of this application, the system comprising:
[0113] The data acquisition module is used to acquire epidemiological information on influenza-like illnesses of all people in a certain region, including their occupations;
[0114] The data segmentation module is used to segment the influenza-like illness epidemiological information of all people in a certain region according to industry sectors and their peak influenza epidemic time.
[0115] The MEM model is used to divide the start and end of the monitoring season by determining the minimum number of consecutive weeks r when the cumulative and maximum percentage of monitoring indicators are determined according to the peak time of the influenza epidemic, and to construct the MAP curve function calculation formula. It is used to select the largest monitoring indicator in the pre-epidemic period of each epidemic season, and then calculate the upper limit of the one-sided 95% confidence interval of its arithmetic mean to obtain multiple epidemic thresholds. After merging, the upper limits of the one-sided 40%, 90%, and 95% confidence intervals of their geometric means are calculated to obtain the first epidemic intensity threshold, the second epidemic intensity threshold, and the third epidemic intensity threshold.
[0116] The cross-validation module is used to evaluate the performance of the MEM model. The evaluation metrics used in this cross-validation module include sensitivity, specificity, positive predictive value, negative predictive value, and Youden index.
[0117] According to a third aspect of this application, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it is able to implement the steps of the influenza monitoring and early warning method based on the mobile epidemic zone method in any of the technical solutions of the first aspect of this application.
[0118] It is understood that in this embodiment, the memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as read-only memory, flash memory, hard disk or solid-state drive; furthermore, the memory may include combinations of the above types of memory. This application does not specifically limit it in this regard.
[0119] Similarly, the processor can implement or execute the various exemplary logical steps described in conjunction with the disclosure of this application. The processor can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logical steps described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.
[0120] According to a fourth aspect of this application, a computer-readable storage medium is also provided, on which a computer program is stored, characterized in that, when executed by a processor, the computer program is capable of implementing the steps of the influenza monitoring and early warning method based on the mobile epidemic interval method in any of the technical solutions of the first aspect of this application.
[0121] In this embodiment, the computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), a register, a hard disk, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof, or any other form of computer-readable storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and storage medium may reside within an application-specific integrated circuit (ASIC). In embodiments of this application, the computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.
[0122] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A method for influenza surveillance and early warning based on the mobile epidemic zone method, characterized in that, Includes the following steps: Based on industry sectors, all people in a certain region are divided into several groups; The peak flu season time is divided for each population group separately; Using the MEM model for analysis, the start and end of the monitoring season are divided by determining the minimum number of consecutive weeks r when the cumulative monitoring indicators reach their maximum percentage based on the peak time of influenza epidemic, and a MAP curve function calculation formula is constructed. Then, select the largest monitoring indicator in the pre-epidemic period of each epidemic season, and calculate the upper limit of the one-sided 95% confidence interval of its arithmetic mean to obtain multiple epidemic thresholds. Then, combine them and calculate the upper limits of the one-sided 40%, 90% and 95% confidence intervals of their geometric means to obtain the first epidemic intensity threshold, the second epidemic intensity threshold and the third epidemic intensity threshold. Cross-validation was used to evaluate the effectiveness of the MEM model. For different populations, each monitoring season was used as the target season for evaluation, and the remaining seasons were used as historical reference seasons. The pre- and post-epidemic thresholds established in the historical seasons were used as the reference standards for judgment. The number of positive cases monitored during the epidemic and non-epidemic periods of the target monitoring season was compared with the reference standards to obtain evaluation data for the diagnostic four-fold table.
2. The influenza surveillance and early warning method based on the mobile epidemic zone method according to claim 1, characterized in that, The formula for calculating the MAP curve function is constructed using the following formula: in, In the formula, This is the MAP function, which represents the percentage of the maximum cumulative sum of monitored indicators over r consecutive weeks in the j-th epidemic season to the total cumulative sum, t i，j Let be the i-th monitoring indicator for the j-th epidemic season, S be the total number of weeks in each epidemic season, and k be the starting week of r. The maximum cumulative sum of monitoring indicators for the j-th consecutive weeks of the epidemic season. This represents the total cumulative sum of monitoring indicators for the j-th epidemic season; right Perform smooth regression to obtain the function. And calculate its weekly increment. When the weekly increase first falls below the preset critical threshold δ, the corresponding value of r is the optimal epidemic period length for the j-th epidemic season. Determining the optimal length of the epidemic period Afterwards, the popular phases were divided by k * The decision was made that the epidemic period would begin from the kth... * Starting in the week, on the The week ended, and the pre-epidemic period lasted from week 1 to week k. * -1 week, the later stage of the epidemic is the 1st week Week S, k * Satisfy the following formula:
3. The influenza surveillance and early warning method based on the mobile epidemic zone method according to claim 2, characterized in that, The specific values for selecting the critical threshold δ include: The functional relationship for setting the critical threshold δ is as follows: In the formula, δ i This refers to the critical threshold corresponding to each group of people. These are weighting coefficients corresponding to the start and end weeks of the peak influenza epidemic for each population group, where 'a' is the week number corresponding to the start week and 'b' is the week number corresponding to the end week. It is one in a thousand of the influenza positivity rate in the corresponding population; Furthermore, the value range of δ is set to 2.0 to 4.
0. If the δ calculated by the above formula exceeds the value range, the endpoint value of the value range closest to the calculation result is selected as the value of δ.
4. The influenza surveillance and early warning method based on the mobile epidemic zone method according to claim 1, characterized in that, The calculation of the popularity threshold specifically includes: Select the n largest monitoring indicators in the pre-epidemic phase of each epidemic season, where n = 30 / N, and N is the number of historical seasons included.
5. The influenza surveillance and early warning method based on the mobile epidemic zone method according to claim 1, characterized in that, Before dividing all people in a region into several groups based on industry sectors, data preprocessing is also included: For data from a given year that exhibits two distinct peaks, a moving average method is used to perform a moving average over several weeks to smooth the data.
6. The influenza surveillance and early warning method based on the mobile epidemic zone method according to claim 1, characterized in that, The aforementioned groups include: cadres and employees, workers, housewives and unemployed, teachers, retirees, farmers, scattered children, commercial service workers, students, and children in childcare facilities.
7. The influenza surveillance and early warning method based on the mobile epidemic zone method according to claim 6, characterized in that, The peak influenza season for each population group is divided into the following periods: The peak period for influenza among cadres and employees is from week 39 of the current year to week 23 of the following year; The peak period for influenza among the working population is from week 41 of the current year to week 20 of the following year; The peak period for influenza among housewives and the unemployed is from week 43 of the current year to week 16 of the following year. The peak period for influenza among teachers is from week 45 of the current year to week 16 of the following year; The peak season for influenza among retirees is from week 41 of the current year to week 23 of the following year; The peak period for influenza among farmers is from week 41 of the current year to week 18 of the following year; The peak period for influenza among scattered children is from week 39 of the current year to week 22 of the following year; The peak period for influenza among commercial service workers is from week 44 of the current year to week 18 of the following year. The peak season for influenza among students is from week 39 of the current year to week 24 of the following year; The peak period for influenza among young children in daycare is from week 39 of the current year to week 25 of the following year.
8. An influenza surveillance system based on the mobile epidemic zone method, characterized in that, The influenza surveillance and early warning method based on the mobile epidemic zone method according to any one of claims 1-7, the system comprising: The data acquisition module is used to acquire epidemiological information on influenza-like illnesses of all people in a certain region, including their occupations; The data segmentation module is used to segment the influenza-like illness epidemiological information of all people in a certain region according to industry sectors and their peak influenza epidemic time. The MEM model is used to divide the start and end of the monitoring season by determining the minimum number of consecutive weeks r when the cumulative and maximum percentage of monitoring indicators are determined according to the peak time of the influenza epidemic, and to construct the MAP curve function calculation formula. It is used to select the largest monitoring indicator in the pre-epidemic period of each epidemic season, and then calculate the upper limit of the one-sided 95% confidence interval of its arithmetic mean to obtain multiple epidemic thresholds. After merging, the upper limits of the one-sided 40%, 90%, and 95% confidence intervals of their geometric means are calculated to obtain the first epidemic intensity threshold, the second epidemic intensity threshold, and the third epidemic intensity threshold. The cross-validation module is used to evaluate the performance of the MEM model. The evaluation metrics used in this cross-validation module include sensitivity, specificity, positive predictive value, negative predictive value, and Youden index.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it is able to implement the steps of the influenza surveillance and early warning method based on the mobile epidemic zone method as described in any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it can implement the steps of the influenza surveillance and early warning method based on the mobile epidemic zone method as described in any one of claims 1-7.