Modeling method and system for spatiotemporal diffusion of infectious diseases within cities

Abstract

The present invention relates to a method for modeling the spatiotemporal diffusion of infectious diseases within a city, comprising: collecting infectious disease-related data and preprocessing the collected data, wherein the data includes: mobile phone positioning data; constructing a spatiotemporal diffusion model of infectious diseases within a city based on the preprocessed data and the pathogenesis and transmission mechanism of infectious diseases; and determining and estimating the parameters of the constructed spatiotemporal diffusion model of infectious diseases within a city. The present invention also relates to a system for modeling the spatiotemporal diffusion of infectious diseases within a city. The present invention can construct a fine-scale spatiotemporal diffusion model of infectious diseases within a city to achieve prediction, early warning, and spatially precise simulation of prevention and control measures.

Description

Technical Field

[0001] The invention relates to a method and system for modeling the spatiotemporal diffusion of infectious diseases within a city. Background Art

[0002] Thanks to the greatly improved convenience of travel, the flow of population within cities has a strong dynamic and time-varying characteristic, which accelerates the spread of infectious diseases, especially respiratory infectious diseases in cities. The historical and current experience of infectious diseases shows that once such infectious diseases break out in megacities with dense populations, developed transportation, and high-speed social operations, the consequences are particularly serious. Modeling the spatiotemporal diffusion of infectious diseases through mathematical and computer means helps to understand the spread of infectious diseases, predict their diffusion trends, and simulate and evaluate the effects of various prevention and control measures.

[0003] Among them, the transmission dynamics model (here refers to the compartment model) is still the mainstream model for modeling the spatiotemporal spread of infectious diseases. The simulation and effect evaluation of various prevention and control measures such as travel restrictions and social distancing are also mainly carried out under this type of model system.

[0004] Typical representatives of the transmission dynamics model are the SIR and SIS models. On this basis, according to the different disease pathogenesis and transmission modes, the SIRS model, SEIR model, SIS model with birth and death migration, and model with age structure have been developed. The above models can be constructed for a single geographical unit, such as a country, province, or city, or the population of multiple geographical units can be regarded as different populations, and a population model based on transmission dynamics (metapopulation model) can be constructed by combining the population flow between populations. Cao Zhidong et al. (2009) constructed a dynamic model reflecting the SARS transmission process, optimally estimated important epidemiological parameters such as the incubation period and infection period, and simulated and evaluated the effects and impacts of four types of intervention measures such as vaccination rate. Li et al. (2020) integrated Internet population migration data and constructed a SEIR-based population model for 375 Chinese cities. Lai et al. (2020) constructed a SEIR-based population model for 340 Chinese cities based on Internet migration data. Zhou et al. (2020) constructed a SEIR-based population model for the 10 administrative districts of Shenzhen based on mobile phone positioning data, and systematically and quantitatively evaluated the effectiveness of four types of travel restriction measures (i.e., reducing travel between all regions, blocking high-risk areas, restricting travel within regions, and restricting travel of infected people).

[0005] Although the above-mentioned transmission dynamics models have the advantages of clear mechanisms and high modeling and operation efficiency, current modeling and prevention and control research based on transmission dynamics is mainly at the national / state / provincial / city levels, and there has not yet been in-depth modeling of the spatiotemporal spread of infectious diseases at the fine scale within the city. Summary of the invention

[0006] In view of this, it is necessary to provide a method and system for modeling the spatiotemporal diffusion of infectious diseases within cities, which can construct a population model based on transmission dynamics at a fine spatial granularity within the city.

[0007] The present invention provides a method for modeling the spatiotemporal diffusion of infectious diseases within a city, the method comprising the following steps: a. collecting infectious disease-related data and preprocessing the collected data, the data comprising: mobile phone positioning data; b. constructing a spatiotemporal diffusion model of infectious diseases within a city based on the preprocessed data and the onset and propagation mechanism of infectious diseases; c. determining and estimating the parameters of the constructed spatiotemporal diffusion model of infectious diseases within the city.

[0008] Furthermore, the preprocessing includes: processing the collected mobile phone positioning data into grid data, wherein the grid data includes the number of residents in the grid, the population flow between grids throughout the day, and the number of residents flowing out of the grid throughout the day.

[0009] Furthermore, the step b comprises:

[0010] Considering the spatiotemporal impact of population mobility within the city on disease transmission, a population model based on susceptible-latent-infected-removed is constructed according to the course characteristics and transmission mechanism of infectious diseases.

[0011] Furthermore, the spatiotemporal diffusion model of infectious diseases within the city is constructed as follows:

[0012]

[0013]

[0014]

[0015]

[0016]

[0017] The suffixes 1 and 0 represent overt infection and latent infection, respectively. That is, the total residential population within the grid remains unchanged.

[0018]

[0019]

[0020] new_infectious1=σδL n

[0021] new_infectious0=(1-σ)δLn

[0022]

[0023]

[0024] in:

[0025] It indicates the number of explicit infected persons flowing from other grids into this grid;

[0026] It indicates the number of latently infected people from other grids flowing into this grid;

[0027] It indicates the number of obvious infected persons encountered by the residents of this grid when going out;

[0028] It indicates the number of latently infected people encountered by residents of this grid when going out;

[0029] f n is the number of people who flow out of the grid throughout the day among the residents of grid n, obtained from the data preprocessed in step S1;

[0030] h mn Represents the number of people who move from grid m to grid n throughout the day, obtained from the data preprocessed in step S1.

[0031] Furthermore, the parameters of the spatiotemporal diffusion model of infectious diseases within the city include:

[0032] β is the effective infection rate of the overtly infected, which needs to be estimated;

[0033] ε is the ratio of the effective infection rate of latently infected persons to that of overtly infected persons, which is taken according to the literature;

[0034] 1 / δ is the latent period, which is taken according to the literature;

[0035] σ is the proportion of overtly infected people, and its value is taken according to the literature;

[0036] 1 / γ1 is the infectious period of an overtly infected person, and the value is taken according to the literature;

[0037] 1 / γ0 is the infection period of latently infected persons, and its value is taken according to the literature.

[0038] Taking different values ​​of β, we simulated the daily new case curve corresponding to a given number of seed cases, and used the R0 toolkit in R language to estimate R0 from each daily new case curve using the exponential growth method. Then we fitted the relationship between β and R0 and obtained the effective infection rate β corresponding to the basic reproduction number R0 of the infectious disease.

[0039] The present invention provides a system for modeling the spatiotemporal diffusion of infectious diseases within a city, the system comprising a data collection and preprocessing unit, a model building unit, and a parameter determination and estimation unit, wherein: the data collection and preprocessing unit is used to collect infectious disease-related data and preprocess the collected data, the data comprising: mobile phone positioning data; the model building unit is used to construct a spatiotemporal diffusion model of infectious diseases within a city based on the preprocessed data and the onset and propagation mechanism of infectious diseases; the parameter determination and estimation unit is used to determine and estimate the parameters of the constructed spatiotemporal diffusion model of infectious diseases within the city.

[0040] Furthermore, the preprocessing includes: processing the collected mobile phone positioning data into grid data, wherein the grid data includes the number of residents in the grid, the population flow between grids throughout the day, and the number of residents flowing out of the grid throughout the day.

[0041] Furthermore, the model building unit is specifically used for:

[0042] Considering the spatiotemporal impact of population mobility within the city on disease transmission, a population model based on susceptible-latent-infected-removed is constructed according to the course characteristics and transmission mechanism of infectious diseases.

[0043] Furthermore, the model building unit is specifically used for:

[0044] Furthermore, the spatiotemporal diffusion model of infectious diseases within the city is constructed as follows:

[0045]

[0046]

[0047]

[0048]

[0049]

[0050] The suffixes 1 and 0 represent overt infection and latent infection, respectively. That is, the total residential population within the grid remains unchanged.

[0051]

[0052]

[0053] new_infectious1=σδL n

[0054] new_infectious0=(1-σ)δLn

[0055]

[0056]

[0057] in:

[0058] It indicates the number of explicit infected persons flowing from other grids into this grid;

[0059] It indicates the number of latently infected people from other grids flowing into this grid;

[0060] It indicates the number of obvious infected persons encountered by the residents of this grid when going out;

[0061] It indicates the number of latently infected people encountered by residents of this grid when going out;

[0062] f n is the number of people who flow out of the grid throughout the day among the resident population of grid n, obtained from the data preprocessed by the data acquisition and preprocessing unit;

[0063] h mn It represents the number of people who move from grid m to grid n throughout the day, and is obtained from the data preprocessed by the data acquisition and preprocessing unit.

[0064] Furthermore, the parameters of the spatiotemporal diffusion model of infectious diseases within the city include:

[0065] β is the effective infection rate of the overtly infected, which needs to be estimated;

[0066] ε is the ratio of the effective infection rate of latently infected persons to that of overtly infected persons, which is taken according to the literature;

[0067] 1 / δ is the latent period, which is taken according to the literature;

[0068] σ is the proportion of overtly infected people, and its value is taken according to the literature;

[0069] 1 / γ1 is the infectious period of an overtly infected person, and the value is taken according to the literature;

[0070] 1 / γ0 is the infection period of latently infected persons, and its value is taken according to the literature.

[0071] Taking different values ​​of β, we simulated the daily new case curve corresponding to a given number of seed cases, and used the R0 toolkit in R language to estimate R0 from each daily new case curve using the exponential growth method. Then we fitted the relationship between β and R0 and obtained the effective infection rate β corresponding to the basic reproduction number R0 of the infectious disease.

[0072] The spatiotemporal diffusion modeling method and system of infectious diseases within cities of the present invention fully consider the population mobility characteristics between areas within cities, and construct a fine-scale spatiotemporal diffusion model of infectious diseases within cities to achieve prediction, early warning and spatially precise simulation of prevention and control measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flow chart of the spatiotemporal diffusion modeling method of infectious diseases within a city according to the present invention;

[0074] Figure 2 A schematic diagram of the COVID-19 infection status and evolution process in a single population provided by an embodiment of the present invention;

[0075] Figure 3 A schematic diagram of the infection relationship within and between grids caused by population mobility provided by an embodiment of the present invention;

[0076] Figure 4 A schematic diagram of the fitting relationship between the effective infection rate β and the R0 estimated value provided in an embodiment of the present invention;

[0077] Figure 5 This is a hardware architecture diagram of the spatiotemporal diffusion modeling system for infectious diseases within cities according to the present invention. DETAILED DESCRIPTION

[0078] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0079] This example uses a city and the infectious disease COVID-19 as an example to illustrate:

[0080] See also Figure 1 The figure shows the operation flow chart of a preferred embodiment of the method for modeling the spatiotemporal diffusion of infectious diseases within a city according to the present invention.

[0081] Step S1, collecting infectious disease related data and preprocessing the collected data. The data includes: mobile phone positioning data. The preprocessing includes: processing the collected mobile phone positioning data into grid data, and the grid data includes the number of grid residents, the population flow between grids throughout the day, and the number of residents who flow out of the grid throughout the day.

[0082] Specifically:

[0083] (1) Mobile phone positioning data: The mobile phone positioning data used in this embodiment is collected from China United Network Communications Group Co., Ltd. (hereinafter referred to as China Unicom), and entrusted to Smart Footprint Data Technology Co., Ltd. (hereinafter referred to as Smart Footprint) to process it into grid data based on 500m×500m. The grid data includes the number of residents in the grid, the population flow between grids throughout the day, and the number of residents who flow out of the grid throughout the day. Since China Unicom mobile phone users only cover part of the population, the population distribution and population flow data are all full population sample data obtained by Smart Footprint based on China Unicom user data.

[0084] ① Number of residents in each grid: This embodiment is based on the observation records for the whole month of December 2019. The grid where the user is observed to stay the longest at night (21:00-8:00 the next day) is taken as his / her residence. On this basis, the number of residents in each grid is aggregated.

[0085] ② Based on ①, this embodiment also counts the number of residents who stay in each grid continuously throughout the day (ie, do not leave the grid throughout the day).

[0086] ③ Inter-grid population flow throughout the day: Identify the user's cross-grid location movement behavior. If the length of stay in the grid before and after the displacement exceeds 30 minutes, the displacement is a valid trip. Aggregate all valid trips of all users on a typical day, such as 2019.12.18, to obtain the population flow between grids throughout the day. Based on the population flow between grids throughout the day, further calculate the population inflow and outflow of each grid throughout the day. Among them, the population flow between grids throughout the day is the OD matrix (Origin-Destination, source-destination matrix).

[0087] It should be noted that mobile phone positioning data is only one data source for obtaining higher-quality data on the number of resident populations in spatial units and the flow of population between spatial units in this embodiment; if there are other data sources that can obtain the number of resident populations in grids or other types of spatial units and the flow of population between spatial units, they can also be used as substitutes and also fall within the scope of protection of the present invention.

[0088] Step S2, constructing a spatiotemporal diffusion model of infectious diseases within the city based on the preprocessed data and the pathogenesis and transmission mechanism of infectious diseases.

[0089] This embodiment uses a 500-meter grid as the population division unit, fully considers the spatiotemporal impact of urban internal population mobility on disease transmission, and constructs a population model based on Susceptible-Latent-Infectious-Removed (SLIR) according to the course characteristics and transmission mechanism of COVID-19.

[0090] Figure 2 The COVID-19 infection status and evolution process in a single population are shown. Based on whether the population is invaded by pathogens and whether it is contagious, this example divides the population living in grid n into N n Divided into susceptible groups S n , latent period population L n , Infectious population (including overtly infected persons I 1 and latently infected persons I 0 ) and remove the crowd R n .

[0091] Among them, the latent period refers to the period from the invasion of the pathogen into the body to the onset of infectiousness; the communicable period refers to the period from the onset of infectiousness to recovery, isolation or death (ie, with the ability to spread). The infectious population, that is, the infected person, is divided into explicit infected persons and latent infected persons. Among them, the explicit infected person develops the disease 1-2 days after entering the infectious period, and 1.5 days is taken in this embodiment. If it is not discovered in advance through close contact tracking and other means, under natural conditions (corresponding to R0≈2.3), the explicit infected person will be hospitalized for isolation about 6 days after the onset of the disease, so the infectious period is about 1.5+6 days; the infectious period of the latent infected person is about 10 days.

[0092] like Figure 3 As shown in the figure, due to the population flow between grids, the susceptible people living in each grid will not only be infected by the infected people in the grid, but also by the infected people flowing into the grid from other grids, and by the infected people in other grids when going out. The infected people in other grids include the local infected people in the grid and the infected people flowing into the grid from other grids.

[0093] Based on the above pathogenesis and transmission mechanism, the model constructed in this embodiment is as follows, wherein the suffix symbols 1 and 0 represent explicit infection and latent infection respectively:

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] in, That is, the total residential population within the grid remains unchanged.

[0100]

[0101]

[0102] new_infectious1=σδL n

[0103] new_infectious0=(1-σ)δL n

[0104]

[0105]

[0106] in:

[0107] It indicates the number of explicit infected persons flowing from other grids into this grid;

[0108] It indicates the number of latently infected people from other grids flowing into this grid;

[0109] Indicates that the residents of this grid are outside

[0110] The number of obvious infected people encountered;

[0111] It indicates the number of latently infected people encountered by residents of this grid when going out;

[0112] f n is the number of people who have left the grid throughout the day among the residents of grid n, obtained from the data pre-processed in step S1;

[0113] h mn Represents the number of people who move from grid m to grid n throughout the day, obtained from the data preprocessed in step S1.

[0114] Step S3, determining and estimating the parameters of the constructed spatiotemporal diffusion model of infectious diseases within the city.

[0115] Specifically:

[0116] The main parameters involved in the spatiotemporal diffusion model of infectious diseases within the city include:

[0117] β is the effective infection rate of an overtly infected person, that is, the average number of people infected by an overtly infected person per day, to be estimated;

[0118] ε is the ratio of the effective infection rate of latently infected persons to that of overtly infected persons, referring to the national CDC report, and is taken as 0.12 in this embodiment;

[0119] 1 / δ is the incubation period, and in this embodiment, 5.2 days (incubation period) - 1.5 days = 4.7 days;

[0120] σ is the proportion of overtly infected people, which is 75% in this embodiment;

[0121] 1 / γ1 is the infectious period of the dominant infected person. According to the early reports of the epidemic (the average time from onset to hospitalization is about 6 days), under natural conditions, this example takes 1.5 days + 6 days = 7.5 days;

[0122] 1 / γ0 is the infection period of a latently infected person, which is 10 days under natural conditions.

[0123] It is worth noting that the above parameters may be updated based on the development of COVID-19 and the latest research results.

[0124] The only parameter that needs to be estimated in the spatiotemporal diffusion model of infectious diseases within the city is the effective infection rate β of the dominant infected person: given a certain number of seed cases (the position of the seed cases can be set according to the population distribution), different values ​​of β are taken to simulate the corresponding daily new curves, and the R0 toolkit of the R language is used to estimate R0 from each daily new curve using the exponential growth method (EG), and then fit the relationship between β and R0 to obtain the effective infection rate β corresponding to the basic reproduction number R0≈2.3 of COVID-19. In this embodiment, given different effective infection rates β, 500 initial seed cases are set, and the daily new curves of different cities are simulated through the spatiotemporal diffusion model of infectious diseases within the city, and then the corresponding R0 is estimated. Figure 4 The relationship between the effective infection rate β and the estimated value of R0 is shown, and the two have a good linear fit relationship. Given R0 = 2.3, the corresponding effective infection rate β value can be obtained as 0.237.

[0125] Substitute the above parameters into the model constructed in step S2, that is, completely construct the spatiotemporal diffusion model of COVID-19 at a 500-meter grid scale in a certain city. The model based on the above parameters conforms to the basic law of COVID-19 transmission under natural conditions. Given the initial case location, a spatiotemporal dynamic forecast of the epidemic development is made. In addition, based on this model, by adjusting the effective infection rate β of the overtly infected person, the time from onset to hospital admission (corresponding to the infection period 1 / γ1 of the overtly infected person), the population flow between grids, etc., it is possible to simulate and evaluate the effectiveness of prevention and control measures such as wearing masks / sanitation and disinfection, isolation and treatment speed, and population flow control.

[0126] See also Figure 5 , which is a hardware architecture diagram of the spatiotemporal diffusion modeling system 10 for infectious diseases within a city according to the present invention. The system comprises: a data acquisition and preprocessing unit 101, a model building unit 102, and a parameter determination and estimation unit 103.

[0127] The data collection and preprocessing unit 101 is used to collect infectious disease related data and preprocess the collected data. The data includes: mobile phone positioning data. The preprocessing includes: processing the collected mobile phone positioning data into grid data, and the grid data includes the number of grid residents and the population flow between grids. Specifically:

[0128] (1) Mobile phone positioning data: The mobile phone positioning data used by the data collection and preprocessing unit 101 is collected from China United Network Communications Group Co., Ltd. (hereinafter referred to as China Unicom), and is processed into grid data based on 500m×500m by Smart Footprint Data Technology Co., Ltd. (hereinafter referred to as Smart Footprint). The grid data includes the number of people living in the grid and the flow of people between grids. Since China Unicom mobile phone users only cover part of the population, the population distribution and population flow data are all full population sample data obtained by Smart Footprint based on the expansion of China Unicom user data.

[0129] ① Number of residents in each grid: This embodiment is based on the observation records for the whole month of December 2019. The grid where the user is observed to stay the longest at night (21:00-8:00 the next day) is taken as his / her residence. On this basis, the number of residents in each grid is aggregated.

[0130] ② The data collection and preprocessing unit 101 also counts the number of residents who stay in each grid continuously throughout the day (ie, do not leave the grid throughout the day).

[0131] ③ Population flow between grids: Identify the user's cross-grid location movement behavior. If the length of stay in the grid before and after the displacement exceeds 30 minutes, the displacement is a valid trip. Aggregate all valid trips of all users on a typical day, such as 2019.12.18, to obtain the population flow between grids throughout the day. Based on the population flow between grids throughout the day, further calculate the population inflow and outflow of each grid throughout the day. Among them, the population flow between grids throughout the day is the OD matrix (OriginationDestination, source-destination matrix).

[0132] It should be noted that mobile phone positioning data is only one data source for obtaining higher-quality data on the number of resident populations in spatial units and the flow of population between spatial units in this embodiment; if there are other data sources that can obtain the number of resident populations in grids or other types of spatial units and the flow of population between spatial units, they can also be used as substitutes and also fall within the scope of protection of the present invention.

[0133] The model building unit 102 is used to build a spatiotemporal diffusion model of infectious diseases within the city based on the pre-processed data and the pathogenesis and transmission mechanism of infectious diseases. Specifically:

[0134] This embodiment uses a 500-meter grid as the population division unit, fully considers the spatiotemporal impact of urban internal population mobility on disease transmission, and constructs a population model based on Susceptible-Latent-Infectious-Removed (SLIR) according to the course characteristics and transmission mechanism of COVID-19.

[0135] Figure 2 The COVID-19 infection status and evolution process in a single population are shown. Based on whether the population is invaded by pathogens and whether it is contagious, this example divides the population living in grid n into N n Divided into susceptible groups S n , latent period population L n , Infectious population (including overtly infected persons I 1 and latently infected persons I 0 ) and remove the crowd R n .

[0136] Among them, the latent period refers to the period from the invasion of the pathogen into the body to the onset of infectiousness; the communicable period refers to the period from the onset of infectiousness to recovery, isolation or death (ie, with the ability to spread). The infectious population, that is, the infected person, is divided into explicit infected persons and latent infected persons. Among them, the explicit infected person develops the disease 1-2 days after entering the infectious period, and 1.5 days is taken in this embodiment. If it is not discovered in advance through close contact tracking and other means, under natural conditions (corresponding to R0≈2.3), the explicit infected person will be hospitalized for isolation about 6 days after the onset of the disease, so the infectious period is about 1.5+6 days; the infectious period of the latent infected person is about 10 days.

[0137] like Figure 3 As shown in the figure, due to the population flow between grids, the susceptible people living in each grid will not only be infected by the infected people in the grid, but also by the infected people flowing into the grid from other grids, and by the infected people in other grids when going out. The infected people in other grids include the local infected people in the grid and the infected people flowing into the grid from other grids.

[0138] Based on the above pathogenesis and transmission mechanism, the model constructed in this embodiment is as follows, wherein the suffix symbols 1 and 0 represent explicit infection and latent infection respectively:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144] in, That is, the total residential population within the grid remains unchanged.

[0145]

[0146]

[0147] new_infectious1=σδL n

[0148] new_infectious0=(1-σ)δL n

[0149]

[0150]

[0151] in:

[0152] It indicates the number of explicit infected persons flowing from other grids into this grid;

[0153] It indicates the number of latently infected people from other grids flowing into this grid;

[0154] It indicates the number of obvious infected persons encountered by the residents of this grid when going out;

[0155] It indicates the number of latently infected people encountered by residents of this grid when going out;

[0156] f n is the number of people who have flowed out of the grid throughout the day among the resident population of grid n, obtained from the data preprocessed by the data acquisition and preprocessing unit 101;

[0157] h mn Indicates the number of people from grid m to grid n throughout the day, obtained from the data preprocessed by the data collection and preprocessing unit 101.

[0158] The parameter determination and estimation unit 103 is used to determine and estimate the parameters of the constructed spatiotemporal diffusion model of infectious diseases within the city. Specifically:

[0159] The main parameters involved in the spatiotemporal diffusion model of infectious diseases within the city include:

[0160] β is the effective infection rate of an overtly infected person, that is, the average number of people infected by an overtly infected person per day, to be estimated;

[0161] ε is the ratio of the effective infection rate of latently infected persons to that of overtly infected persons, referring to the national CDC report, and is taken as 0.12 in this embodiment;

[0162] 1 / δ is the incubation period, and in this embodiment, 5.2 days (incubation period) - 1.5 days = 4.7 days;

[0163] σ is the proportion of overtly infected people, which is 75% in this embodiment;

[0164] 1 / γ1 is the infectious period of the dominant infected person. According to the early reports of the epidemic (the average time from onset to hospitalization is about 6 days), under natural conditions, this example takes 1.5 days + 6 days = 7.5 days;

[0165] 1 / γ0 is the infection period of a latently infected person, which is 10 days under natural conditions.

[0166] It is worth noting that the above parameters may be updated based on the development of COVID-19 and the latest research results.

[0167] The only parameter that needs to be estimated in the urban intra-city infectious disease spatiotemporal diffusion model is the effective infection rate β of the overtly infected persons: given a certain number of seed cases (the positions of the seed cases can be set according to the population distribution), different values ​​of β are taken to simulate the corresponding daily new curves, and R0 is estimated from each daily new curve using the R0 toolkit of the R language using the exponential growth method (EG), and then the relationship between β and R0 is fitted to obtain the effective infection rate β corresponding to the basic reproduction number of COVID-19 R0≈2.3.

[0168] In this embodiment, given different effective infection rates β, 500 initial seed cases are set, and the daily new case curves of different cities are simulated through the spatiotemporal diffusion model of infectious diseases within the city, and then the corresponding R0 is estimated. Figure 4 The relationship between the effective infection rate β and the estimated value of R0 is shown, and the two have a good linear fit relationship. Given R0 = 2.3, the corresponding effective infection rate β value can be obtained as 0.237.

[0169] Substituting the above parameters into the model constructed by the model construction unit 102, a COVID-19 spatiotemporal diffusion model at a 500-meter grid scale in a certain city is completely constructed. The model based on the above parameters conforms to the basic law of COVID-19 transmission under natural conditions. Given the initial case location, a spatiotemporal dynamic forecast of the epidemic development is made. In addition, based on this model, by adjusting the effective infection rate β of the dominant infected person, the time from onset to hospital admission (corresponding to the infection period 1 / γ1 of the dominant infected person), the population flow between grids, etc., it is possible to simulate and evaluate the effectiveness of prevention and control measures such as wearing masks / sanitation and disinfection, isolation and treatment speed, and population flow control.

[0170] The method and system for modeling the spatiotemporal diffusion of infectious diseases within cities of the present invention incorporate population mobility data and construct a population model based on transmission dynamics at a fine spatial granularity (500-meter grid level) within the city to achieve spatially refined spatiotemporal diffusion modeling of infectious diseases, especially COVID-19, and serve the spatiotemporal prediction, early warning, and prevention and control simulation of infectious diseases, especially COVID-19.

[0171] The spread of infectious diseases is a highly nonlinear spatiotemporal process, and it is difficult to use intuition to judge future development trends. Constructing an infectious disease model is not only to approximate a specific truth, but more to systematically and scientifically study the development trend of this complex process under different hypothetical intervention scenarios, providing policy makers, clinicians, and public health practitioners with crucial planning tools. The spatiotemporal diffusion model of infectious diseases within cities constructed by the present invention can be effectively applied to the prediction of spatiotemporal trends in epidemics and the simulation of prevention and control measures while ensuring that the data is authentic and reliable and the model expression form and parameters are reasonable.

[0172] Although the present invention has been described with reference to the current preferred embodiments, those skilled in the art should understand that the above-mentioned preferred embodiments are only used to illustrate the present invention and are not used to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for modeling the spatiotemporal diffusion of infectious diseases within a city, characterized in that: The method comprises the following steps: a. Collecting infectious disease related data and preprocessing the collected data, the data including: mobile phone positioning data; b. Construct a spatiotemporal diffusion model of infectious diseases within the city based on the preprocessed data and the pathogenesis and transmission mechanism of infectious diseases; c. Determine and estimate the parameters of the constructed spatiotemporal diffusion model of infectious diseases within the city; The constructed spatiotemporal diffusion model of infectious diseases within the city is: The suffixes 1 and 0 represent overt infection and latent infection, respectively. That is, the total residential population within the grid remains unchanged; new_infectious1=σδL n new_infectious0=(1-σ)δL n in: It indicates the number of explicit infected persons flowing from other grids into this grid; It indicates the number of latently infected people from other grids flowing into this grid; It indicates the number of obvious infected persons encountered by the residents of this grid when going out; It indicates the number of latently infected people encountered by residents of this grid when going out; β is the effective infection rate of the dominant infected person; ε is the ratio of the effective infection rate of latently infected persons to that of overtly infected persons; 1 / δ is the latent period; σ is the proportion of overtly infected people; 1 / γ1 is the infectious period of the dominant infected person; 1 / γ0 is the infection period of latently infected persons; f n is the number of people who have left the grid throughout the day among the residents of grid n, obtained from the data preprocessed in step a; h mn Represents the number of people who move from grid m to grid n throughout the day, obtained from the preprocessed data in step a.

2. The method for modeling the spatiotemporal diffusion of infectious diseases within a city as claimed in claim 1, characterized in that: The preprocessing includes: processing the collected mobile phone positioning data into grid data, wherein the grid data includes the number of grid residents, the population flow between grids throughout the day, and the number of residents flowing out of the grid throughout the day.

3. The method for modeling the spatiotemporal diffusion of infectious diseases within a city as claimed in claim 2, characterized in that: The step b comprises: Considering the spatiotemporal impact of population mobility within the city on disease transmission, a population model based on susceptible S-latent L-infected I-removed R is constructed according to the course characteristics and transmission mechanism of infectious diseases.

4. The method for modeling the spatiotemporal diffusion of infectious diseases within a city as claimed in claim 3, characterized in that: The parameters of the spatiotemporal diffusion model of infectious diseases within the city include: β needs to be estimated; ε is taken according to the literature; 1 / δ is taken according to the literature; σ is taken according to the literature; 1 / γ1 is taken according to the literature; 1 / γ0 is taken according to the literature; Taking different values ​​of β, simulating the daily new curve corresponding to a given number of seed cases, using the R0 toolkit of R language, using the exponential growth method to estimate R0 from each daily new curve, and then fitting the relationship between β and R0, to obtain the effective infection rate β corresponding to the basic reproduction number R0 of the infectious disease; wherein the seed cases are set according to the population distribution location and number.

5. A spatiotemporal diffusion modeling system for infectious diseases within cities, characterized in that: The system includes a data acquisition and preprocessing unit, a model building unit, and a parameter determination and estimation unit, wherein: The data collection and preprocessing unit is used to collect infectious disease related data and preprocess the collected data, and the data includes: mobile phone positioning data; The model building unit is used to build a spatiotemporal diffusion model of infectious diseases within the city based on the preprocessed data and the pathogenesis and spread mechanism of infectious diseases; The parameter determination and estimation unit is used to determine and estimate the parameters of the constructed spatiotemporal diffusion model of infectious diseases within the city; The model building unit is specifically used for: The constructed spatiotemporal diffusion model of infectious diseases within the city is: The suffixes 1 and 0 represent overt infection and latent infection, respectively. That is, the total residential population within the grid remains unchanged; new_infectious1=σδL n new_infectious0=(1-σ)δL n in: It indicates the number of explicit infected persons flowing from other grids into this grid; It indicates the number of latently infected people from other grids flowing into this grid; It indicates the number of obvious infected persons encountered by the residents of this grid when going out; It indicates the number of latently infected people encountered by residents of this grid when going out; β is the effective infection rate of the dominant infected person; ε is the ratio of the effective infection rate of latently infected persons to that of overtly infected persons; 1 / δ is the latent period; σ is the proportion of overtly infected people; 1 / γ1 is the infectious period of the dominant infected person; 1 / γ0 is the infection period of latently infected persons; f n is the number of people who flow out of the grid throughout the day among the resident population of grid n, obtained from the data preprocessed by the data acquisition and preprocessing unit; h mn It represents the number of people who move from grid m to grid n throughout the day, and is obtained from the data preprocessed by the data acquisition and preprocessing unit.

6. The spatiotemporal diffusion modeling system for infectious diseases within a city as claimed in claim 5, characterized in that: The preprocessing includes: processing the collected mobile phone positioning data into grid data, wherein the grid data includes the number of grid residents, the population flow between grids throughout the day, and the number of residents flowing out of the grid throughout the day.

7. The spatiotemporal diffusion modeling system for infectious diseases within a city as claimed in claim 6, characterized in that: The model building unit is specifically used for: Considering the spatiotemporal impact of population mobility within the city on disease transmission, a population model based on susceptible S-latent L-infected I-removed R is constructed according to the course characteristics and transmission mechanism of infectious diseases.

8. The spatiotemporal diffusion modeling system for infectious diseases within a city as claimed in claim 7, characterized in that: The parameters of the spatiotemporal diffusion model of infectious diseases within the city include: β needs to be estimated; ε is taken according to the literature; 1 / δ is taken according to the literature; σ is taken according to the literature; 1 / γ1 is taken according to the literature; 1 / γ0 is taken according to the literature; Taking different values ​​of β, simulating the daily new curve corresponding to a given number of seed cases, using the R0 toolkit of R language, using the exponential growth method to estimate R0 from each daily new curve, and then fitting the relationship between β and R0, to obtain the effective infection rate β corresponding to the basic reproduction number R0 of the infectious disease; wherein the seed cases are set according to the population distribution location and number.

Images