A method for epidemic risk assessment and prediction

Abstract

The present invention relates to the technical field of epidemic monitoring, and in particular to an epidemic risk assessment and prediction method, which calculates the epidemic diagnosis correlation of each infection-related indicator in the dimension of the impact of the infection-related indicator on the epidemic diagnosis, thereby screening out the epidemic impact indicators that will have a practical impact on the epidemic diagnosis; and further screening out the key patients who have a greater impact on the spread of the epidemic according to the various epidemic impact indicators involved in the sick group; and then determining the epidemic spread performance more accurately according to the spread of the key patients through the region before the diagnosis and the medical capacity of the region, so as to further perform more accurate epidemic risk assessment and prediction according to the numerical distribution of the epidemic spread performance of the key patients combined with cluster analysis, so as to achieve better effects of epidemic risk assessment and prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of epidemic monitoring, and in particular to an epidemic risk assessment and prediction method. Background Art

[0002] In the current public health field, disease prediction and analysis are important means of controlling the spread of infectious diseases. Accurate risk prediction or assessment of epidemics before they expand on a large scale can provide more reasonable prevention and control measures and response ideas for epidemic prevention and control.

[0003] The existing methods for predicting or assessing epidemic risks usually rely on historical data as a whole, such as transmission speed, mortality rate, and cure period, but this method ignores the impact of heterogeneity between individual cases on disease transmission, and fails to fully consider the role of social behavioral characteristics such as patient contact history and residential population density in disease spread; in addition, the level of hospital treatment also has potential to contribute to slowing the development of the epidemic. Obviously, the existing technology usually does not consider or consider the above characteristics simultaneously, resulting in low accuracy of existing technology in epidemic risk assessment and prediction. Summary of the invention

[0004] The present application provides an epidemic risk assessment and prediction method, which first calculates the epidemic diagnosis correlation for each infection-related indicator in the dimension of the impact of the infection-related indicator on the epidemic diagnosis, so as to screen out the epidemic impact indicators that will have a practical impact on the epidemic diagnosis; and further screen out the key patients who have a greater impact on the spread of the epidemic according to the various epidemic impact indicators involved in the sick population; and then determine the epidemic transmission performance more accurately according to the transmission situation of the key patients through the region before the diagnosis and the medical capacity of the region, so as to further perform more accurate epidemic risk assessment and prediction according to the numerical distribution of the epidemic transmission performance of the key patients combined with cluster analysis, thereby solving the problem of low accuracy of epidemic risk assessment and prediction in the prior art, and making the effect of epidemic risk assessment and prediction better.

[0005] The present application provides a method for epidemic risk assessment and prediction, the method comprising:

[0006] Obtain the number of newly confirmed cases of the epidemic and the number of newly confirmed cases of each infection-related indicator every day in the preset time period of the assessment area;

[0007] Determine the epidemiological confirmed correlation of each infection-related indicator based on the temporal correlation between the number of newly confirmed cases of the epidemic and the number of newly confirmed cases of each infection-related indicator; and select the epidemiological impact indicators based on the epidemiological confirmed correlation;

[0008] Determine the degree of concern for each epidemic confirmed patient according to the overall magnitude of the epidemic confirmed correlation of each epidemic impact indicator corresponding to each epidemic confirmed patient; and screen out all key patients according to the degree of concern;

[0009] Determine the degree of transmission impact of each key patient based on the change in the number of confirmed cases of the epidemic in the region after the diagnosis of each key patient; determine the epidemiological transmission performance of each key patient based on the epidemiological recovery situation of the regional hospitals before the diagnosis of each key patient and the degree of transmission impact;

[0010] According to the epidemic transmission performance, the degree of concern of each key patient and the corresponding epidemic recovery period, the clustering reference value of each key patient is determined; and the epidemic risk is assessed and predicted based on the clustering results after cluster analysis based on the clustering reference value.

[0011] Furthermore, the process of obtaining the epidemiological diagnosis correlation includes:

[0012] Arrange the number of new infections in each infection-related indicator in a time sequence within the preset time period of the assessment area, and determine the sequence of new infections in each infection-related indicator; Arrange the number of new infections in each infection-related indicator in a time sequence within the preset time period of the assessment area, and determine the sequence of new infections in each infection-related indicator;

[0013] The epidemiological confirmation correlation of each infection-related indicator is determined based on the Pearson correlation coefficient between the series of newly confirmed epidemic numbers and the series of newly confirmed infection numbers of infection-related indicators.

[0014] Furthermore, the process of obtaining the epidemic impact index includes:

[0015] The infection-related indicators whose corresponding epidemic diagnosis correlation is greater than the preset correlation threshold are used as epidemic impact indicators.

[0016] Furthermore, the process of obtaining the degree of attention includes:

[0017] All the epidemic impact indicators matched to each epidemic confirmed patient are obtained; the cumulative value of the epidemic confirmed correlation of all the epidemic impact indicators corresponding to each epidemic confirmed patient is used as the attention level of each epidemic confirmed patient.

[0018] Furthermore, the process of acquiring the focused patients includes:

[0019] The average of the corresponding attention levels of all confirmed epidemic patients is taken as the key attention threshold; all confirmed epidemic patients whose corresponding attention levels are greater than the key attention threshold are taken as key attention patients.

[0020] Furthermore, the process of obtaining the degree of propagation influence includes:

[0021] Obtain the movement path of each key patient within the preset first statistical time period before diagnosis; use the area corresponding to the preset radiation range of each path point on the movement path as the epidemic spread area of ​​each path point; use all the infection radiation areas corresponding to all the path points on the movement path as the overall spread area of ​​each key patient; use the preset first statistical time period before diagnosis and the preset second statistical time period after diagnosis of each key patient as the spread influence time period of each key patient;

[0022] During the transmission impact time period, the ratio between the number of new epidemic people in the overall transmission area corresponding to each key patient and the total population in the overall transmission area is used as the transmission impact degree of each key patient.

[0023] Furthermore, the process of obtaining the epidemic transmission performance includes:

[0024] Determine the epidemiological cure index for each key patient based on the overall epidemiological treatment capacity of each hospital in the overall transmission area corresponding to each key patient;

[0025] The product of the negative correlation mapping value of the epidemic cure index and the degree of transmission impact is used as the epidemic transmission performance of each key patient.

[0026] Furthermore, the process of obtaining the epidemic cure index includes:

[0027] The product of the number of doctors treating epidemics and the number of medical facilities in each hospital is used as the epidemic treatment resource parameter of each hospital; the normalized value of the total number of people cured of the epidemic in each hospital within a preset time period is used as the epidemic cure rate parameter of each hospital; the product of the epidemic treatment resource parameter and the epidemic cure rate parameter is used as the reference value of the treatment effect of each hospital on the epidemic; the average of the reference values ​​of the treatment effect of all hospitals on the epidemic in the overall transmission area is used as the epidemic cure index of each key patient.

[0028] Furthermore, the process of obtaining the cluster reference value includes:

[0029] The product of the degree of concern of each key patient, the epidemic cure cycle and the epidemic transmission performance is used as the clustering reference value when each key patient participates in clustering.

[0030] Furthermore, the process of performing epidemic risk assessment and prediction based on the clustering results after cluster analysis according to the clustering reference value includes:

[0031] Perform k-means cluster analysis on the cluster reference values ​​of all key patients to obtain at least two clusters; take the mean of the cluster reference values ​​of all key patients in each cluster as the overall reference value of each cluster; take the ratio between the number of key patients in the cluster with the largest overall reference value and the total number of key patients in all clusters as the reference ratio; take the normalized value of the overall reference value corresponding to the cluster with the largest overall reference value as the risk reference value; take the product of the reference ratio and the risk reference value as the risk prediction assessment value;

[0032] When the risk prediction evaluation value is greater than a preset first prediction threshold, the epidemic is evaluated as a high-risk epidemic;

[0033] When the risk prediction evaluation value is less than or equal to a preset first prediction threshold and greater than a preset second prediction threshold, the epidemic is evaluated as a medium-risk epidemic;

[0034] When the risk prediction evaluation value is less than or equal to a preset second prediction threshold, the epidemic is evaluated as a low-risk epidemic.

[0035] This application includes the following benefits:

[0036] For each infection-related indicator, the closer the time series change of the number of new cases per day in the preset time period is to the time series change of the number of new cases per day in the confirmed epidemic, the higher the correlation between the corresponding infection-related indicator and the confirmed epidemic, that is, the higher the importance of the corresponding infection-related indicator to disease prediction and evaluation, so as to further analyze the patients including infection-related indicators, so as to more significantly characterize the adverse effects caused by the epidemic, so as to conduct a more accurate epidemic impact analysis. Therefore, for each patient confirmed by the epidemic, the larger the overall corresponding epidemic impact indicators are, the higher the importance of the confirmed patient to the disease prediction and evaluation, and the higher the corresponding degree of attention, so as to further screen out the key patients according to the degree of attention, and conduct a more accurate epidemic risk analysis according to the epidemic impact caused by the key patients.

[0037] For each key patient, the path they traveled before being diagnosed will reflect the transmission characteristics of the epidemic, and the more significant the change in the number of confirmed epidemics after the key patients passed through these areas, the greater the corresponding degree of transmission impact; in addition, the medical level in the area passed by each key patient needs to be considered. If the medical level of each hospital in the area passed through is relatively high, the control or management ability of the epidemic will be stronger to a certain extent, and the degree of impact will be lower. Therefore, according to the epidemic cure situation of the hospitals in the area passed by each key patient before being diagnosed and the degree of transmission impact, the epidemic transmission performance of each key patient is determined. After determining the epidemic transmission performance, combined with the characteristics that the length of the epidemic cure period will affect its risk, further on the basis of the epidemic transmission performance and the cure period, a comprehensive analysis is performed with the corresponding degree of attention for each key patient as a weight. After cluster analysis is performed according to the cluster reference value obtained in the analysis, a more accurate epidemic risk assessment and prediction is performed based on the cluster clusters with larger transmission performance in the clustering results. This allows the disease risk assessment and prediction to simultaneously combine the heterogeneity between individual cases, patient contact history, social behavioral characteristics such as residential population density, and the impact of hospital treatment level on disease transmission, making the epidemic risk assessment and prediction more effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A flow chart of an epidemic risk assessment and prediction method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of an epidemic risk assessment and prediction method proposed in accordance with the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are used only for descriptive purposes and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0041] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0042] The specific scheme of an epidemic risk assessment and prediction method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0043] This application embodiment provides a method for evaluating and predicting epidemic risk. Figure 1 , which shows a flow chart of an epidemic risk assessment and prediction method provided by an embodiment of the present invention, the method comprising:

[0044] Step S101: Obtain the number of newly confirmed cases of the epidemic and the number of newly confirmed cases of each infection-related indicator every day in the preset time period in the assessment area.

[0045] In a specific implementation of an embodiment of the present invention, the preset time period is set to a time period corresponding to half a year before the evaluation or prediction moment; the types of infection-related indicators include: confirmed respiratory diseases, confirmed immune system diseases, confirmed cardiovascular diseases, confirmed nervous system diseases, confirmed infectious fevers, and confirmed viral colds. The types of infection-related indicators can be determined by the specific implementation environment, and no further elaboration is made here; for example, for confirmed respiratory diseases, the corresponding daily increase in the number of infection-related indicators is also the daily increase in the number of confirmed respiratory diseases, and no further elaboration is made here. It should be noted that all data in the embodiments of the present invention are obtained through reports from the health department of the evaluation area, and the data used are all authorized data for use, and no further elaboration is made here.

[0046] Step S102: Determine the epidemic confirmation correlation of each infection-related indicator according to the temporal correlation between the number of newly confirmed epidemic cases and the number of newly added infection-related indicators; and select the epidemic impact indicators according to the epidemic confirmation correlation.

[0047] For each infection-related indicator, the closer the time series change in the number of new cases per day within the preset time period is to the time series change in the number of new cases confirmed per day, the higher the correlation between the corresponding infection-related indicator and the confirmed epidemic, that is, the higher the importance of the corresponding infection-related indicator to disease prediction and evaluation, so as to further analyze the patients including the infection-related indicators, so as to more significantly characterize the adverse effects caused by the epidemic and conduct a more accurate epidemic impact analysis.

[0048] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the epidemiological diagnosis correlation includes:

[0049] Within the preset time period of the assessment area, the number of new people for each infection-related indicator is arranged in chronological order every day, and the sequence of new people of infection-related indicators corresponding to each infection-related indicator is determined; after the number of new people confirmed with the epidemic every day within the preset time period of the assessment area is arranged in chronological order, the sequence of new people confirmed with the epidemic is determined; based on the Pearson correlation coefficient between the sequence of new people confirmed with the epidemic and the sequence of new people of infection-related indicators, the epidemiological confirmed correlation of each infection-related indicator is determined. The larger the Pearson correlation coefficient is, the greater the correlation between the infection-related indicator and the epidemiological confirmed indicator in the dimension of positive correlation, then the greater the epidemiological confirmed correlation of the corresponding infection-related indicator, and the more accurate the characterization of the impact on the epidemic, so further, according to the epidemiological confirmed correlation, the infection-related indicators with relatively large correlation with the epidemiological confirmed indicators can be screened out, and used as epidemiological impact indicators for subsequent analysis. Preferably, in some possible implementation methods of the embodiments of the present invention, the process of obtaining the epidemiological impact indicator includes:

[0050] The infection-related indicators whose corresponding epidemic diagnosis correlation is greater than the preset correlation threshold are used as epidemic impact indicators. It should be noted that the Pearson correlation coefficient is a technical means well known to those skilled in the art and will not be further described here. In a specific implementation of an embodiment of the present invention, the preset correlation threshold is set to 0.53, which can be adjusted according to the specific implementation environment. It should be noted that when there is no infection-related indicator greater than the preset correlation threshold, the three largest infection-related indicators are used as epidemic impact indicators so that subsequent analysis can be carried out.

[0051] Step S103: Determine the degree of concern for each epidemic confirmed patient according to the overall magnitude of the epidemic confirmed correlation of each epidemic impact indicator corresponding to each epidemic confirmed patient; and select all key patients according to the degree of concern.

[0052] For each patient confirmed with an epidemic, when the corresponding epidemic impact indicators are generally large, then based on the characteristics that the epidemic impact indicators have a greater impact on the confirmed epidemic, the importance of the patient confirmed with an epidemic to disease prediction and evaluation should be higher, and the corresponding level of attention will be higher, so as to further screen out key patients based on the level of attention, and conduct a more accurate epidemic risk analysis based on the epidemic impact caused by the key patients.

[0053] Preferably, in a specific implementation of the embodiment of the present invention, the process of acquiring the degree of attention includes:

[0054] Obtain all the epidemic impact indicators matched to each epidemic confirmed patient; take the cumulative value of the epidemic confirmed correlation of all the epidemic impact indicators corresponding to each epidemic confirmed patient as the attention level of each epidemic confirmed patient. It should be noted that for each epidemic confirmed patient, if the corresponding epidemic impact indicator of the epidemic confirmed patient is confirmed within the preset time period, it means that the epidemic confirmed patient matches the corresponding epidemic impact indicator; for each epidemic confirmed patient, the more epidemic impact indicators it matches and the greater the epidemic confirmed correlation of each matched epidemic impact indicator, it means that the epidemic confirmed patient is more affected by the epidemic, and the higher the importance of disease prediction and evaluation should be, so a higher level of attention should be given.

[0055] In a specific implementation of the embodiment of the present invention, the process of obtaining the degree of attention is expressed by the formula: ;in, For the The level of attention paid to confirmed patients of the epidemic; For the The number of types of all epidemic impact indicators matched to each confirmed epidemic patient; For the The first patient matched with the Epidemiological confirmed correlation of 10 epidemiological impact indicators.

[0056] Preferably, in a specific implementation of the embodiment of the present invention, the process of acquiring the patient includes:

[0057] The mean of the corresponding attention levels of all confirmed epidemic patients is used as the key attention threshold; all confirmed epidemic patients whose corresponding attention levels are greater than the key attention threshold are used as key attention patients. For each confirmed epidemic patient, the greater the corresponding attention level, the higher the importance in disease prediction and assessment. In order to more accurately assess the epidemic risk, it is necessary to extract the key attention patients with a greater degree of attention for analysis. The mean can represent the overall numerical size of a set of data, so data greater than the mean can be considered to be relatively large as a whole; by screening out key attention patients for analysis separately, the amount of calculation can be reduced to a certain extent, while more accurately analyzing the spread of the epidemic.

[0058] Step S104: Determine the degree of transmission impact of each key patient according to the change in the number of confirmed epidemic cases in the region after the diagnosis of each key patient; determine the epidemic transmission performance of each key patient according to the epidemic cure status and transmission impact of the local hospitals in the region before the diagnosis of each key patient.

[0059] For each key patient, the path they traveled before being diagnosed will reflect the transmission characteristics of the epidemic, and the more significant the change in the number of confirmed cases of the epidemic after the key patient passes through these areas, the greater the corresponding degree of transmission impact; in addition, the medical level in the area passed by each key patient needs to be considered. If the medical level of each hospital in the passing area is relatively high, the control or management ability of the epidemic will be stronger to a certain extent, and the degree of impact will be lower. Therefore, the epidemic transmission performance of each key patient is determined based on the epidemic cure situation and transmission impact of the hospitals in the areas passed by each key patient before being diagnosed.

[0060] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the degree of propagation influence includes:

[0061] Obtain the moving path of each key patient within the preset first statistical time period before diagnosis; use the area corresponding to the preset radiation range of each path point on the moving path as the epidemic transmission area of ​​each path point; use all the infected radiation areas corresponding to all path points on the moving path as the overall transmission area of ​​each key patient; use the preset first statistical time period before diagnosis of each key patient and the preset second statistical time period after diagnosis as the transmission impact time period of each key patient; within the transmission impact time period, use the ratio between the number of new epidemics in the overall transmission area corresponding to each key patient and the total population in the overall transmission area as the transmission impact degree of each key patient.

[0062] First, considering that epidemics usually have an incubation period, the moving path of each key patient in the preset first statistical time period before diagnosis is first obtained for analysis, and in a specific implementation of the embodiment of the present invention, the preset first statistical time period and the preset second statistical time period are both set to 14 days, which can be adjusted according to the specific implementation environment, and are used to measure an empirical value of the incubation period. Then, on the corresponding moving path, the area corresponding to the preset radiation range of each path point is used as the epidemic transmission area or the control area, so as to determine the area affected by the incubation period of each key patient. After the key patient passes through the epidemic transmission area, the more new epidemics in the overall transmission area, the stronger the epidemic transmission ability, that is, the greater the degree of transmission impact. In a specific implementation of the embodiment of the present invention, the preset radiation range is set to an area consisting of a 150-meter circle centered on each path point, which can be adjusted according to the specific implementation environment. It should be noted that the moving path of each key patient can be obtained with the help of the positioning information authorized by the patient, and in other specific implementations of the embodiment of the present invention, the moving path of the patient can also be obtained in the form of a questionnaire. And the total number of people in the epidemic transmission area is the total number of people who have passed through the epidemic transmission area within a preset first statistical time period before the diagnosis of the key patients.

[0063] Preferably, in some possible implementations of the embodiments of the present invention, the process of acquiring the epidemic spread performance includes:

[0064] The epidemic cure index of each key patient is determined according to the overall situation of the epidemic treatment capacity of each hospital in the overall transmission area corresponding to each key patient. The degree of transmission impact represents the transmission capacity of the epidemic, and in the transmission area, that is, the overall transmission area, the stronger the epidemic treatment capacity of each corresponding hospital is, the more the risk of epidemic transmission will be contained and the harm caused will be reduced. Therefore, the epidemic transmission performance is further measured by calculating the epidemic cure index. Preferably, in some possible implementation methods of the embodiments of the present invention, the process of obtaining the epidemic cure index includes:

[0065] The product of the number of doctors treating epidemics and the number of medical facilities in each hospital is used as the epidemic treatment resource parameter of each hospital; the normalized value of the total number of people cured of epidemics in each hospital within the preset time period is used as the epidemic cure rate parameter of each hospital; the product of the epidemic treatment resource parameter and the epidemic cure rate parameter is used as the reference value of the treatment effect of each hospital on the epidemic; the average of the reference values ​​of the treatment effect of all hospitals on the epidemic in the overall transmission area is used as the epidemic cure index of each key patient. For each hospital, the more doctors and medical facilities it has, the more medical resources the corresponding hospital has, and the stronger its ability to respond to the epidemic. The more the total number of people cured of the epidemic in the preset time period, the higher the hospital's cure level for the epidemic; by combining the hospital's medical resources and the epidemic cure level, the treatment effect reference value obtained by the product of the epidemic treatment resource parameter and the epidemic cure rate parameter comprehensively determines the restriction capacity of each hospital on the epidemic. Therefore, in the overall transmission area, the higher the overall restriction capacity of all hospitals on the epidemic, that is, the larger the epidemic cure index, the smaller the degree of epidemic impact reflected by the key patients corresponding to the overall transmission area should be.

[0066] In a specific implementation of the embodiment of the present invention, the process of obtaining the epidemic cure index is expressed by the formula: ;in, For the An epidemiological cure index focusing on patients; For the the number of hospitals in the overall transmission area of ​​the focused patient; For the Focus on the overall spread of patients The total number of people cured of the epidemic in a hospital within a preset time period; For the Focus on the overall spread of patients The number of doctors treating epidemics in each hospital; For the Focus on the overall spread of patients The number of medical facilities for treating epidemic diseases in each hospital; For the Focus on the overall spread of patients Epidemiological cure rate parameters of each hospital; For the Focus on the overall spread of patients Epidemiological treatment resource parameters of each hospital; For the Focus on the overall spread of patients The reference value of treatment effect of each hospital; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment.

[0067] The larger the epidemic cure index is, the stronger the epidemic treatment capacity of each hospital in the overall transmission area is, and the smaller the impact of the epidemic will be. The degree of transmission impact represents the transmission capacity of the epidemic, which is obviously positively correlated with the impact of the epidemic. Therefore, the product of the negative correlation mapping value of the epidemic cure index and the degree of transmission impact is further used as the epidemic transmission performance of each key patient, and the epidemic transmission performance is used to represent the impact capacity of the epidemic of each key patient.

[0068] In a specific implementation of the embodiment of the present invention, the process of obtaining the epidemic transmission performance is expressed by the formula: ;in, For the A focus on the epidemiological performance of patients; For the An epidemiological cure index focusing on patients; For the The number of new cases in the overall transmission area of ​​the focused patients during the corresponding transmission impact period; For the The total population within the overall transmission area of ​​the focused patient; For the The focus is on the degree of transmission impact of patients; For the A negative correlation mapping value of the epidemic cure index focusing on patients; It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment.

[0069] Step S105: Determine the clustering reference value of each key patient according to the epidemic transmission performance, the degree of concern of each key patient and the corresponding epidemic cure period; perform epidemic risk assessment and prediction based on the clustering results after cluster analysis based on the clustering reference value.

[0070] After determining the epidemic transmission performance, combined with the characteristics that the length of the epidemic cure period will affect its risk, further on the basis of the epidemic transmission performance and the cure period, a comprehensive analysis is performed with the corresponding degree of attention for each key patient as a weight. After cluster analysis is performed according to the cluster reference value obtained in the analysis, a more accurate epidemic risk assessment and prediction is performed based on the cluster clusters with larger transmission performance in the clustering results. This allows the disease risk assessment and prediction to simultaneously combine the heterogeneity between individual cases, patient contact history, social behavioral characteristics such as residential population density, and the impact of hospital treatment level on disease transmission, making the epidemic risk assessment and prediction more effective.

[0071] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the cluster reference value includes:

[0072] The product of the degree of concern of each key patient, the epidemic cure cycle and the epidemic transmission performance is used as the clustering reference value for each key patient when participating in the clustering. For each key patient, the greater the corresponding degree of concern, the greater the reference value of the epidemic impact of the key patient, so it is used as a weight to participate in the product operation. The longer the epidemic cure cycle, the greater the impact of the epidemic in the time dimension; and the epidemic transmission performance can characterize the impact of the epidemic, so the epidemic cure cycle and the epidemic transmission performance are combined through the product; so that the obtained clustering reference value can comprehensively characterize the impact of the epidemic, thereby indirectly characterizing the epidemic risk. In a specific implementation method of an embodiment of the present invention, the process of obtaining the clustering reference value is expressed by the formula: ;in, For the The first one focuses on the cluster reference value when the patient participates in the cluster; For the A focus on the epidemiological performance of patients; For the The focus is on the patient's level of concern; For the It should be noted that the epidemic cure period in the embodiment of the present invention specifically refers to the number of days from the diagnosis of the epidemic to the cure of the epidemic for the key patients, and if it is less than one day, one day is taken as the epidemic cure period for analysis.

[0073] Preferably, in some possible implementations of the embodiments of the present invention, the process of performing epidemic risk assessment and prediction based on the clustering results after cluster analysis according to the clustering reference values ​​includes:

[0074] Perform k-means cluster analysis on the cluster reference values ​​of all key patients to obtain at least two clusters; take the mean of the cluster reference values ​​of all key patients in each cluster as the overall reference value of each cluster; take the ratio between the number of key patients in the cluster with the largest overall reference value and the total number of key patients in all clusters as the reference ratio; take the normalized value of the overall reference value corresponding to the cluster with the largest overall reference value as the risk reference value; take the product of the reference ratio and the risk reference value as the risk prediction assessment value. In a specific implementation of an embodiment of the present invention, the k value in the k-means cluster analysis is set to 4, that is, the number of clusters is set to 4, which can be adjusted according to the specific implementation environment, but needs to be greater than or equal to 2.

[0075] Since the cluster reference value can comprehensively characterize the impact of the epidemic and indirectly characterize the epidemic risk, cluster analysis will classify the key patients with similar epidemic impact into the same cluster. For each cluster obtained, the larger the overall reference value, the greater the epidemic impact caused by the key patients corresponding to the cluster; therefore, the more key patients in the cluster with the largest overall reference value, that is, the greater the proportion of the total number of key patients in all clusters, that is, the larger the reference ratio, the greater the harm or risk of the epidemic reflected in the epidemic impact; and since the overall reference value is the mean of the cluster reference values ​​of all key patients in the cluster, and the cluster reference value indirectly characterizes the epidemic risk; therefore, the larger the risk prediction assessment value obtained by multiplying the reference ratio and the risk reference value, the greater the risk of the epidemic assessed should be.

[0076] In a specific implementation of the embodiment of the present invention, the process of obtaining the risk prediction evaluation value is expressed by the formula: ;in, To estimate the risk; The number of patients in the cluster with the largest overall reference value; The cluster with the largest overall reference value The cluster reference value of the focused patients is also the overall reference value of the cluster; is the risk reference value; The total number of patients in all clusters. It is a linear normalization function. The implementer can adjust the normalization method according to the specific implementation environment.

[0077] Further, the epidemic is evaluated and predicted based on the risk prediction evaluation value. When the risk prediction evaluation value is greater than the preset first prediction threshold, the epidemic is evaluated as a high-risk epidemic; when the risk prediction evaluation value is less than or equal to the preset first prediction threshold and greater than the preset second prediction threshold, the epidemic is evaluated as a medium-risk epidemic; when the risk prediction evaluation value is less than or equal to the preset second prediction threshold, the epidemic is evaluated as a low-risk epidemic. In a specific implementation of an embodiment of the present invention, the first prediction threshold is set to 0.4, and the second prediction threshold is set to 0.2, which can be adjusted according to the specific implementation environment.

[0078] In summary, a method for assessing and predicting epidemic risk first calculates the correlation between each infection-related indicator and the epidemic diagnosis in the dimension of the impact of the infection-related indicator on the epidemic diagnosis, so as to screen out the epidemic impact indicators that will have a practical impact on the epidemic diagnosis; and further screen out the key patients with a greater impact on the spread of the epidemic according to the various epidemic impact indicators involved in the sick population; and then determine the epidemic transmission performance more accurately according to the regional transmission situation of the key patients before their diagnosis and the medical capacity of the region where they are located, so as to further conduct more accurate epidemic risk assessment and prediction according to the numerical distribution of the epidemic transmission performance of the key patients combined with cluster analysis, thereby solving the problem of low accuracy of epidemic risk assessment and prediction in the existing technology and making the effect of epidemic risk assessment and prediction better.

[0079] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0080] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for epidemic risk assessment and prediction, characterized in that: The method comprises: Obtain the number of newly confirmed cases of the epidemic and the number of newly confirmed cases of each infection-related indicator every day in the preset time period of the assessment area; Determine the epidemiological confirmed correlation of each infection-related indicator based on the temporal correlation between the number of newly confirmed cases of the epidemic and the number of newly confirmed cases of each infection-related indicator; and select the epidemiological impact indicators based on the epidemiological confirmed correlation; Determine the degree of concern for each epidemic confirmed patient according to the overall magnitude of the epidemic confirmed correlation of each epidemic impact indicator corresponding to each epidemic confirmed patient; and screen out all key patients according to the degree of concern; Determine the degree of transmission impact of each key patient based on the change in the number of confirmed cases of the epidemic in the region after the diagnosis of each key patient; determine the epidemiological transmission performance of each key patient based on the epidemiological recovery situation of the regional hospitals before the diagnosis of each key patient and the degree of transmission impact; Determine a cluster reference value for each key patient according to the epidemic transmission performance, the degree of concern of each key patient, and the corresponding epidemic cure period; perform epidemic risk assessment and prediction based on the cluster results after cluster analysis based on the cluster reference values; The process of performing epidemic risk assessment and prediction based on the clustering results after cluster analysis according to the clustering reference value includes: Perform k-means cluster analysis on the cluster reference values ​​of all key patients to obtain at least two clusters; take the mean of the cluster reference values ​​of all key patients in each cluster as the overall reference value of each cluster; take the ratio between the number of key patients in the cluster with the largest overall reference value and the total number of key patients in all clusters as the reference ratio; take the normalized value of the overall reference value corresponding to the cluster with the largest overall reference value as the risk reference value; take the product of the reference ratio and the risk reference value as the risk prediction assessment value; When the risk prediction evaluation value is greater than a preset first prediction threshold, the epidemic is evaluated as a high-risk epidemic; When the risk prediction evaluation value is less than or equal to a preset first prediction threshold and greater than a preset second prediction threshold, the epidemic is evaluated as a medium-risk epidemic; When the risk prediction evaluation value is less than or equal to a preset second prediction threshold, the epidemic is evaluated as a low-risk epidemic.

2. The method for epidemic risk assessment and prediction according to claim 1, characterized in that: The process of obtaining the epidemiological confirmed association includes: Arrange the number of new infections in each infection-related indicator in a time sequence within the preset time period of the assessment area, and determine the sequence of new infections in each infection-related indicator; Arrange the number of new infections in each infection-related indicator in a time sequence within the preset time period of the assessment area, and determine the sequence of new infections in each infection-related indicator; The epidemiological confirmation correlation of each infection-related indicator is determined based on the Pearson correlation coefficient between the series of newly confirmed epidemic numbers and the series of newly confirmed infection numbers of infection-related indicators.

3. The method for epidemic risk assessment and prediction according to claim 1, characterized in that: The process of obtaining the epidemic impact indicator includes: The infection-related indicators whose corresponding epidemic diagnosis correlation is greater than the preset correlation threshold are used as epidemic impact indicators.

4. The method for epidemic risk assessment and prediction according to claim 1, characterized in that: The process of obtaining the degree of attention includes: All the epidemic impact indicators matched to each epidemic confirmed patient are obtained; the cumulative value of the epidemic confirmed correlation of all the epidemic impact indicators corresponding to each epidemic confirmed patient is used as the attention level of each epidemic confirmed patient.

5. The method for epidemic risk assessment and prediction according to claim 1, characterized in that: The focus of the patient acquisition process includes: The average of the corresponding attention levels of all confirmed epidemic patients is taken as the key attention threshold; all confirmed epidemic patients whose corresponding attention levels are greater than the key attention threshold are taken as key attention patients.

6. The method for epidemic risk assessment and prediction according to claim 1, characterized in that: The process of obtaining the degree of communication impact includes: Obtain the movement path of each key patient within the preset first statistical time period before diagnosis; use the area corresponding to the preset radiation range of each path point on the movement path as the epidemic spread area of ​​each path point; use all the infection radiation areas corresponding to all the path points on the movement path as the overall spread area of ​​each key patient; use the preset first statistical time period before diagnosis and the preset second statistical time period after diagnosis of each key patient as the spread influence time period of each key patient; During the transmission impact time period, the ratio between the number of new epidemic people in the overall transmission area corresponding to each key patient and the total population in the overall transmission area is used as the transmission impact degree of each key patient.

7. The method for epidemic risk assessment and prediction according to claim 6, characterized in that: The process of obtaining the epidemic transmission performance includes: Determine the epidemiological cure index for each key patient based on the overall epidemiological treatment capacity of each hospital in the overall transmission area corresponding to each key patient; The product of the negative correlation mapping value of the epidemic cure index and the degree of transmission impact is used as the epidemic transmission performance of each key patient.

8. The method for epidemic risk assessment and prediction according to claim 7, characterized in that: The process of obtaining the epidemic cure index includes: The product of the number of doctors treating epidemics and the number of medical facilities in each hospital is used as the epidemic treatment resource parameter of each hospital; the normalized value of the total number of people cured of the epidemic in each hospital within a preset time period is used as the epidemic cure rate parameter of each hospital; the product of the epidemic treatment resource parameter and the epidemic cure rate parameter is used as the reference value of the treatment effect of each hospital on the epidemic; the average of the reference values ​​of the treatment effect of all hospitals on the epidemic in the overall transmission area is used as the epidemic cure index of each key patient.

9. The method for epidemic risk assessment and prediction according to claim 1, characterized in that: The process of obtaining the cluster reference value includes: The product of the degree of concern of each key patient, the epidemic cure cycle and the epidemic transmission performance is used as the clustering reference value when each key patient participates in clustering.

Images