CN104615884A - Severe infectious disease and mortality risk early warning system and method based on virus detection rate - Google Patents
Severe infectious disease and mortality risk early warning system and method based on virus detection rate Download PDFInfo
- Publication number
- CN104615884A CN104615884A CN201510056900.2A CN201510056900A CN104615884A CN 104615884 A CN104615884 A CN 104615884A CN 201510056900 A CN201510056900 A CN 201510056900A CN 104615884 A CN104615884 A CN 104615884A
- Authority
- CN
- China
- Prior art keywords
- risk
- infectious disease
- severe
- node
- mortality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention relates to a severe infectious disease and mortality risk early warning system and method based on the virus detection rate. The severe infectious disease and mortality risk early warning system comprises a risk definition module, a risk element selection module, a risk network construction module, a risk evaluation prediction module and a graded early warning module. The severe infectious disease and mortality risk early warning system and method can effectively resolve the problems that an early warning mode based on infectious disease abnormal distribution judgment is inexact and not timely in judgment, prone to generating excessive early warning and the like and can be used for complementing and optimizing an infectious disease early-stage space-time early warning mode of a domestic infectious disease monitoring automatic early warning information platform.
Description
Technical field
The present invention relates to a kind of infectious disease severe based on viral recall rate, mortality risk prognoses system and method, can according to the recall rate of certain Causative virus of infectious disease, in conjunction with weather environment and risk population distribution situation, determine that this infectious disease severe, dead risk will occur at future fast, and send early warning targetedly thus, can be used for the early stage space-time modes of warning of infectious disease supplementing and optimize country of China monitoring of infectious disease automatic early-warning information platform.
Background technology
In recent years, the growth of day by day frequent and some areas ultranationalism and the religion force that exchange between excessively intensive, the countries and regions of the public health service ability of the drug-resistant variants of the environmental pollution be on the rise and scarcity of resources, original bio-pathogen, the appearance of emerging infectious disease, economically underdeveloped area and country's weakness, population of city districts, in the period occurred frequently of the whole world an is progressed into public health emergency, the life security of the mankind receives serious threat.How effectively to take precautions against the harm that accident brings, become the severe problem that national governments face.Setting up the public health emergency emergency reaction system comprising infectious disease early warning system is effectively control outbreaks of infectious diseases and popular and effectively tackle the important measures of great public health event.Only has the symptom of a trend of the health event that notes abnormalities early; science, effectively early warning is made to " symptom of a trend " public health emergency; could be implement various counter-measure to gain time; public health emergency is controlled unlikelyly in bud to cause crisis, or the extent of injury of crisis can be reduced to greatest extent.China started in 2003 " public health emergency emergency reaction Mechanism for Building project ", pass through project implementation, country establishes the Communicable disease information reporting management system (abbreviation reporting directly through network) based on case report, starts again major infectious diseases monitoring automatic early-warning Information System configuration and pilots project in 2005.Thus, through the effort of nearly 10 years, China sets up the monitoring of infectious disease automatic early-warning infosystem covering the maximum crowd in the world.The time Nuisance alarms way that this system adopts dynamic percentile control chart method and spacescan statistic to combine, detects the abnormal aggregation distribution of disease on one's own initiative.This method detects the aggregation of the case produced, not forward-looking.And only make use of total cases and incidence of disease information, comprehensively do not consider other Surveillance kind (meteorology, movement of population, article flow and purchase and sale, emphasis place and key population medical ill etc.), easily cause the imprecise, not in time of early warning.Equally, the determination of threshold value of warning is also a difficult problem.The complicacy of algorithm calculating in addition, utilizes model to carry out large-scale infectious disease time-place clustering and detects the time and the computational resource that need at substantial.
Summary of the invention
Technology of the present invention is dealt with problems and is: the present invention takes to utilize certain specific infectious disease pathology of current monitored area, environmental risk usually will predict following a period of time inner region infectious disease severe, dead occurrence risk, carry out early warning thus, the problem that the modes of warning of existing monitoring of infectious disease automatic early-warning infosystem is not forward-looking can be overcome.And the present invention is the different emergency conditioies according to following a period of time infection severe, dead occurrence risk, send early warning signal targetedly, can effectively solve the judgement of modes of warning that tradition judges based on infectious disease spatial abnormal feature imprecise, produce the problems such as excessive early warning not in time, easily; Can be used for carrying out supplementing and optimizing to the early stage space-time modes of warning of infectious disease of country of China monitoring of infectious disease automatic early-warning information platform.
Technical solution of the present invention is: a kind of infectious disease severe based on viral recall rate, mortality risk early warning system and method, it makes full use of the specific infectious disease severe in monitored area of current slot, dead relevant pathology, environmental element carry out the next time period infectious disease severe, mortality risk early warning, comprise as shown in Figure 1:
Risk definition module: the monitoring of infectious disease network data base of scanning monitored area, to choose in monitoring network the specific infectious disease average case in the past period in each month and historical time section patch and severe, death, utilize mobile percentile method to determine the historical baseline of infectious disease severe, incidence of mortality, thus define monitored area in the past each moon of following period of time this infectious disease severe, mortality risk grade.Risk class is defined result stored in risk sample database.
Risk elements chooses module: after risk definition module determines risk sample, risk elements is chosen module and is adopted the means such as expert graded or reading document, analyze in the numerous environmental element in monitored area, choose over in the time period to pathology, the environmental element of the severe of this infectious disease every month, the previous moon that mortality risk is relevant, set up risk elements system.Recycling correlation analysis is determined final to carry out infectious disease severe, pathology, environmental element needed for mortality risk prediction.These key elements are carried out sliding-model control, and result is stored in environmental element sample database.Simultaneously also same process is carried out to the related pathologies of the current moon and environmental element data, be also deposited in environmental element sample database.
Risk network struction module: in conjunction with the data in risk sample database and environmental element sample database, utilize based on information-theoretical network structure learning method, be based upon the Bayesian network that can reflect relation between pathology, environmental element and severe, dead occurrence risk of monitored area in the past period; Node in the network set up represents instruction infectious disease severe, dead occurrence risk grade variables and related pathologies, environmental element variable respectively.This module set up risk network by be pushed to carry out this infectious disease of current monitored area in risk assessment prediction module severe, mortality risk estimate.
Risk assessment prediction module: risk assessment prediction module reads related pathologies, environmental element data in environmental element sample database selected by the current moon, utilize the risk network that risk network struction module is set up, the possibility size occurred by joint probability formulae discovery next month monitored area infectious disease severe, dead various risk class.On this basis, risk assessment prediction module select the maximum risk class of possibility as next month severe, mortality risk grade, be pushed to grading forewarning system module as Rule of judgment.
Grading forewarning system module: according to risk assessment prediction module push next month monitored area this infectious disease severe, mortality risk grade, send two kinds of urgency level early warning targetedly or do not carry out early warning.Early warning signal comprises risk class, early warning relates to region and time period information.Early warning signal can with note and network page form, and the full-time worker to prevention and control of diseases department pushes.
Described risk network struction module implementation procedure is as follows:
(1) first set up and carry out with patterned way the Bayesian network initial configuration G=<V that expresses, E>, corresponding to the node V wherein in structural drawing G is instruction infectious disease severe, dead occurrence risk grade variables and the pathology selected, environmental element variable.Temporarily there is no fillet E between node V, indicate between variable temporarily without dependence;
(2) on the basis of monitored area pathology, environmental risk and infectious disease severe, dead occurrence risk historical summary, formula (1) is utilized to calculate often couple of internodal interactive information I (v
i, v
j)
.
Pr (v
i, v
j) represent that node is to v
i, v
jthe simultaneous possibility size of situation shown in two kinds of representative variablees, Pr (v
i), Pr (v
j) represent individual node v respectively
ior v
jthe simultaneous possibility size of situation shown in representative variable.All interactive information are greater than to the node pair of threshold epsilon, insert in chained list L successively from big to small according to interactive information value size, now set up first node pair that a pointer P points to chained list L.
(3) from chained list L, shift out the first two node pair, corresponding fillet is put into fillet E, and pointer P is moved on to next node pair.The node pair pointed by pointer P is taken out from chained list L, if there is no access path between these two nodes, just the limit of correspondence to be joined in fillet E and the node of correspondence is deleted from chained list L, then pointer P being pointed to the next node pair in chained list L.Repeat edged and delete step a little, until pointer P points to the tail of chained list L, or when containing h-1 bar limit in figure G till.H is the nodes in figure G.When figure G comprises h-1 bar limit, if add a limit again will form loop.
(4) pointer P is pointed to first node pair of chained list L, the node pointed by taking-up pointer P is to v again
i, v
j, find in structural drawing G and can cut off this partition node set to node contacts.Formula (2) is utilized to calculate each node centering individual node v
nwith other a certain node v in structural drawing G
sbetween conditional mutual information I (v
n, v
s| C), C is set of node, v
nfor node is to v
i, v
jin any node.
Pr (v
n, v
s| C) represent that set of node C interior joint is to v
i, v
jthe simultaneous possibility size of situation shown in two kinds of representative variablees, Pr (v
n| C), Pr (v
s| C) represent individual node v in set of node C respectively
nor v
sthe simultaneous possibility size of situation shown in representative variable.If I is (v
n, v
s| C) be less than threshold epsilon, then node is to v
n, v
scan not conditional sampling in partition set of node, so in fillet E deletion of node to v
n, v
sbetween fillet, increase node to partition internodal fillet.Repeat this step, until pointer P points to the tail of chained list L.
(5) for the limit E (v in fillet E
i, v
j), if at node v
ior v
jbetween also have other limit, so by limit E (v
i, v
j) temporarily delete from fillet E.If these two nodes can not be spaced, by limit E (v
i, v
j) rejoin fillet E; Otherwise permanent delet this edge.Finally give the limit in current figure directed.
(6) when bayesian network structure once determine, based on historical data data, in computational grid structure in the past under the pathology of the following period of time previous moon, environmental element combination condition, the joint probability distribution that the various infectious disease severes of next month, dead occurrence risk grade occur, obtains the possibility size of various infectious disease severe in the next month of monitored area, the generation of dead occurrence risk grade thus.Network structure can be optimized and revised according to historical summary.
Based on infectious disease severe, the mortality risk method for early warning of viral recall rate, it is characterized in that step is as follows:
(1) according to every month in the past period in the monitoring of infectious disease network of monitored area and earlier month historical time section patch in specific infectious disease average case and severe, death, by mobile percentile method, observation is cycle severe, the incidence of mortality level of chronomere's movement according to the moon, and synchronizing moving calculate history severe, incidence of mortality level percentile as historical baseline, define severe, the mortality risk grade of monitored area following period of time this infectious disease every month in the past thus.
(2) means such as expert graded or reading document are utilized, set up specific infectious disease severe, the pathology of dead occurrence risk and environmental risk factors system, in the previous moon that factors system relates to risk assessment, Causative virus recall rate, meteorological condition and risk population distribute multiple key element.Then determine final to carry out pathology needed for infectious disease severe, mortality risk prediction and environmental risk key element according to statistic correlation analysis; Because Bayesian Network Inference input data are type or level data, so before setting up evaluation of risk network, by method of equal intervals, equifrequency method or based on entropy method, risk elements data are carried out sliding-model control, automatically determine risk elements data from continuous type attribute to the corresponding relation of discrete type attribute.Method of equal intervals is the interval number according to specifying, and by several intervals of the codomain of numerical attribute, and makes each interval width equal.Equifrequency method is roughly the same with method of equal intervals, is carry out equidistant division to the frequency that data occur unlike method of equal intervals.And be a kind of supervision, top-down discrete method based on the method for entropy.In order to discrete values attribute, the method selects to have the value of minimum entropy as split point in numerical attribute, and recursively division result is interval, obtains layering discretize result.
(3) utilize based on information-theoretical network structure learning method, utilize the monitored area severe of following period of time this infectious disease every month, mortality risk and the pathology of the previous moon, environmental element data in the past, set up the Bayesian network that can reflect relation between pathology, environmental element and severe, dead occurrence risk of monitored area; Node in the network set up represents instruction infectious disease severe, dead occurrence risk grade variables and related pathologies, environmental element variable respectively.
(4) risk network structure is once determine, on the basis of the pathology of the monitored area current moon, environmental risk data, calculate the probability distribution of infectious disease severe, dead occurrence risk node in the Bayesian network set up, obtain the possibility size of next month this infectious disease severe of monitored area, dead various risk class generation thus, the risk class selecting probability of happening maximum is as next month the most contingent risk.
(5) according to the risk class of next month, judge that monitored area is probably in following three kinds of situations at this infectious disease severe of next month, mortality risk thus: be positioned at the high-order level of historical baseline (this infectious disease severe of next month, incidence of mortality higher than period of history infectious disease severe of the same race, incidence of mortality mxm. 80% value, namely >=P80, P represents percentage numerical digit); Be positioned at historical baseline by-level (incidence of next month is positioned at the value interval of 80% and 50% of the mxm. of period of history incidence, i.e. P80-P50); Be positioned at historical baseline reduced levels (incidence of next month lower than period of history average originating rate, namely≤P50); Then according to next month this infectious disease severe of monitored area, mortality risk different emergency conditioies, send early warning targetedly.When determining severe, mortality risk is when being positioned at the high-order level of historical baseline, and control and prevention of disease personnel should take prevention and control measure the strictest, such as isolation etc.And determine severe, mortality risk is when being positioned at historical baseline by-level, control and prevention of disease personnel will consider prevention and control measure in two kinds of situation.A kind of situation may be that epidemic situation occurs, and is in and eliminates the stage gradually, and so illustrate that early stage, prevention and control measure was had an effect, control and prevention of disease personnel should continue to use these measures; Another kind of situation may be that epidemic situation does not also occur, and is in the amount of the accumulateing stage, and so control and prevention of disease personnel should watch with the deepest concern the development of this kind of infectious disease in this area, and being necessary can targetedly to the general knowledge of public awareness campaign infectious disease prevention.
The present invention's advantage is compared with prior art: the present invention can utilize disease severe and dead relevant risk usually will predict infectious disease severe and dead occurrence risk, carries out early warning thus, forward-looking; And uncertainty analysis can be had to risk profile result, comparatively science; In addition can according to following a period of time infectious disease severe, dead occurrence risk size, carry out early warning to point rank, the aspect such as epidemic situation judgement, prevention and control measure selection is carried out to the edidemic control personnel of Disease Prevention and Control Institutions and administrative department of public health and provides decision basis.And after this invention establishes network reasoning, once can also forecasting risk when lacking some risk elements, and constantly can adjust network according to expertise, this meets the real work needs of infectious disease early warning very much.The present invention is applicable to the sick early warning of common transmittable in territory, local cell, can supplement, can improve China's existing key area infectious disease early warning pattern as of the early stage space-time modes of warning of national monitoring of infectious disease automatic early-warning information platform infectious disease.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of a kind of infectious disease severe based on viral recall rate of the present invention, mortality risk early warning system and method.
Fig. 2 is the bayesian network structure for assessment of cities and counties of Hunan Province 6 4-7 month hand-foot-and-mouth disease severe in 2013, dead occurrence risk
Embodiment
For instantiation, the concrete implementation step (as shown in Figure 1) based on the infectious disease severe of viral recall rate, mortality risk early warning system and method in the present invention is specifically introduced below with " cities and counties of Hunan Province 6 (Changde City, Xiangxi Autonomous Prefecture, Xiangtan City, Loudi City, Huaihua City and Chenzhou City) 4-7 month hand-foot-and-mouth disease severe, dead occurrence risk prediction and warning in 2013 ":
As shown in Figure 1, a kind of infectious disease severe based on viral recall rate of the present invention, mortality risk early warning system and method choose module by risk definition module, risk elements, risk network struction module, risk assessment prediction module, grading forewarning system module form.
Whole implementation procedure is as follows:
(1) the monitoring of infectious disease network data base of 6 cities and counties is scanned, to choose in monitoring network the hand-foot-and-mouth disease average case in the 4-7 month in 2012 each moon and historical time section patch (this month of 2010-2011 and front and back each two months) thereof and severe, death, utilize mobile percentile method to determine the historical baseline of the salty disease of brothers, incidence of mortality, define 6 cities and counties this infectious disease severe, mortality risk grade in the 4-7 month in 2012 thus.Risk class is defined result stored in risk sample database.
(2) after risk definition module determines risk sample, risk elements is chosen module and is adopted the means such as expert graded or reading document, analyze in the 6 numerous environmental elements of cities and counties, be chosen in the 3-6 month in 2012 to the severe of hand-foot-and-mouth disease every month, pathology, environmental element that mortality risk is relevant, set up risk elements system.Recycling correlation analysis is determined final to carry out infectious disease severe, pathology, environmental element needed for mortality risk prediction.These key elements are carried out sliding-model control, and result is stored in environmental element sample database.Simultaneously also same process is carried out to the related pathologies of every month in the 3-6 month in 2013 and environmental element data, be also deposited in environmental element sample database.
(3) in conjunction with the data in risk sample database and environmental element sample database, utilize based on information-theoretical network structure learning method, be based upon the Bayesian network that can reflect relation between pathology, environmental element and severe, dead occurrence risk of 6 cities and counties in the past period; Node in the network set up represents instruction hand-foot-and-mouth disease severe, dead occurrence risk grade variables and related pathologies, environmental element variable respectively.The risk network that this module is set up will be pushed to and carry out in the 4-7 month in 2013 cities and counties' hand-foot-and-mouth disease severe every month 6, mortality risk estimation in risk assessment prediction module.
(4) risk assessment prediction module reads related pathologies, environmental element data selected in the 3-6 month in 2013 in environmental element sample database, utilize the risk network that risk network struction module is set up, by the joint probability formulae discovery 4-7 month in 2013 every month 6 cities and counties' hand-foot-and-mouth disease severe, the possibility size that occurs of dead various risk class.On this basis, the risk class that risk assessment prediction module selects possibility maximum, as severe, the mortality risk grade of every month in the 4-7 month in 2013, is pushed to grading forewarning system module as Rule of judgment.
(5) according to risk assessment prediction module push the 4-7 months in 2013 in every month every cities and counties this infectious disease severe, mortality risk grade, send two kinds of urgency level early warning targetedly or do not carry out early warning.Early warning signal comprises risk class, early warning relates to region and time period information.Early warning signal can with note and network page form, and the full-time worker to prevention and control of diseases department pushes.
The specific implementation process of above-mentioned each module is as follows:
1. risk definition module
According to the hand-foot-and-mouth disease average case in the historical time section patch of each moon in the 4-7 month in 2012 in the monitoring of infectious disease network of 6 cities and counties and previously same period (this month of 2010-2012 and front and back each two months) thereof and severe, death, by mobile percentile method, observation is cycle severe, the incidence of mortality level of chronomere's movement according to the moon, and synchronizing moving calculates history severe, incidence of mortality horizontal base line (percentile), defines 6 cities and counties' hand-foot-and-mouth disease severes, mortality risk thus.(incidence of each month is lower than period of history average originating rate for≤P50, P represents percentage numerical digit) be wind force 0 danger, P80-P50 (value that the incidence of each month is positioned at 80% and 50% of the mxm. of period of history incidence is interval) is light breeze danger, >=P80 (this infectious disease severe of each month, incidence of mortality higher than period of history infectious disease severe of the same race, incidence of mortality mxm. 80% value) be gentle breeze danger.Risk class is defined result stored in risk sample database.
2. risk elements chooses module
Utilize the means such as expert graded or reading document, set up hand-foot-and-mouth disease severe, the pathology of dead occurrence risk and environmental risk factors system, factors system relates to the severe with hand-foot-and-mouth disease every month in the 3-6 month in 2012, the total recall rate of EV71 virus of less than the 5 years old children that mortality risk is correlated with, each age group (less than 1:6 month, 2:6 month-1 year old, : 3:1-2 year, 4:2-4 year, 5:4-5 year, more than 6:5 year) the viral recall rate of EV71, hand-foot-and-mouth disease severe, incidence of mortality, the current moon meteorological condition (minimum relative humidity, average relative humidity, sunshine duration, the lowest temperature, temperature on average, the highest temperature, maximum wind velocity, mean wind speed, average gas pressure, rainfall amount), multiple key element such as Susceptible population (less than 5 years old) density of population and the level of urbanization etc.
Determine final to carry out pathology needed for hand-foot-and-mouth disease severe, mortality risk prediction and environmental risk key element according to statistic correlation analysis.The risk elements of final input Bayesian network method comprises the severe with hand-foot-and-mouth disease every month in the 3-6 month in 2012, the total recall rate of EV71 virus of less than the 5 years old children that mortality risk is correlated with, 6 months-1 year old, 1-2 year, 4-5 year and more than 5 years old children EV71 virus recall rate, hand-foot-and-mouth disease severe, incidence of mortality, the current moon meteorological condition (minimum relative humidity, average relative humidity, sunshine duration, the lowest temperature, temperature on average, the highest temperature, maximum wind velocity, mean wind speed, average gas pressure, rainfall amount), Susceptible population's (less than 5 years old) density of population and the level of urbanization.With based on entropy method, risk elements data are carried out sliding-model control, automatically determine risk elements data from continuous type attribute to the corresponding relation of discrete type attribute.There is the value of minimum entropy as split point based in the method choice numerical attribute of entropy, and recursively division result is interval, thus obtains layering discretize result.Simultaneously also same process is carried out to the related pathologies of every month in the 3-6 month in 2013 and environmental element data.Two groups of pathology, environmental element data are all deposited in environmental element sample database.
3. risk network struction module
In conjunction with the data in risk sample database and environmental element sample database, utilize based on information-theoretical network structure learning method, be based upon the Bayesian network that can reflect relation between pathology, environmental element and severe, dead occurrence risk of 6 cities and counties in the past period; Node in the network set up represents instruction hand-foot-and-mouth disease severe, dead occurrence risk grade variables and related pathologies, environmental element variable respectively.Step is as follows:
1. first set up and carry out with patterned way the Bayesian network initial configuration G=<V that expresses, E>, corresponding to the node V wherein in structural drawing G is instruction hand-foot-and-mouth disease severe, dead occurrence risk grade variables and the pathology selected, environmental element variable.Temporarily there is no fillet E between node V, indicate between variable temporarily without dependence;
2., on the basis of 6 cities and counties' pathology, environmental risk and hand-foot-and-mouth disease severe, dead occurrence risk historical summary, formula (1) is utilized to calculate often couple of internodal interactive information I (v
i, v
j).
Pr (v
i, v
j) represent that node is to v
i, v
jthe simultaneous possibility size of situation shown in two kinds of representative variablees, Pr (v
i), Pr (v
j) represent individual node v respectively
ior v
jthe simultaneous possibility size of situation shown in representative variable.All interactive information are greater than to the node pair of 0.5, insert in chained list L successively from big to small according to interactive information value size, now set up first node pair that a pointer P points to chained list L.
3. from chained list L, shift out the first two node pair, corresponding fillet is put into fillet E, and pointer P is moved on to next node pair.The node pair pointed by pointer P is taken out from chained list L, if there is no access path between these two nodes, just the limit of correspondence to be joined in fillet E and the node of correspondence is deleted from chained list L, then pointer P being pointed to the next node pair in chained list L.Repeat edged and delete step a little, until pointer P points to the tail of chained list L, or when containing h-1 bar limit in figure G till.H is the nodes in figure G.When figure G comprises h-1 bar limit, if add a limit again will form loop.
4. pointer P is pointed to first node pair of chained list L, the node pointed by taking-up pointer P is to v again
i, v
j, find in structural drawing G and can cut off this partition node set to node contacts.Formula (2) is utilized to calculate each node centering individual node v
nwith other a certain node v in structural drawing G
sbetween conditional mutual information I (v
n, v
s| C), C is set of node, v
nfor node is to v
i, v
jin any node.
Pr (v
n, v
s| C) represent that set of node C interior joint is to v
i, v
jthe simultaneous possibility size of situation shown in two kinds of representative variablees, Pr (v
n| C), Pr (v
s| C) represent individual node v in set of node C respectively
nor v
sthe simultaneous possibility size of situation shown in representative variable.If I is (v
n, v
s| C) be less than threshold epsilon, then node is to v
n, v
scan not conditional sampling in partition set of node, so in fillet E deletion of node to v
n, v
sbetween fillet, increase node to partition internodal fillet.Repeat this step, until pointer P points to the tail of chained list L.
5. for the limit E (v in fillet E
i, v
j), if at node v
ior v
jbetween also have other limit, so by limit E (v
i, v
j) temporarily delete from fillet E.If these two nodes can not be spaced, by limit E (v
i, v
j) rejoin fillet E; Otherwise permanent delet this edge.Finally give the limit in current figure directed.
6. when bayesian network structure once determine, based on pathology, the environmental element data of each cities and counties 3-6 every month month in 2012, the joint probability distribution that in computational grid structure, 3 kinds of hand-foot-and-mouth disease severes, dead occurrence risk grades occur under various pathology, environmental element combination condition, obtains the corresponding 6 cities and counties possibility size that each cities and counties every month 3 kinds of hand-foot-and-mouth disease severes, dead occurrence risk grade occur in the 4-7 month in 2012 thus.Utilize true Monitoring Data to verify and adjust bayesian network structure.
4. risk assessment prediction module
Read related pathologies, environmental element data selected in the 3-6 month in 2013 in environmental element sample database, utilize the risk network that risk network struction module is set up, by the joint probability formulae discovery 4-7 month in 2013 every month 6 cities and counties' hand-foot-and-mouth disease severe, the possibility size that occurs of dead various risk class.On this basis, the risk class that risk assessment prediction module selects possibility maximum, as severe, the mortality risk grade of every month in the 4-7 month in 2013, is pushed to grading forewarning system module as Rule of judgment.
5. grading forewarning system module
The risk class result of coming is pushed according to risk assessment prediction module, judge the hand-foot-and-mouth disease severe of 6 cities and counties in the 4-7 month in 2013 in each cities and counties next month every month thus, mortality risk be probably in following three kinds of situations: be positioned at the high-order level of historical baseline (this infectious disease severe of next month, incidence of mortality higher than period of history infectious disease severe of the same race, incidence of mortality mxm. 80% value, namely >=P80, P represents percentage numerical digit); Be positioned at historical baseline by-level (incidence of next month is positioned at the value interval of 80% and 50% of the mxm. of period of history incidence, i.e. P80-P50); Be positioned at historical baseline reduced levels (incidence of next month lower than period of history average originating rate, namely≤P50); Then according to the different emergency conditioies of hand-foot-and-mouth disease severe, mortality risk in each cities and counties next month every month, early warning is sent targetedly.When determining severe, mortality risk is when being positioned at the high-order level of historical baseline, and control and prevention of disease personnel should take prevention and control measure the strictest, such as isolation etc.And determine severe, mortality risk is when being positioned at historical baseline by-level, control and prevention of disease personnel will consider prevention and control measure in two kinds of situation.A kind of situation may be that epidemic situation occurs, and is in and eliminates the stage gradually, and so illustrate that early stage, prevention and control measure was had an effect, control and prevention of disease personnel should continue to use these measures; Another kind of situation may be that epidemic situation does not also occur, and is in the amount of the accumulateing stage, and so control and prevention of disease personnel should watch with the deepest concern the development of this kind of infectious disease in this area, and being necessary can targetedly to the general knowledge of public awareness campaign infectious disease prevention.
The content be not described in detail in instructions of the present invention belongs to the known prior art of professional and technical personnel in the field.
Claims (3)
1. based on infectious disease severe, the mortality risk early warning system of viral recall rate, it is characterized in that comprising: risk definition module, risk elements choose module, risk network struction module, risk assessment prediction module and grading forewarning system module; Wherein:
Risk definition module: the monitoring of infectious disease network data base of scanning monitored area, to choose in monitoring network the specific infectious disease average case in the past period in each month and historical time section patch and severe, death, utilize mobile percentile method to determine the historical baseline of infectious disease severe, incidence of mortality, thus define monitored area in the past each moon of following period of time this infectious disease severe, mortality risk grade; Risk class is defined result stored in risk sample database;
Risk elements chooses module: after risk definition module determines risk sample, risk elements is chosen module and is adopted the means such as expert graded or reading document, analyze in the numerous environmental element in monitored area, choose over in the time period to pathology, the environmental element of the severe of this infectious disease every month, the previous moon that mortality risk is relevant, set up risk elements system; Recycling correlation analysis is determined final to carry out infectious disease severe, pathology, environmental element needed for mortality risk prediction; These key elements are carried out sliding-model control, and result is stored in environmental element sample database; Simultaneously also same process is carried out to the related pathologies of the current moon and environmental element data, be also deposited in environmental element sample database;
Risk network struction module: in conjunction with the data in risk sample database and environmental element sample database, utilize based on information-theoretical network structure learning method, be based upon the Bayesian network that can reflect relation between pathology, environmental element and severe, dead occurrence risk of monitored area in the past period; Node in the network set up represents instruction infectious disease severe, dead occurrence risk grade variables and related pathologies, environmental element variable respectively; This module set up risk network by be pushed to carry out this infectious disease of monitored area of the current moon in risk assessment prediction module severe, mortality risk estimate;
Risk assessment prediction module: risk assessment prediction module reads related pathologies, environmental element data in environmental element sample database selected by the current moon, utilize the risk network that risk network struction module is set up, the possibility size occurred by joint probability formulae discovery next month monitored area infectious disease severe, dead various risk class; On this basis, risk assessment prediction module select the maximum risk class of possibility as next month severe, mortality risk grade, be pushed to grading forewarning system module as Rule of judgment;
Grading forewarning system module: according to risk assessment prediction module push next month monitored area this infectious disease severe, mortality risk grade, send two kinds of urgency level early warning targetedly or do not carry out early warning; Early warning signal comprises risk class, early warning relates to region and time period information; Early warning signal can with note and network page form, and the full-time worker to prevention and control of diseases department pushes.
2. a kind of infectious disease severe based on viral recall rate according to claim 1, mortality risk early warning system, is characterized in that: described risk network struction module implementation procedure is as follows:
(1) first set up and carry out with patterned way the Bayesian network initial configuration G=<V that expresses, E>, corresponding to the node V wherein in structural drawing G is instruction infectious disease severe, dead occurrence risk grade variables and the pathology selected, environmental element variable; Temporarily there is no fillet E between node V, indicate between variable temporarily without dependence;
(2) on the basis of monitored area pathology, environmental risk and infectious disease severe, dead occurrence risk historical summary, formula (1) is utilized to calculate often couple of internodal interactive information I (v
i, v
j);
Pr (v
i, v
j) represent that node is to v
i, v
jthe simultaneous possibility size of situation shown in two kinds of representative variablees, Pr (v
i), Pr (v
j) represent individual node v respectively
ior v
jthe simultaneous possibility size of situation shown in representative variable; All interactive information are greater than to the node pair of threshold epsilon, insert in chained list L successively from big to small according to interactive information value size, now set up first node pair that a pointer P points to chained list L;
(3) from chained list L, shift out the first two node pair, corresponding fillet is put into fillet E, and pointer P is moved on to next node pair; The node pair pointed by pointer P is taken out from chained list L, if there is no access path between these two nodes, just the limit of correspondence to be joined in fillet E and the node of correspondence is deleted from chained list L, then pointer P being pointed to the next node pair in chained list L; Repeat edged and delete step a little, until pointer P points to the tail of chained list L, or when containing h-1 bar limit in figure G till; H is the nodes in figure G; When figure G comprises h-1 bar limit, if add a limit again will form loop;
(4) pointer P is pointed to first node pair of chained list L, the node pointed by taking-up pointer P is to v again
i, v
j, find in structural drawing G and can cut off this partition node set to node contacts; Formula (2) is utilized to calculate each node centering individual node v
nwith other a certain node v in structural drawing G
sbetween conditional mutual information I (v
n, v
s| C), C is set of node, v
nfor node is to v
i, v
jin any node;
Pr (v
n, v
s| C) represent that set of node C interior joint is to v
i, v
jthe simultaneous possibility size of situation shown in two kinds of representative variablees, Pr (v
n| C), Pr (v
s| C) represent individual node v in set of node C respectively
nor v
sthe simultaneous possibility size of situation shown in representative variable; If I is (v
n, v
s| C) be less than threshold epsilon, then node is to v
n, v
scan not conditional sampling in partition set of node, so in fillet E deletion of node to v
n, v
sbetween fillet, increase node to partition internodal fillet; Repeat this step, until pointer P points to the tail of chained list L;
(5) for the limit E (v in fillet E
i, v
j), if at node v
ior v
jbetween also have other limit, so by limit E (v
i, v
j) temporarily delete from fillet E, if these two nodes can not be spaced, by limit E (v
i, v
j) rejoin fillet E; Otherwise permanent delet this edge; Finally give the limit in current figure directed;
(6) when bayesian network structure once determine, based on historical data data, in computational grid structure in the past under the pathology of the following period of time previous moon, environmental element combination condition, the joint probability distribution that the various infectious disease severes of next month, dead occurrence risk grade occur, obtains the possibility size of various infectious disease severe in the next month of monitored area, the generation of dead occurrence risk grade thus; Network structure is optimized and revised according to historical summary.
3., based on infectious disease severe, the mortality risk method for early warning of viral recall rate, it is characterized in that step is as follows:
(1) according to every month in the past period in the monitoring of infectious disease network of monitored area and earlier month historical time section patch in specific infectious disease average case and severe, death, by mobile percentile method, observation is cycle severe, the incidence of mortality level of chronomere's movement according to the moon, and synchronizing moving calculate history severe, incidence of mortality level percentile as historical baseline, define severe, the mortality risk grade of monitored area following period of time this infectious disease every month in the past thus;
(2) means such as expert graded or reading document are utilized, set up specific infectious disease severe, the pathology of dead occurrence risk and environmental risk factors system, in the previous moon that factors system relates to risk assessment, Causative virus recall rate, meteorological condition and risk population distribute multiple key element; Then determine final to carry out pathology needed for infectious disease severe, mortality risk prediction and environmental risk key element according to statistic correlation analysis; Because Bayesian Network Inference input data are type or level data, so before setting up evaluation of risk network, by method of equal intervals, equifrequency method or based on entropy method, risk elements data are carried out sliding-model control, automatically determine risk elements data from continuous type attribute to the corresponding relation of discrete type attribute;
(3) utilize based on information-theoretical network structure learning method, utilize the monitored area severe of following period of time this infectious disease every month, mortality risk and the pathology of the previous moon, environmental element data in the past, set up the Bayesian network that can reflect relation between pathology, environmental element and severe, dead occurrence risk of monitored area; Node in the network set up represents instruction infectious disease severe, dead occurrence risk grade variables and related pathologies, environmental element variable respectively;
(4) risk network structure is once determine, on the basis of the pathology of the monitored area current moon, environmental risk data, calculate the probability distribution of infectious disease severe, dead occurrence risk node in the Bayesian network set up, obtain the possibility size of next month this infectious disease severe of monitored area, dead various risk class generation thus, the risk class selecting probability of happening maximum is as next month the most contingent risk;
(5) according to the risk class of next month, judge that monitored area is probably in following three kinds of situations at this infectious disease severe of next month, mortality risk thus: be positioned at the high-order level of historical baseline, the high-order level of historical baseline refer to this infectious disease severe of next month, incidence of mortality higher than period of history infectious disease severe of the same race, incidence of mortality mxm. 80% value, namely >=P80, P represents percentage numerical digit; Be positioned at historical baseline by-level, historical baseline by-level refers to that the incidence of next month is positioned at the value interval of 80% and 50% of the mxm. of period of history incidence, i.e. P80-P50; Be positioned at historical baseline reduced levels, historical baseline reduced levels refers to that the incidence of next month is lower than period of history average originating rate, namely≤P50; Then according to next month this infectious disease severe of monitored area, mortality risk different emergency conditioies, send early warning targetedly; When determining severe, mortality risk is when being positioned at the high-order level of historical baseline, and control and prevention of disease personnel should take prevention and control measure the strictest; And determine severe, mortality risk is when being positioned at historical baseline by-level, control and prevention of disease personnel will consider prevention and control measure in two kinds of situation; A kind of situation may be that epidemic situation occurs, and is in and eliminates the stage gradually, then illustrate that early stage, prevention and control measure was had an effect, control and prevention of disease personnel should continue to use these measures; Another kind of situation may be that epidemic situation does not also occur, and is in the amount of the accumulateing stage, then control and prevention of disease personnel should watch with the deepest concern the development of this kind of infectious disease in this area, and being necessary can targetedly to the general knowledge of public awareness campaign infectious disease prevention.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510056900.2A CN104615884B (en) | 2015-02-04 | 2015-02-04 | A kind of infectious disease severe based on viral recall rate, mortality risk early warning system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510056900.2A CN104615884B (en) | 2015-02-04 | 2015-02-04 | A kind of infectious disease severe based on viral recall rate, mortality risk early warning system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104615884A true CN104615884A (en) | 2015-05-13 |
| CN104615884B CN104615884B (en) | 2017-07-28 |
Family
ID=53150325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510056900.2A Active CN104615884B (en) | 2015-02-04 | 2015-02-04 | A kind of infectious disease severe based on viral recall rate, mortality risk early warning system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104615884B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107133901A (en) * | 2017-05-05 | 2017-09-05 | 中国科学院深圳先进技术研究院 | A kind of infectious disease preventing control method and system |
| CN107680685A (en) * | 2017-10-24 | 2018-02-09 | 山东浪潮云服务信息科技有限公司 | A kind of disease pre-warning method and system |
| CN109949930A (en) * | 2019-03-22 | 2019-06-28 | 武汉工程大学 | Animal epidemics diagnostic method and system based on Bayesian network |
| CN110517770A (en) * | 2019-08-28 | 2019-11-29 | 中国科学院重庆绿色智能技术研究院 | The critical adverse events Decision-Making Intervention method of peri-operation period based on Bayesian network and utility systems |
| CN112185586A (en) * | 2020-11-17 | 2021-01-05 | 北京嘉和海森健康科技有限公司 | Infectious disease monitoring and early warning method and device |
| CN112185583A (en) * | 2020-10-14 | 2021-01-05 | 天津之以科技有限公司 | Data mining quarantine method based on Bayesian network |
| CN112205965A (en) * | 2020-08-28 | 2021-01-12 | 北京大学 | Health risk key event detection method and system based on time window cutting |
| CN112365942A (en) * | 2020-10-20 | 2021-02-12 | 哈尔滨学院 | Infectious disease epidemic risk prediction analysis method |
| CN113159571A (en) * | 2021-04-20 | 2021-07-23 | 中国农业大学 | Cross-border foreign species risk level determination and intelligent identification method and system |
| CN115810425A (en) * | 2022-11-30 | 2023-03-17 | 广州中医药大学第一附属医院 | Method and device for predicting mortality risk level of septic shock patient |
| CN117133461A (en) * | 2023-10-23 | 2023-11-28 | 北京肿瘤医院(北京大学肿瘤医院) | Method and related equipment for postoperative short-term death risk assessment of aged lung cancer patient |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894309A (en) * | 2009-11-05 | 2010-11-24 | 南京医科大学 | Epidemic situation predicting and early warning method of infectious diseases |
| CN202711253U (en) * | 2012-03-09 | 2013-01-30 | 李晓松 | Detecting and early-stage pre-warning system for infectious disease aggregation |
| CN103259855A (en) * | 2013-04-27 | 2013-08-21 | 北京福乐云检测科技有限公司 | Cloud platform of infectious disease and chronic disease detecting and monitoring pre-warning system |
| WO2014066270A1 (en) * | 2012-10-23 | 2014-05-01 | Theranos, Inc. | Assisted medical and associated lifestyle decision making |
-
2015
- 2015-02-04 CN CN201510056900.2A patent/CN104615884B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894309A (en) * | 2009-11-05 | 2010-11-24 | 南京医科大学 | Epidemic situation predicting and early warning method of infectious diseases |
| CN202711253U (en) * | 2012-03-09 | 2013-01-30 | 李晓松 | Detecting and early-stage pre-warning system for infectious disease aggregation |
| WO2014066270A1 (en) * | 2012-10-23 | 2014-05-01 | Theranos, Inc. | Assisted medical and associated lifestyle decision making |
| CN103259855A (en) * | 2013-04-27 | 2013-08-21 | 北京福乐云检测科技有限公司 | Cloud platform of infectious disease and chronic disease detecting and monitoring pre-warning system |
Non-Patent Citations (3)
| Title |
|---|
| 李小洲等: "基于不同参数的空间扫描统计量方法在传染病暴发探测中的效果比较", 《中国卫生统计》 * |
| 王小莉等: "传染病疫情早期预警的主要模型", 《现代预防医学》 * |
| 陈勇: "重点传染病预防控制综合分析系统的设计与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107133901B (en) * | 2017-05-05 | 2020-11-03 | 中国科学院深圳先进技术研究院 | Infectious disease prevention and control method and system |
| CN107133901A (en) * | 2017-05-05 | 2017-09-05 | 中国科学院深圳先进技术研究院 | A kind of infectious disease preventing control method and system |
| CN107680685A (en) * | 2017-10-24 | 2018-02-09 | 山东浪潮云服务信息科技有限公司 | A kind of disease pre-warning method and system |
| CN109949930A (en) * | 2019-03-22 | 2019-06-28 | 武汉工程大学 | Animal epidemics diagnostic method and system based on Bayesian network |
| CN110517770B (en) * | 2019-08-28 | 2021-08-17 | 中国科学院重庆绿色智能技术研究院 | Perperative critical adverse event intervention decision method based on Bayesian network and utility system |
| CN110517770A (en) * | 2019-08-28 | 2019-11-29 | 中国科学院重庆绿色智能技术研究院 | The critical adverse events Decision-Making Intervention method of peri-operation period based on Bayesian network and utility systems |
| CN112205965A (en) * | 2020-08-28 | 2021-01-12 | 北京大学 | Health risk key event detection method and system based on time window cutting |
| CN112185583B (en) * | 2020-10-14 | 2022-05-31 | 天津之以科技有限公司 | Data mining quarantine method based on Bayesian network |
| CN112185583A (en) * | 2020-10-14 | 2021-01-05 | 天津之以科技有限公司 | Data mining quarantine method based on Bayesian network |
| CN112365942A (en) * | 2020-10-20 | 2021-02-12 | 哈尔滨学院 | Infectious disease epidemic risk prediction analysis method |
| CN112185586A (en) * | 2020-11-17 | 2021-01-05 | 北京嘉和海森健康科技有限公司 | Infectious disease monitoring and early warning method and device |
| CN113159571A (en) * | 2021-04-20 | 2021-07-23 | 中国农业大学 | Cross-border foreign species risk level determination and intelligent identification method and system |
| CN115810425A (en) * | 2022-11-30 | 2023-03-17 | 广州中医药大学第一附属医院 | Method and device for predicting mortality risk level of septic shock patient |
| CN115810425B (en) * | 2022-11-30 | 2023-12-08 | 广州中医药大学第一附属医院 | Method and device for predicting mortality risk level of sepsis shock patient |
| CN117133461A (en) * | 2023-10-23 | 2023-11-28 | 北京肿瘤医院(北京大学肿瘤医院) | Method and related equipment for postoperative short-term death risk assessment of aged lung cancer patient |
| CN117133461B (en) * | 2023-10-23 | 2024-01-30 | 北京肿瘤医院(北京大学肿瘤医院) | Method and device for postoperative short-term death risk assessment of aged lung cancer patient |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104615884B (en) | 2017-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104615884A (en) | Severe infectious disease and mortality risk early warning system and method based on virus detection rate | |
| CN110544402B (en) | Chemical industry park virtual simulation emergency deduction system based on big data fusion | |
| Lindell et al. | A hurricane evacuation management decision support system (EMDSS) | |
| Li et al. | A household-level approach to staging wildfire evacuation warnings using trigger modeling | |
| CN107331132B (en) | A kind of method and system of Urban Fires hidden danger dynamic prediction monitoring | |
| Boneh et al. | Fog forecasting for Melbourne Airport using a Bayesian decision network | |
| Li et al. | Application of the artificial neural network and support vector machines in forest fire prediction in the guangxi autonomous region, China | |
| CN104408900A (en) | Dynamic optimization based neural network flood warning device and method | |
| Vicari et al. | LAV@ HAZARD: a web-GIS interface for volcanic hazard assessment | |
| KR101646587B1 (en) | Hybrid seasonal prediction of the Western North Pacific Tropical Cyclone Activity | |
| CN104732762A (en) | Traffic abnormal road section probability identification method | |
| Whitehead et al. | Method selection in short-term eruption forecasting | |
| Dacre et al. | Chilean wildfires: probabilistic prediction, emergency response, and public communication | |
| Sorensen et al. | Community response to hurricane threat: Estimates of warning issuance time distributions | |
| CN102456109A (en) | Training method for Trojan horse incident prediction least squares support vector machine, and prediction method | |
| CN112233381B (en) | Debris flow early warning method and system based on mechanism and machine learning coupling | |
| Gao et al. | Using multilayer perceptron to predict forest fires in jiangxi province, southeast china | |
| CN105894706A (en) | Forest fire prediction method and system | |
| CN103886177A (en) | Lightning fire daily occurrence probability predicting method based on space grids | |
| CN110399537B (en) | Artificial intelligence technology-based warning situation space-time prediction method | |
| KR20240016233A (en) | Disaster response system that detects and responds to disaster situations in real time | |
| CN111861276A (en) | Event chain-oriented dynamic risk assessment method for green tide disasters | |
| Meng et al. | A dynamic emergency decision support model for emergencies in urban areas | |
| Chiu et al. | A watershed-based debris flow early warning system using sensor web enabling techniques in heterogeneous environments | |
| CN107705001A (en) | A kind of method that basin pop-up threat message sense is established |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |