A method, system, terminal and storage medium for simulating the spread of infectious diseases

Abstract

The present invention relates to a method, system, terminal and storage medium for simulating the spread of infectious diseases. It includes: constructing an individual mobility model based on multi-source data in a prevention and control area; constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model; modeling non-drug intervention measures implemented in the prevention and control area; constructing an individual-based infectious disease propagation model based on the dynamic contact network and the intervention measures modeling, and simulating the infectious disease propagation intensity in the prevention and control area through the infectious disease propagation model. The present invention simulates the infectious disease propagation intensity in the prevention and control area through the infectious disease propagation model, and can support the fusion modeling of various non-drug intervention measures, which is helpful to evaluate the individual's compliance with various intervention measures, and can accurately evaluate the effectiveness of various non-drug intervention measures in inhibiting the spread of infectious diseases, thereby systematically and scientifically assisting in responding to future epidemic prevention and control.

Description

Technical Field

[0001] The present invention belongs to the field of infectious disease computing technology, and in particular relates to an infectious disease propagation simulation method, system, terminal and storage medium. Background Art

[0002] The novel coronavirus (COVID-19) is a virus that can be transmitted through respiratory droplets. After an individual is infected with the COVID-19 virus, the main symptoms are fever, fatigue, dry cough, and other symptoms, with radiological features of pneumonia. According to the latest real-time statistics from the World Health Organization, as of December 9, 2020, Beijing time, there were 67,780,361 confirmed cases of COVID-19 and 1,551,214 deaths worldwide. The dynamic changes in human activities in time and space are the key driving factors for the spread of infectious diseases between people. Most countries in the world have begun to implement various non-drug intervention measures to respond to the spread of COVID-19 in their countries, including city closures, restrictions on going out, and prohibitions on gatherings. These measures have seriously interfered with people's normal life needs and caused serious losses to the economy. Therefore, in the face of the reality that people are gradually returning to their normal life order, it has become a huge challenge to scientifically take simple and effective measures to suppress the re-emergence of the epidemic.

[0003] By building a refined infectious disease transmission model based on individuals, it is helpful to track the travel locations of individuals at different times, thereby reconstructing the dynamic contact network between individuals and accurately simulating the spatiotemporal transmission process of infectious diseases between people. The refined infectious disease transmission model based on individuals can support individual-level spatiotemporal diffusion simulation and refined prevention and control measures evaluation, but the generation of individual mobility models relies on detailed individual attribute information and individual activity trajectories to construct individual dynamic contact networks. To study the spread of infectious diseases at the city level, the individual mobility model needs to match the study area in terms of population attributes, travel characteristics, etc., and the model needs to be verified with real urban infection case data. The evaluation of the effectiveness of intervention measures can only measure the effect of intervention measures in the spread of the epidemic, and cannot provide a reasonable solution for suppressing the outbreak of the epidemic based on actual scenarios. In addition, the huge computing resources and computing time required to run the refined infectious disease transmission model based on individuals are still a major challenge.

[0004] At present, the number of infections in countries that implement high-intensity mask-wearing and close contact tracing is significantly lower than that in countries that resist mask-wearing and are unable to implement close contact tracing. Timely testing of infected people after onset directly shortens the transmission time of infected people and improves the effectiveness of close contact tracing. However, very few studies have focused on the impact of the three intervention measures of wearing masks, close contact tracing, and timely testing speed after onset on the possibility of an epidemic outbreak. Summary of the invention

[0005] The present invention provides a method, system, terminal and storage medium for simulating the spread of an infectious disease, aiming to solve at least one of the above-mentioned technical problems in the prior art to a certain extent.

[0006] In order to solve the above problems, the present invention provides the following technical solutions:

[0007] A method for simulating the spread of an infectious disease, comprising:

[0008] Constructing an individual mobility model based on multi-source data within the prevention and control area; the multi-source data includes population census data, building census data, travel survey data and mobile phone signaling data;

[0009] constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model;

[0010] Modeling of non-pharmaceutical interventions implemented in the control area;

[0011] An individual-based infectious disease transmission model is constructed based on the dynamic contact network and the intervention measures modeling, and the infectious disease transmission intensity in the prevention and control area is simulated through the infectious disease transmission model.

[0012] The technical solution adopted by the embodiment of the present invention also includes: the individual mobility model is constructed according to the multi-source data in the control area specifically:

[0013] Using the population census data to generate synthetic individuals corresponding to the total population, and reconstructing individual attribute information of the synthetic individuals according to population statistical characteristics; the individual attribute information includes age, gender, family structure, and place of residence attributes;

[0014] Taking the mobile phone signaling data as the main data and the travel survey data as the supplementary data, and taking the residence attribute of the synthetic individual as the common semantics, the travel trajectory of the synthetic individual is matched through data mining and fusion to generate the 24-hour travel activity chain of the synthetic individual and the travel locations corresponding to various travel activities; the travel activities include activities at home, work, school and other public places.

[0015] The technical solution adopted by the embodiment of the present invention also includes: the step of constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model is specifically as follows:

[0016] According to the 24-hour travel activity chain of the synthetic individual, with the time granularity of hour, the individuals who perform the same travel activity in the same building within the same hour are regarded as spatiotemporal co-occurring individuals;

[0017] Based on the time-space co-occurring individuals, fixed contact circles for a small number of acquaintances and random contact circles for strangers are constructed respectively.

[0018] The technical solution adopted by the embodiment of the present invention also includes: the non-drug intervention measures include isolation, timely testing, wearing masks, social distancing, and city closure; the modeling of the non-drug intervention measures implemented in the prevention and control area is specifically:

[0019] When the susceptible individual comes into contact with an infected individual without any intervention measures, the probability p of the susceptible individual being infected is:

[0020] p=pTrans×I c ×

[0021] When wearing a mask, the probability of the susceptible individual being infected is:

[0022] p=pTrans×I c ×r×(1-θ)

[0023] When the infected person is self-isolating at home after falling ill, the probability of contact with the infected person within the family is:

[0024] p=pTrans×I c ×r×(1-δ)

[0025] In the above formula: pTrans is the transmission probability of each contact, I c is the degree of contact, θ is the effectiveness of the mask, δ is the effectiveness of self-isolation at home, and r is the ability of the infected person to spread the virus before and after the onset of symptoms; when the susceptible individual comes into contact with the infected person, a random number x uniformly distributed in [0,1] is generated, and the random number x is compared with the infection probability p. If x≤p, it means that the susceptible individual is infected.

[0026] The technical solution adopted by the embodiment of the present invention also includes: the modeling and constructing of an individual-based infectious disease transmission model based on the dynamic contact network and the intervention measures also includes:

[0027] The infectious disease transmission model is verified using the actual incidence data in the prevention and control area. The verification process is specifically as follows: fitting the transmission probability pTrans based on the given R0 value in a scenario without any intervention measures; in a baseline epidemic scenario, fitting the two parameters of mask effectiveness θ and home isolation effectiveness δ based on the criterion of minimizing the root mean square error between the simulation results and the actual incidence data curve in the prevention and control area. The baseline epidemic scenario refers to the scenario of the first wave of infectious disease epidemics experienced in the prevention and control area.

[0028] The technical solution adopted by the embodiment of the present invention also includes: the modeling and constructing of an individual-based infectious disease transmission model based on the dynamic contact network and the intervention measures also includes:

[0029] For sporadic epidemics under the normalized epidemic prevention scenario, the outbreak probability of the infectious disease is simulated under the combination of close contact tracing, wearing masks and timely testing, and a combination of intervention measures to suppress the spread of infectious diseases is generated; the normalized epidemic prevention scenario refers to the scenario where normal life is resumed after the infectious disease is suppressed.

[0030] The technical solution adopted by the embodiment of the present invention also includes: the probability of the outbreak of the infectious disease under the combination of the three measures of close contact tracking, wearing masks and timely detection is:

[0031] The close contact tracing level is divided into five close contact tracing levels, namely, no close contact tracing, tracing close contacts of families, tracing close contacts of fixed contact circles, tracing fixed contact circles and 50% of random close contacts, and tracing all close contacts. Two scenarios of low and high mask effectiveness are set respectively to simulate the outbreak probability, and the probability of an outbreak in each epidemic prevention scenario within a set time after the first case is discovered is calculated to generate a combination of intervention measures for suppressing the spread of infectious diseases under the corresponding epidemic prevention scenarios.

[0032] Another technical solution adopted by the embodiment of the present invention is: a system for simulating the spread of infectious diseases, comprising:

[0033] Mobility model building module: used to build individual mobility models based on multi-source data in the prevention and control area; the multi-source data includes population census data, building census data, travel survey data and mobile phone signaling data;

[0034] A contact network construction module: used for constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model;

[0035] Infection probability calculation module: used to model non-drug intervention measures implemented in the prevention and control area;

[0036] Transmission model construction module: used to construct an individual-based infectious disease transmission model based on the dynamic contact network and the intervention measures modeling, and simulate the infectious disease transmission intensity in the prevention and control area through the infectious disease transmission model.

[0037] The technical solution adopted by the embodiment of the present invention also includes: a terminal, the terminal includes a processor and a memory coupled to the processor, wherein:

[0038] The memory stores program instructions for implementing the above-mentioned infectious disease transmission simulation method;

[0039] The processor is used to execute the program instructions stored in the memory to control an infectious disease transmission simulation.

[0040] The technical solution adopted by the embodiment of the present invention also includes: a storage medium storing program instructions executable by a processor, wherein the program instructions are used to execute the above-mentioned infectious disease transmission simulation method.

[0041] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows: the infectious disease transmission simulation method, system, terminal and storage medium of the embodiments of the present invention construct an individual mobility model by integrating multi-source data such as population census data, building census data, travel survey data, mobile phone signaling data, etc., so that the individual mobility model is consistent with the real environment of the study area, and a spatiotemporal dynamic contact network based on individuals is constructed based on the individual mobility model, so that the activity trajectory of individuals at the urban scale and the contact network between individuals are explicitly expressed and tracked, which is conducive to inversion and prediction of the spread of infectious diseases within the city. An individual-based infectious disease transmission model is constructed based on the dynamic contact network, and the infectious disease transmission intensity of the prevention and control area is simulated through the infectious disease transmission model. The infectious disease transmission model describes the individual's workplace and hourly travel location in detail, and tracks the contacts that the individual may encounter in various travel activities, so as to simulate the infectious disease transmission intensity in the prevention and control area. The embodiments of the present invention can support the integrated modeling of various non-drug intervention measures such as close contact tracking and isolation, isolation of outsiders, wearing masks, reducing travel, restricting inter-city transportation, suspending school, and gradually resuming work. It not only helps to evaluate the individual's compliance with various intervention measures, but also can accurately evaluate the effectiveness of various non-drug intervention measures in suppressing the spread of infectious diseases. In the face of sporadic epidemics in normalized epidemic prevention scenarios, the embodiments of the present invention further evaluate the probability of outbreaks of infectious diseases under the combination of close contact tracking, wearing masks and timely detection, and generate a combination of intervention measures to suppress sporadic epidemics of infectious diseases, thereby systematically and scientifically assisting in responding to future epidemic prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of a method for simulating the spread of infectious diseases according to an embodiment of the present invention;

[0043] Figure 2 is a diagram of the infectious disease propagation model architecture of an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the impact of different levels of close contact tracing on the re-outbreak of the epidemic in an embodiment of the present invention;

[0045] Figure 4 Schematic diagram of the reconstruction result of the individual spatiotemporal contact model of the embodiment of the present invention, wherein (a) is a comparison between the age of the synthetic individual and the census data, (b) is a comparison between the family structure of the synthetic individual and the travel survey data, (c) is a comparison between the number of contacts of the synthetic individual in each age group and the survey data of City B, and (d) is the degree distribution of the number of contacts of the synthetic individual;

[0046] Figure 5 It is a schematic diagram of the fitting results of the effectiveness of different masks and the effectiveness of home isolation on the cumulative number of cases in an embodiment of the present invention;

[0047] Figure 6 A schematic diagram of the fitting results between the actual disease data observed in the embodiment of the present invention and the simulation results;

[0048] Figure 7 A schematic diagram of spatial accuracy analysis between simulation results and observation results in an embodiment of the present invention;

[0049] Figure 8 This is a schematic diagram of the probability of an epidemic outbreak within four weeks under different intensities of intervention measures implemented in an embodiment of the present invention;

[0050] Fig. 9 This is a schematic diagram of the probability of an epidemic outbreak within eight weeks under different intensities of implementation of intervention measures in an embodiment of the present invention;

[0051] Fig.10 is a schematic diagram of the structure of an infectious disease propagation simulation system according to an embodiment of the present invention;

[0052] Fig.11 A schematic diagram of the terminal structure of an embodiment of the present invention;

[0053] Fig.12 Schematic diagram of the structure of a storage medium according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] In view of the shortcomings of the prior art, the infectious disease propagation simulation method of the embodiment of the present invention is based on urban population census data, building census data, family travel survey data, mobile phone user's mobile phone location data and other data to build an individual-based infectious disease propagation model containing individual family attributes and home addresses (refined to the building level). Based on the infectious disease propagation model, it is possible to support the implementation of isolation measures such as home isolation, building isolation, and community isolation. In addition, the infectious disease propagation model details the individual's workplace and hourly travel location, and tracks their colleagues / classmates, as well as contacts that may be encountered in other activities such as leisure and entertainment, which can support the fusion modeling of various intervention measures such as close contact tracking and isolation, isolation of outsiders, wearing masks, reducing travel, restricting intercity transportation, suspending school, and gradually resuming work, so as to not only evaluate the effects of the intervention measures that have been implemented, but also systematically and scientifically assist in responding to future epidemic prevention and control. In terms of computing resources and computing efficiency, the embodiment of the present invention transplants the infectious disease transmission model to the Sugon supercomputer computing platform. Its powerful computing power meets the needs of propagation simulation and computing of various infectious diseases such as COVID-19, influenza, norovirus, dengue fever, etc., enabling it to cope with large-scale simulations in various scenarios and analyze in detail the effectiveness of various intervention measures in epidemic prevention and control.

[0056] For details, please refer to Figure 1 , is a flow chart of a method for simulating the spread of infectious diseases according to an embodiment of the present invention. The method for simulating the spread of infectious diseases according to an embodiment of the present invention comprises the following steps:

[0057] S10: Integrate multi-source data such as population census data, building census data, travel survey data, and mobile phone signaling data to build an individual mobility model;

[0058] In this step, the individual mobility model has both individual attribute characteristics and individual mobility characteristics. The individual mobility model is constructed as follows:

[0059] S11: Generate synthetic individuals corresponding to the total population using census data, and reconstruct individual attribute information of the synthetic individuals based on population statistical characteristics;

[0060] S12: Based on mobile phone signaling data and supplemented by travel survey data, with the individual's place of residence as the common semantics, through data mining and fusion, the travel trajectory matching of synthetic individuals is achieved, and a 24-hour travel activity chain of synthetic individuals is generated; among which, the number of synthetic individuals is equivalent to the total population of the prevention and control area, with individual attribute information, 24-hour travel activity chain and travel location during various travel activities. Among them, individual attribute information includes age, gender, family structure, occupation, place of residence (building level) and other information, and activity types include activities at home, work, school and other public places.

[0061] S20: Constructing individual-based spatiotemporal dynamic contact networks based on individual mobility models;

[0062] In this step, the spatiotemporal dynamic contact network construction method specifically includes:

[0063] S21: Based on the 24-hour travel activity chain of the synthetic individual, with the hour as the time granularity, individuals who perform the same type of activities in the same building within the same hour are regarded as spatiotemporal co-occurring individuals;

[0064] S22: Based on the spatiotemporal co-occurrence individuals, fixed contact circles for a small number of acquaintances and random contact circles for strangers are constructed respectively; among them, according to the actual situation of the prevention and control area, fixed contact circles for the first four contact environments are set up for the five contact environments of family, kindergarten, primary and secondary schools, workplace and public places, including family members, people in the same class in kindergarten (up to 25 people), people in the same class in primary and secondary schools (up to 50 people), and members of the same working group in the workplace (up to 10 people). According to the existing urban research results, it is assumed that family members will have contact with each other every day; children in kindergartens have contact with 10 classmates in the fixed contact circle every day, and have random contact with 2 children in other classes in the same kindergarten; each student in primary and secondary schools has contact with 20 classmates in the fixed contact circle, and has random contact with 5 classmates in other classes; each individual in the workplace has contact with 7 colleagues in the fixed contact circle, and has random contact with 3 colleagues in other groups. The specific number of contacts that spatiotemporal co-occurrence individuals have in public places is subject to the overall constraint of the long-tail distribution of the total number of contacts that individuals have every day.

[0065] S30: Modeling of non-drug intervention measures such as isolation, timely testing, wearing masks, social distancing, and city lockdowns implemented in prevention and control areas;

[0066] In this step, non-drug intervention measures are not limited to isolation, timely testing, wearing masks, social distancing, and city lockdowns, but also include travel restrictions, environmental disinfection, and other intervention measures. The start and end time of each intervention measure can be set in the model according to the actual situation of the prevention and control area. Taking COVID-19 (new coronavirus) as an example, the specific modeling content includes:

[0067] a) Various types of isolation: simulated isolation of apparent and latently infected persons after diagnosis, centralized isolation of close contacts of confirmed patients for 14 days, home isolation of neighbors living in the building where confirmed patients live for 14 days, home isolation of migrants entering the prevention and control area for 14 days, and self-isolation of apparent infected persons after onset of illness and before seeking medical treatment;

[0068] b) Timely detection: Based on the epidemiological report, the model shortens the time from onset to diagnosis in stages;

[0069] c) Wearing masks: According to the actual situation, it is assumed that the probability of wearing masks before the outbreak of COVID-19 was 0, and since the outbreak of COVID-19, citizens have gradually started to wear masks when going out. The probability of wearing masks in the model changes over time, which is specifically described as P t =1 / (1+e-0.5*(t-t0)), where t0 is January 23rd.

[0070] d) Social distancing: The model simulates social distancing by canceling outings, closing schools, gradually resuming work in batches, and canceling extracurricular activities.

[0071] e) City lockdown: The model simulates the population migration in and out of the city based on Baidu Migration Index.

[0072] The probability p of a susceptible individual being infected when they come into contact with an infected individual without any intervention is:

[0073] p=pTrans×I c ×r (1)

[0074] When wearing a mask, the probability of a susceptible individual becoming infected is adjusted based on the effectiveness of the mask:

[0075] p=pTrans×I c ×r×(1-θ) (2)

[0076] When an infected person self-isolates at home after falling ill, the probability of contact with an infected person within the family is adjusted to:

[0077] p=pTrans×I c ×r×(1-δ) (3)

[0078] In formulas (1), (2), and (3), pTrans is the transmission probability of each contact, I c is the degree of contact, θ is the effectiveness of the mask, δ is the effectiveness of self-isolation at home, and r is the ability of the infected person to spread the virus before and after the onset of symptoms. Finally, the Bernoulli trial was used to determine whether the susceptible individual was infected after contact with the infected person. The specific description is: when the susceptible individual comes into contact with the infected person, a random number x uniformly distributed between [0,1] is generated, and the random number x is compared with the infection probability p. If x≤p, it means that the susceptible person is infected, otherwise it means that he is not infected.

[0079] S40: Construct an individual-based infectious disease transmission model based on dynamic contact networks and intervention measures modeling. After verifying the infectious disease transmission model based on real epidemic data, simulate the infectious disease transmission intensity in the prevention and control area through the infectious disease transmission model;

[0080] In this step, the infectious disease transmission model architecture is as follows Figure 2 As shown. Among them, S represents susceptible population, L represents population in latent period, I represents population with transmission ability, Q represents population in centralized isolation / diagnosis and treatment, R represents population removed after recovery or death, Inflow and Outflow represent the daily inflow and outflow of population in the prevention and control area respectively. After being infected, susceptible individual S will have a 25% chance of becoming a latent infected person, otherwise he will develop symptoms after the end of the incubation period and become an obvious infected person. The obvious infected person will experience an average of 5.2 days of incubation period and have the ability to spread the virus 2 days before the onset of the disease. The transmission stage of these patients before the onset of the disease is recorded as I pre_sym The transmission stage after the onset of disease is recorded as I onset After a latent period of 4.6 days, latently infected people begin to be able to spread the virus and enter the infectious period I asym And lasted for 9.5 days.

[0081] Furthermore, the embodiment of the present invention uses the real disease data of the prevention and control area to verify the infectious disease transmission model; the model verification process is specifically as follows: in the scenario without any intervention measures, the transmission probability pTrans is fitted based on the given R0 value; in the baseline epidemic scenario, the two unknown parameters of mask effectiveness θ and home isolation effectiveness δ are fitted based on the standard of minimizing the root mean square error (RMSE) between the simulation results and the real disease curve of the prevention and control area; according to relevant research, assuming that θ has a value range of 0.5-0.9 and δ has a value range of 0-0.9, and the simulation is performed in a grid search manner (first with a step size of 0.1, then with a step size of 0.05), and each group of parameters is simulated 1000 times to reduce the randomness of the simulation results. Based on the verified model, the secondary incidence rate within the family, the age distribution of the infected, and the spatial location accuracy of the infected are analyzed to verify the simulation ability of the model again.

[0082] In an embodiment of the present invention, constructing an individual-based infectious disease transmission model is helpful to demonstrate the heterogeneity between individuals. The individual-based infectious disease transmission model describes in detail the individual's workplace and hourly travel location, can track colleagues / classmates who come into contact with them, and contacts that may be encountered in other activities such as leisure and entertainment, and can support the integrated modeling of various intervention measures such as close contact tracking and isolation, isolation of outsiders, wearing masks, reducing travel, restricting inter-city transportation, suspending school, and gradually resuming work. At the same time, the non-drug intervention measures implemented in the real world are mapped to each individual, which not only helps to evaluate the individual's compliance with various intervention measures, but also can accurately evaluate the effectiveness of various intervention measures in inhibiting the spread of infectious diseases.

[0083] S50: For sporadic epidemics under normal epidemic prevention scenarios, simulate the outbreak probability of infectious diseases under the combination of close contact tracking, wearing masks and timely testing, and generate a combination of intervention measures to suppress the spread of infectious diseases;

[0084] In this step, due to the uncertainty of the effectiveness of masks, two scenarios of low (0.5) and high (0.7) mask effectiveness are set to simulate the probability of outbreak. The mask wearing rate and timely detection rate are set from 0 to 1 at intervals of 0.2, with 6 groups of parameters each. The close contact tracking level is divided into 5 levels: no close contact tracking (level 0), tracking close contacts of the family (level I), tracking close contacts of fixed contact circles (family / work unit / school) (level II), tracking fixed contact circles and 50% of random close contacts (level III), and tracking all close contacts (level IV). Based on the above parameters, a total of 360 (=6×5×6×2) epidemic prevention scenarios were simulated, and the probability of an outbreak within 4 weeks or 8 weeks after the first case was discovered was calculated for each epidemic prevention scenario, and then a combination of intervention measures to suppress the spread of infectious diseases under the corresponding epidemic prevention scenario was generated for reference by decision makers. Specifically, Figure 3 As shown in FIG. 1 , it is a schematic diagram of the impact of different levels of close contact tracing on the re-emergence of the epidemic. In the embodiment of the present invention, the epidemic suppression effect is defined as follows: If the effective reproduction number R t If the reading drops below 1 within 4 weeks (or 8 weeks, whichever is greater, depending on actual conditions) after the first case is discovered, it means the epidemic is under effective control.

[0085] It can be understood that the present invention can be widely applied to the transmission simulation of various infectious diseases such as COVID-19, influenza, norovirus, dengue fever, etc.

[0086] Based on the above, the infectious disease propagation simulation method of the embodiment of the present invention constructs an individual mobility model by integrating multi-source data such as population census data, building census data, travel survey data, and mobile phone signaling data, so that the individual mobility model is consistent with the real environment of the study area, and constructs an individual-based spatiotemporal dynamic contact network based on the individual mobility model, so that the activity trajectory of the individual at the urban scale and the contact network between individuals are explicitly expressed and tracked, which is conducive to inversion and prediction of the spread of infectious diseases within the city. An individual-based infectious disease propagation model is constructed based on a dynamic contact network, and the individual's workplace and hourly travel location are detailed through the infectious disease propagation model, and the contacts that the individual may encounter in various travel activities are tracked, which can support the fusion modeling of various non-drug intervention measures such as close contact tracking and isolation, isolation of outsiders, wearing masks, reducing travel, restricting inter-city transportation, suspending school, and gradually resuming work. In addition, by corresponding the non-drug intervention measures implemented in the real world to each individual, it is helpful to evaluate the individual's compliance with various intervention measures, and can accurately evaluate the effectiveness of various non-drug intervention measures in inhibiting the spread of infectious diseases. In the face of sporadic epidemics in normalized epidemic prevention scenarios, the embodiments of the present invention further evaluate the outbreak probability of infectious diseases under the combination of close contact tracking, wearing masks and timely detection, and generate a combination of intervention measures to suppress sporadic epidemics of infectious diseases, thereby systematically and scientifically assisting in responding to future epidemic prevention and control.

[0087] In order to verify the feasibility and effectiveness of the embodiments of the present invention, the following embodiments take City A as the research area and take the spread simulation of COVID-19 as an example to construct a COVID-19 spread model at the "individual-hour-building" level, and simulate the spread of the epidemic under the baseline epidemic scenario and the normalized epidemic prevention scenario respectively to test this solution. Among them, the baseline epidemic scenario is the first wave of COVID-19 epidemic experienced by City A, and the normalized epidemic prevention scenario refers to the scenario where people resume normal life after the epidemic is suppressed. Specifically:

[0088] (1) First, the population census data, Baidu migration data, desensitized signaling data of 5.8 million mobile phone users, travel survey records of 980,000 households, social contact survey data of 1,000 households, and census data of 600,000 buildings are integrated to build an individual mobility model for 11.2 million individual residents in complex urban environments and population heterogeneity. Using the population census data, synthetic individuals corresponding to the total population of City A are generated, and the age, gender, family structure, residence and other attributes of the synthetic individuals are reconstructed according to the statistical characteristics of the group. Secondly, with mobile phone signaling data as the main data and travel survey data as the auxiliary data, and the individual's residence as the common semantics, the travel trajectory of the synthetic individuals is matched through data mining and fusion, and a 24-hour travel activity chain is generated. Finally, based on the synthetic individual travel activity chain, with the hour as the time granularity, individuals who perform the same type of activities in the same building within the same hour are regarded as spatiotemporal co-occurring individuals. On the basis of the spatiotemporal co-occurring individuals, fixed contacts for a small number of acquaintances and random contacts for strangers are constructed according to the survey data of City B. Specifically, Figure 4 The figure shows the reconstruction results of the individual spatiotemporal contact model, where (a) is the comparison between the age of the synthetic individual and the census data, (b) is the comparison between the family structure of the synthetic individual and the travel survey data, (c) is the comparison between the number of contacts of the synthetic individual in each age group and the survey data of City B, and (d) is the degree distribution of the number of contacts of the synthetic individual.

[0089] (2) Combined with the disease transmission characteristics of COVID-19, an individual-level COVID-19 transmission model was constructed, and the COVID-19 transmission model was verified using the real incidence data of City A. First, pTrans was fitted based on R0=2.4. In the absence of any non-drug intervention scenario, 1,000 individuals were randomly selected as the starting cases, and pTrans=0.165 was obtained through systematic simulation. In the baseline epidemic scenario, 1,000 simulations were performed, and δ=0.7 minimized the root mean square error RMSE, and the model simulation results were insensitive to the value of θ, so it was difficult to determine its exact value; specifically Figure 5 As shown in the figure, it is a schematic diagram of the fitting results of different mask effectiveness and home isolation effectiveness on the cumulative number of cases. In the end, the model produced an average of 416 cases of apparent infection, which is highly consistent with the 418 real cases in City A. The observed number of new cases per day fits well with the average number of new cases per day in the simulation results, with a root mean square error of 1.354; Figure 6 The figure shows the fitting results between the observed real incidence data and the simulation results. The average household secondary infection rate is 11.02%, which is consistent with the epidemiological survey. In addition, the number of cases in each administrative district simulated by the model is highly consistent with the number of cases reported in each district (R 2 =0.95), with high spatial accuracy, as shown in Figure 7The figure shows a schematic diagram of the spatial accuracy analysis between the simulation results and the observation results.

[0090] In the above, in the baseline epidemic scenario, based on the actual situation of City A and the urban population flow data from Baidu Migration, the model takes into account the daily inflow and outflow of the city. In addition, due to the high vigilance of the population in the first wave of the epidemic and the high-intensity intervention measures implemented by the government, according to the epidemiological survey report, some patients will seek medical treatment immediately after the onset of symptoms, while some patients will self-isolate at home before seeking medical treatment. Therefore, it is assumed that patients in the first wave of the epidemic will seek medical treatment or home isolation measures immediately after the onset of symptoms. In the normalized epidemic prevention scenario, due to the gradual decrease in people's compliance with non-drug intervention measures, it is assumed that the delay time for patients to take measures (self-isolation at home or seeking testing / medical treatment) after the onset of symptoms is 1 to 2 days. Based on the city size and individual activity intensity of City A, different contact levels I are given to individuals who co-occur in time and space according to the average contact time of the activity type. c Specifically, the degree of contact among family members in home activities is 0.37, the degree of contact between kindergarten and primary and secondary school students is 0.25, the degree of contact between adults in work activities is 0.26, and the degree of contact in other public activities is 0.1.

[0091] (3) Based on the COVID-19 transmission model, the combined intervention effects of three types of non-drug intervention measures, namely close contact tracing, wearing masks and timely testing, as well as the risk of re-outbreak of the epidemic under different intervention intensities are simulated to provide scientific quantitative guidance for the normalized epidemic prevention in large cities. Based on the verified benchmark model, facing the sporadic epidemic in the normalized epidemic prevention scenario, the present invention further evaluates the outbreak probability of the COVID-19 epidemic under the combination of close contact tracing, wearing masks and timely testing. By setting two scenarios with low (0.5) and high (0.7) mask effectiveness, both scenarios show that close contact tracing has the most significant effect on suppressing the epidemic, followed by wearing masks and timely testing.

[0092] The large-scale simulation results based on this model show that the "close contact II + 80 + 40" strategy is used as the minimum prevention and control level for City A and other large cities, namely: tracking close contacts in a fixed contact circle + 80% of the public wearing masks + 40% of patients being tested promptly after onset. Other cities can refer to the "close contact II + 80 + 40" strategy based on their own prevention and control goals. Figure 8 and Fig. 9 Based on the risk simulation results, different levels of close contact tracing, mask wearing and timely testing ratios are implemented to adapt to local medical capabilities and national conditions. Figure 8 This is a schematic diagram of the probability of an epidemic outbreak within four weeks under different intensities of intervention measures (the black dotted line indicates the dividing line of 5% outbreak rate). Fig. 9Schematic diagram of the probability of an epidemic breaking out within eight weeks under different intensities of intervention measures (the black dotted line indicates the dividing line of 5% outbreak rate).

[0093] See also Fig.10 , is a schematic diagram of the structure of the infectious disease propagation simulation system according to an embodiment of the present invention. The infectious disease propagation simulation system 40 according to an embodiment of the present invention comprises:

[0094] Mobility model building module 41: used to integrate multi-source data such as population census data, building census data, travel survey data, mobile phone signaling data, etc. to build an individual mobility model;

[0095] A contact network construction module 42: used for constructing an individual-based spatiotemporal dynamic contact network based on an individual mobility model;

[0096] Infection probability calculation module 43: used to model non-drug intervention measures such as isolation, timely testing, wearing masks, social distancing, and city closures implemented in prevention and control areas;

[0097] The transmission model construction module 44 is used to construct an individual-based infectious disease transmission model based on a dynamic contact network and intervention measures modeling, and after verifying the infectious disease transmission model according to real epidemic data, simulate the infectious disease transmission intensity in the prevention and control area through the infectious disease transmission model;

[0098] Outbreak Probability Simulation Module 45: It is used to simulate the outbreak probability of infectious diseases under the combination of close contact tracking, wearing masks and timely testing for sporadic epidemics in normalized epidemic prevention scenarios, and generate a combination of intervention measures to inhibit the spread of infectious diseases.

[0099] See also Fig.11 , is a schematic diagram of a terminal structure of an embodiment of the present invention. The terminal 50 includes a processor 51 and a memory 52 coupled to the processor 51 .

[0100] The memory 52 stores program instructions for implementing the above-mentioned infectious disease transmission simulation method.

[0101] The processor 51 is used to execute program instructions stored in the memory 52 to control the simulation of infectious disease transmission.

[0102] The processor 51 may also be referred to as a CPU (Central Processing Unit). The processor 51 may be an integrated circuit chip having the ability to process signals. The processor 51 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0103] See also Fig.12 , which is a schematic diagram of the structure of the storage medium of the embodiment of the present invention. The storage medium of the embodiment of the present invention stores a program file 61 that can implement all the above methods, wherein the program file 61 can be stored in the above storage medium in the form of a software product, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (processor) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0104] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined in the present invention may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown in the present invention, but will conform to the widest scope consistent with the principles and novel features disclosed in the present invention.

Claims

1. A method for simulating the spread of infectious diseases, characterized in that: include: Constructing an individual mobility model based on multi-source data within the prevention and control area; the multi-source data includes population census data, building census data, travel survey data and mobile phone signaling data; constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model; Modeling of non-pharmaceutical interventions implemented in the control area; An individual-based infectious disease transmission model is constructed based on the dynamic contact network and the intervention measures modeling, and the infectious disease transmission intensity in the prevention and control area is simulated by the infectious disease transmission model; wherein: The individual mobility model is constructed based on multi-source data within the control area as follows: Using the population census data to generate synthetic individuals corresponding to the total population, and reconstructing individual attribute information of the synthetic individuals according to population statistical characteristics; the individual attribute information includes age, gender, family structure, and place of residence attributes; The mobile phone signaling data is mainly used, the travel survey data is supplemented, and the residence attribute of the synthetic individual is used as the common semantics. The travel trajectory of the synthetic individual is matched through data mining and fusion to generate a 24-hour travel activity chain of the synthetic individual and travel locations corresponding to various travel activities; the travel activities include activities at home, work, school and other public places; The step of constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model is specifically as follows: According to the 24-hour travel activity chain of the synthetic individual, with the time granularity of hour, the individuals who perform the same travel activity in the same building within the same hour are regarded as spatiotemporal co-occurring individuals; Based on the time-space co-occurring individuals, a fixed contact circle for a small number of acquaintances and a random contact circle for strangers are respectively constructed; The non-drug intervention measures include isolation, timely testing, wearing masks, social distancing, and city closures; the modeling of the non-drug intervention measures implemented in the prevention and control area is specifically as follows: When a susceptible individual comes into contact with an infected individual without any intervention measures, the probability p of the susceptible individual being infected is: p=pTrans×I c ×r When wearing a mask, the probability of the susceptible individual being infected is: p=pTrans×I c ×r×(1-θ) When the infected person is self-isolating at home after falling ill, the probability of contact with the infected person within the family is: p=pTrans×I c ×r×(1-δ) In the above formula: pTrans is the transmission probability of each contact, I c is the degree of contact, θ is the effectiveness of the mask, δ is the effectiveness of self-isolation at home, and r is the ability of the infected person to spread the virus before and after the onset of symptoms; when the susceptible individual comes into contact with the infected person, a random number x uniformly distributed between [0,1] is generated, and the random number x is compared with the infection probability p. If x≤p, it means that the susceptible individual is infected; The constructing of an individual-based infectious disease transmission model based on the dynamic contact network and the intervention measures modeling includes: The infectious disease transmission model is verified using the actual incidence data in the prevention and control area. The verification process is specifically as follows: fitting the transmission probability pTrans based on the given R0 value in a scenario without any intervention measures; in a baseline epidemic scenario, fitting the two parameters of mask effectiveness θ and home isolation effectiveness δ based on the criterion of minimizing the root mean square error between the simulation results and the actual incidence data curve in the prevention and control area. The baseline epidemic scenario refers to the scenario of the first wave of infectious disease epidemics experienced in the prevention and control area.

2. The infectious disease propagation simulation method according to claim 1, characterized in that: The constructing of an individual-based infectious disease transmission model based on the dynamic contact network and the intervention measures modeling also includes: For sporadic epidemics under the normalized epidemic prevention scenario, the outbreak probability of the infectious disease is simulated under the combination of close contact tracing, wearing masks and timely testing, and a combination of intervention measures to suppress the spread of infectious diseases is generated; the normalized epidemic prevention scenario refers to the scenario where normal life is resumed after the infectious disease is suppressed.

3. The infectious disease propagation simulation method according to claim 2, characterized in that: The probability of an outbreak of the infectious disease under the combination of close contact tracing, wearing masks and timely testing is: The close contact tracing level is divided into five close contact tracing levels, including no close contact tracing, tracing close contacts of families, tracing close contacts of fixed contact circles, tracing fixed contact circles and 50% of random close contacts, and tracing all close contacts. Two scenarios of low and high mask effectiveness are set respectively to simulate the outbreak probability, and the probability of an outbreak in each epidemic prevention scenario within a set time after the first case is discovered is calculated to generate a combination of intervention measures to suppress the spread of infectious diseases under the corresponding epidemic prevention scenarios.

4. An infectious disease transmission simulation system using the infectious disease transmission simulation method according to claim 1, characterized in that: include: Mobility model building module: used to build individual mobility models based on multi-source data in the prevention and control area; the multi-source data includes population census data, building census data, travel survey data and mobile phone signaling data; A contact network construction module: used for constructing an individual-based spatiotemporal dynamic contact network based on the individual mobility model; Infection probability calculation module: used to model non-drug intervention measures implemented in the prevention and control area; Transmission model construction module: used to construct an individual-based infectious disease transmission model based on the dynamic contact network and the intervention measures modeling, and simulate the infectious disease transmission intensity in the prevention and control area through the infectious disease transmission model.

5. A terminal, characterized in that: The terminal includes a processor and a memory coupled to the processor, wherein: The memory stores program instructions for implementing the infectious disease propagation simulation method according to any one of claims 1 to 3; The processor is used to execute the program instructions stored in the memory to control an infectious disease transmission simulation.

6. A storage medium, characterized in that: Processor-executable program instructions are stored, and the program instructions are used to execute the infectious disease transmission simulation method described in any one of claims 1 to 3.

Images