CN110299208A - Disease surveillance data exception detection method, system, equipment and storage medium - Google Patents
Disease surveillance data exception detection method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN110299208A CN110299208A CN201910430065.2A CN201910430065A CN110299208A CN 110299208 A CN110299208 A CN 110299208A CN 201910430065 A CN201910430065 A CN 201910430065A CN 110299208 A CN110299208 A CN 110299208A
- Authority
- CN
- China
- Prior art keywords
- data
- doubtful
- abnormal
- morbidity statistics
- morbidity
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
The present invention is applicable in field of computer technology, provide a kind of disease surveillance data exception detection method, system, equipment and storage medium, this method comprises: obtaining in specified section after the present illness statistical data of discrete distribution, using doubtful anomaly data detection strategy, screening obtains doubtful abnormal data from present illness statistical data；Correction strategy is recycled, screening obtains non-abnormal data from doubtful abnormal data, with finally screening obtains abnormal data or normal data from present illness statistical data.In this way, normal data and abnormal data can be efficiently differentiated, the data property of can refer to finally taken is made to be increased dramatically, to correctly guide the related works such as subsequent prevention and control, guarantees control effect.
Description
Technical field
The invention belongs to field of computer technology more particularly to a kind of disease surveillance data exception detection method, system, set
Standby and storage medium.
Background technique
It is usually to pass through: in Sentinel point hospital or health and epidemic prevention station currently, carrying out the understanding of the popular situations of diseases such as influenza
Corresponding influenza-like case data are collected week Deng pressing, it is then for statistical analysis, obtain corresponding influenza pandemic situational data.Stream
Induced current row situational data is provided to related disease control expert or doctor etc. and refers to, to carry out the prevention and control of influenza, and guides
The data collection effort of Sentinel point hospital or health and epidemic prevention station.
But since influenza-like case data collected abnormal feelings may occur because of human error or accidentalia
Condition, so that the property of can refer to of influenza-like case data and/or influenza pandemic situational data reduces, and due in data
Abnormal conditions are usually to be easy to be concerned or need to be paid close attention to, accordingly, it is possible to can be to later period correlation disease control expert or doctor
Raw prevention and control measure or the guidance for instructing work to carry out mistake, to not have corresponding control effect or even negative effect can be brought
Fruit.
Summary of the invention
The purpose of the present invention is to provide a kind of disease surveillance data exception detection method, system, equipment and storage medium,
It aims to solve the problem that present in the prior art, because of the not high problem of the data property of can refer to caused by not excluding abnormal data.
On the one hand, the present invention provides a kind of disease surveillance data exception detection methods, which comprises
Obtain the present illness statistical data of the discrete distribution in specified section；
Using doubtful anomaly data detection strategy, screening obtains doubtful abnormal number from the present illness statistical data
According to；
Using correction strategy, screening obtains non-abnormal data from the doubtful abnormal data, finally from described current
Screening obtains abnormal data or normal data in morbidity statistics data.
Further, it using doubtful anomaly data detection strategy, screens and is doubted from the present illness statistical data
Like abnormal data, specifically include:
According to the current portions morbidity statistics data in the present illness statistical data in specified range, obtain for anti-
Reflect in the current portions morbidity statistics data, between a current point morbidity statistics data and consecutive points morbidity statistics data it is real
The current variation degree designation date of border variation degree；
When the current variation degree designation date be less than preset threshold when, using the current point morbidity statistics data as
The doubtful abnormal data.
Further, according to the current portions morbidity statistics data in the present illness statistical data in specified range,
It obtains for being reflected in the current portions morbidity statistics data, a current point morbidity statistics data and consecutive points morbidity statistics
The current variation degree designation date of actual change degree between data, specifically includes:
First variance and the first average value are acquired to the current portions morbidity statistics data；
Using the quotient of the first variance and first average value as the current variation degree designation date.
Further, it using doubtful anomaly data detection strategy, screens and is doubted from the present illness statistical data
Like abnormal data, specifically include:
According to the history morbidity statistics data for corresponding to discrete distribution in section with the specified section corresponding one, used
In the first threshold data for judging whether current point morbidity statistics data are too high or too low in the present illness statistical data；
When the current point morbidity statistics data are more than that first threshold data requires, with current point disease system
It counts as the doubtful abnormal data.
Further, according to the history morbidity statistics number for corresponding to discrete distribution in section with the specified section corresponding one
According to obtaining first for judging whether current point morbidity statistics data are too high or too low in the present illness statistical data
Data are limited, are specifically included:
To in the history morbidity statistics data, with current point morbidity statistics Data Position corresponding first history portion
Point morbidity statistics data assign relatively high weight, in the history morbidity statistics data, far from the first history portion disease
Second history portion morbidity statistics data of sick statistical data assign relatively low weight, obtain intermediate data；
Calculate the second variance and the second average value of the intermediate data；
Using the sum of second average value and the first float value as first threshold data, first float value is
The reasonable multiple of the second variance.
Further, using correction strategy, screening obtains non-abnormal data from the doubtful abnormal data, with finally from
Screening obtains abnormal data or normal data in the present illness statistical data, specifically includes:
Obtain multiple history morbidity statisticses that discrete distribution in section is corresponded to the specified section corresponding at least two
In data, the first history portion morbidity statistics corresponding with the doubtful abnormal data position of current point in the doubtful abnormal data
Data；
Calculate the third party of the first history portion morbidity statistics data in multiple history morbidity statistics data
Difference and third average value；
Using the sum of the third average value and the second float value as the second threshold data, second float value is described
The reasonable multiple of third variance；
It is doubtful with the current point when the doubtful abnormal data of the current point, which is less than second threshold data, to be required
Abnormal data is as the non-abnormal data.
Further, a present illness statistical data counts gained by a data statistics station, using correction strategy, from
In the doubtful abnormal data screening obtain non-abnormal data, with finally from the present illness statistical data screening obtain it is different
Regular data or normal data, specifically include:
It obtains at least two data statistics stations out of the same area and counts resulting, at least two parts of present illness
Statistical data；
When by part part or whole part present illness statistical data via the disease data method for detecting abnormality institute
Result it is same or like when, confirm using the doubtful abnormal data as the abnormal data.
On the other hand, the present invention also provides a kind of disease surveillance data exception detection system, the system comprises:
Acquiring unit, for obtaining the present illness statistical data of the discrete distribution in specified section；
First screening unit is sieved from the present illness statistical data for utilizing doubtful anomaly data detection strategy
Choosing obtains doubtful abnormal data；And
Second screening unit, for utilizing correction strategy, screening obtains non-abnormal data from the doubtful abnormal data,
With finally screening obtains abnormal data or normal data from the present illness statistical data.
On the other hand, the present invention also provides a kind of calculating equipment, including memory and processor, the processor is executed
It realizes when the computer program stored in the memory such as the step in the above method.
On the other hand, the present invention also provides a kind of computer readable storage medium, the computer readable storage mediums
It is stored with computer program, is realized when the computer program is executed by processor such as the step in the above method.
The present invention after the present illness statistical data of discrete distribution, utilizes doubtful abnormal data in specified section in acquisition
Inspection policies, screening obtains doubtful abnormal data from present illness statistical data；Correction strategy is recycled, from doubtful abnormal number
Non- abnormal data is obtained according to middle screening, with finally screening obtains abnormal data or normal data from present illness statistical data.
In this way, normal data and abnormal data can be efficiently differentiated, the data property of can refer to finally taken is made to be increased dramatically, thus
The related works such as subsequent prevention and control are correctly guided, guarantee control effect.
Detailed description of the invention
Fig. 1 is the implementation flow chart for the disease surveillance data exception detection method that the embodiment of the present invention one provides；
Fig. 2 is the refined flow chart of step S102 in the embodiment of the present invention two；
Fig. 3 is an acquisition modes schematic diagram of current variation degree designation date in the embodiment of the present invention two；
Fig. 4 is another acquisition modes schematic diagram of current variation degree designation date in the embodiment of the present invention two；
Fig. 5 is the refined flow chart of step S102 in the embodiment of the present invention three；
Fig. 6 is the refined flow chart of step S501 in the embodiment of the present invention three；
Fig. 7 is the refined flow chart of step S103 in the embodiment of the present invention four；
Fig. 8 is the refined flow chart of step S103 in the embodiment of the present invention five；
Fig. 9 is the structural schematic diagram for the disease surveillance data exception monitoring system that the embodiment of the present invention six provides；
Figure 10 is the structural schematic diagram for the calculating equipment that the embodiment of the present invention seven provides.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Specific implementation of the invention is described in detail below in conjunction with specific embodiment:
Disease surveillance data exception detection method involved in the application mainly can be identified first from morbidity statistics data
Doubtful abnormal data, then doubtful abnormal data is modified, the non-abnormal data confirmed in doubtful abnormal data is excluded, most
Screening obtains abnormal data or normal data from present illness statistical data eventually, to efficiently differentiate normal data and exception
Data make the data property of can refer to finally taken be increased dramatically, to correctly guide the related works such as subsequent prevention and control, guarantee
Control effect.
Embodiment one:
As shown in Figure 1, the disease surveillance data exception detection method of the present embodiment mainly can be by data processing system reality
It is existing, it specifically includes that
Step S101 obtains the present illness statistical data of the discrete distribution in specified section.
Specifically, after the data statistics station such as Sentinel point hospital or health and epidemic prevention station collects morbidity statistics data, it can be by disease
Sick statistical data, which is entered into data processing system, to be handled.
Morbidity statistics data are the data in specified Interval Discrete distribution, such as: it can be day granularity data or weekly granularity number
According to etc..Specified section can be data comprising current point in time, in some cycles, be also possible to comprising a certain historical time
Data put, in some cycles, current point in time typically refer to the time point for carrying out data processing instantly, present illness statistics
Data typically refer to the targeted morbidity statistics data of data processing instantly.Such as: it is past that specified section can be current point in time
When data or history within the first half of data in the previous year, the data before and after current point in time in 6 months, historical time point
Between point to the data etc. between current point in time.
Morbidity statistics data generally can be the disease sample case load that a certain Sentinel point hospital or health and epidemic prevention station etc. collect
Measure data, disease sample case percent data, disease virus positive detection rate data etc..This method is for the popular disease such as influenza
Disease has the effect become apparent.
Step S102, using doubtful anomaly data detection strategy, screening is obtained doubtful different from present illness statistical data
Regular data.
Specifically, step S102 is mainly according to a preset data processing policy, from present illness statistical data
In doubtful abnormal data or normal data are screened, thus from present illness statistical data distinguish normal data and doubt
Like abnormal data.Whether it is normal data or doubtful abnormal data, is that can voluntarily be distinguished by data processing system by processing
's.
In this application, normal data and doubtful abnormal data are concept that is opposite and determining, data processing system
The normal data filtered out can substantially or entirely reflect that those data human error or accidentalia etc. do not occur and made
At abnormal conditions, and the doubtful abnormal data that data processing system filters out can substantially or completely reflect those data
It is doubtful that exception is caused by human error or accidentalia etc..
Normal data usually can behave as: in a certain selected period (section), the data of continuous acquisition exist certain
It fluctuates without the steady of exception, correspondingly, doubtful abnormal data usually can behave as: for normal data, occurring
Abnormal steady situation；Alternatively, normal data usually can behave as: data collected are generally not in more than normal model
The abnormal extreme value enclosed, correspondingly, doubtful abnormal data usually can behave as: for normal data, abnormal pole occur
Value.Other normal, abnormal conditions equally can similarly understand, and the similar data processing scheme for being applicable in the embodiment of the present application.It is positive because
In this way, data processing system just can determine that corresponding data processing policy, more accurately data screening is carried out.
Step S103, using correction strategy, screening obtains non-abnormal data from doubtful abnormal data, finally from current
Screening obtains abnormal data or normal data in morbidity statistics data.
Specifically, step S103 is mainly according to another preset data processing policy, it is right from doubtful abnormal data
Non- abnormal data or real abnormal data are screened, thus further discriminated between from doubtful abnormal data non-abnormal data with
And real abnormal data.Whether be non-abnormal data or real abnormal data, passed through by data processing system
What reason can be distinguished voluntarily.
In this application, non-abnormal data and real abnormal data are concept that is opposite and determining, data processing
Screening system go out non-abnormal data can substantially or entirely reflect those data although show doubtful feature singularly but
Do not occur abnormal conditions caused by human error or accidentalia etc. really, and that data processing system filters out is real
Abnormal data can substantially or completely reflect that those data validations are to cause exception by human error or accidentalia etc..
For abnormal data real in doubtful abnormal data, non-abnormal data therein is usually also shown:
For multiple selected periods (section) (the usually history same period), there are periodic breaks situations for data, namely each selected
Period can occur similar catastrophe (such as: influenza periodically outburst or the Spring Festival during go to a doctor number it is periodically relatively low),
To occur more than the abnormal extreme value of normal range (NR), but these catastrophes really occur, the abnormal extreme value accordingly generated
It should be excluded from doubtful abnormal data and should be considered being non-abnormal data, correspondingly, other remaining, aperiodicity
Doubtful abnormal data corresponding to catastrophe then accordingly can be considered as real abnormal data；Alternatively, relative to doubtful different
In regular data for real abnormal data, non-abnormal data therein is usually also shown: for numbers multiple in the same area
For station statistics obtains the morbidity statistics data of the same period according to statistics, if there is base in each data statistics station statistics the data obtained
This simultaneous, similar catastrophe, and the abnormal extreme value of normal range (NR) is occurred more than, but since these catastrophes are
Really occur, the abnormal extreme value accordingly generated should be excluded from doubtful abnormal data and should be considered being non-abnormal number
According to correspondingly, other are remaining, cannot obtain other data statistics stations statistics gained corresponding data proves not to be abnormal data
Doubtful abnormal data then accordingly can be considered as real abnormal data.Equally, just because of this, data processing system just can determine that
Corresponding data processing policy carries out more accurately data screening.
Doubtful abnormal data can be carried out to present illness statistical data using data processing system by implementing the present embodiment
Identification, then doubtful abnormal data is modified, the non-abnormal data confirmed in doubtful abnormal data is excluded, finally from current
Screening obtains abnormal data or normal data in morbidity statistics data, to efficiently differentiate normal data and abnormal data, makes
The data property of can refer to finally taken is increased dramatically, to correctly guide the related works such as subsequent prevention and control, guarantees prevention and control effect
Fruit.
Embodiment two:
The present embodiment is on the basis of embodiment one, it is further provided following content:
As shown in Fig. 2, step S102 is specifically included:
Step S201 is obtained according to the current portions morbidity statistics data in present illness statistical data in specified range
For being reflected in current portions morbidity statistics data, between a current point morbidity statistics data and consecutive points morbidity statistics data
The current variation degree designation date of actual change degree.
Specifically, whether being doubtful exception to every bit morbidity statistics data when needing in present illness statistical data
Data are judged, then can accordingly obtain the part morbidity statistics data comprising this morbidity statistics data, part morbidity statistics
Data occupy the specified range (such as: comprising 5 weeks before including current point in time) in specified section.Part morbidity statistics data
It is middle there are current point morbidity statistics data (the point morbidity statistics data for currently needing to judge whether it is doubtful abnormal data) with
And the consecutive points morbidity statistics data adjacent with current point morbidity statistics data.Consecutive points morbidity statistics data can be with currently
Point morbidity statistics data closest some morbidity statistics data or one group of point morbidity statistics data.If morbidity statistics data
For day granularity data, then, point morbidity statistics data refer to a certain day morbidity statistics data, correspondingly, part morbidity statistics number
According to containing multiple days morbidity statistics data；If morbidity statistics data are weekly granularity data, point morbidity statistics data are
Refer to a certain all morbidity statistics data, correspondingly, part morbidity statistics data contain multiple all morbidity statistics data.
Current variation degree designation date can reflect that current point morbidity statistics data and consecutive points disease adjacent thereto are united
Actual change degree between counting.Such as: as shown in figure 3, can by calculate specified range in, every two adjacent disease
Line slope k between statistical data1,k2,...,kn, n is specified range intraconnections quantity, and obtains all companies in specified range
The average value of line slope absolute value is as current variation degree designation date, line slope average value
Current variation degree designation date can also obtain in the following way, as shown in figure 4, step S201 is specifically wrapped
It includes:
Step S401, to the current portions morbidity statistics data a in specified range1,a2,...,amAcquire first variance σ1
And first average valueWherein, m is the point morbidity statistics data bulk in specified range:
Step S402, by first variance σ1With the first average valueQuotient as current variation degree designation date A:
Step S202, when current variation degree designation date is less than preset threshold, with current point morbidity statistics data work
For doubtful abnormal data.
Specifically, preset threshold reflects under normal circumstances, in certain a part of morbidity statistics data, some morbidity statistics numbers
According to the thresholding of variation degree between consecutive points morbidity statistics data adjacent thereto, an empirical value generally can be.So, when
To after above-mentioned current variation degree designation date, if current variation degree designation date is less than preset threshold, work as then reflecting
The feature of preceding morbidity statistics data meets abnormal stable defining standard, so that the current point morbidity statistics data are counted
It is determined as doubtful abnormal data according to processing system.
Implement the present embodiment, abnormal smoothly doubtful abnormal data can be identified from present illness statistical data, thus
It can provide effective reference data.
Embodiment three:
The present embodiment is on the basis of embodiment one or two, it is further provided following content:
As shown in figure 5, step S102 can also include:
Step S501, the history morbidity statistics data for corresponding to discrete distribution in section according to corresponding with specified section one,
Obtain the first threshold data for judging whether current point morbidity statistics data are too high or too low in present illness statistical data.
Specifically, history morbidity statistics data may generally be the history contemporaneous data of present illness statistical data, then right
Answering section then is the corresponding history same period, such as: the specified section of present illness statistical data is 1 day-December of January in 2018 31
Day, then the correspondence section of history contemporaneous data can are as follows: 1 day-December 31 January in 2017, or, the correspondence of history contemporaneous data
Section can include: 1 day-December 31 January in 2017 and 31 days 1 day-December of January in 2016 etc..
First threshold data can be used to be compared with current point morbidity statistics data, to judge current point morbidity statistics
Whether data are doubtful abnormal data.Such as: peak and minimum in history morbidity statistics data can be obtained, to the highest
Value or minimum assign corresponding, corresponding with population growth rate weight, and then obtain the first threshold data.
First threshold data can also obtain in the following way, as shown in fig. 6, step S501 is specifically included:
Step S601, in history morbidity statistics data, the first history corresponding with current point morbidity statistics Data Position
Morbidity statistics data in part assign relatively high weight, in history morbidity statistics data, far from the first history portion morbidity statistics
Second history portion morbidity statistics data of data assign relatively low weight, obtain intermediate data.
Specifically, the history contemporaneous data is united comprising the first history portion disease by taking 1 year history contemporaneous data as an example
Count x1,x2,...,xpAnd the second history portion morbidity statistics data y1,y2,...,yq, wherein p is the first history portion
Divide morbidity statistics data midpoint morbidity statistics data bulk, q is the second history portion morbidity statistics data midpoint morbidity statistics number
The value of data bulk, p can be delimited rule of thumb, and remaining morbidity statistics data bulk is q in history contemporaneous data.It can be right
Each point morbidity statistics data assign relatively high weight α respectively in first history portion morbidity statistics data1,α2,...,αp, to
Each point morbidity statistics data assign relatively low weight beta in two history portion morbidity statistics data1,β2,...,βp, obtain mediant
According to b, i.e. x1α1,x2α2,...xpαp,y1β1,y2β2,...,yqβq.Wherein, the height of weight is opposite, in the present embodiment,
Weight α is high relative to weight beta, and weight α1,α2,...,αpAnd β1,β2,...,βpIn, closer to current point morbidity statistics number
The weight assigned according to corresponding position is higher, gets over further away from the weight assigned with current point morbidity statistics data corresponding position
Low, Gaussian Profile can be used in the distribution of weight α, β.In this way, may make in history contemporaneous data, with current point morbidity statistics data
Several history point morbidity statistics data corresponding to corresponding position, when calculating the first threshold data, importance is relatively higher,
More reference value, to further strengthen the technical effect of the application.
Step S602 calculates the second variance σ of intermediate data b2And second average value
Step S603, with the second average valueWith the first float value ε1Sum as the first threshold data B, the first float value
ε1For second variance σ2Reasonable multiple.
Such as: ε1=5 σ2Or ε1=-5 σ2, then, the first threshold data B can be with are as follows:Alternatively,WhenWhen, the first threshold data B corresponds to Upper threshold, whenWhen, the first threshold number
Lower Threshold is corresponded to according to B.
Certainly, in other embodiments, reasonable multiple can also use other multiple values.
Step S502, when current point morbidity statistics data are more than that the first threshold data requires, with current point morbidity statistics
Data are as doubtful abnormal data.
Specifically, the first threshold data reflects data limiting case corresponding to normal, historical data, thus available
In judging whether current point morbidity statistics data are too high or too low in present illness statistical data.So, when obtaining above-mentioned first
After threshold data, if current point morbidity statistics data are higher than Upper threshold or are lower than Lower Threshold, then reflecting current point disease
The feature of statistical data meets abnormal too high or too low defining standard, so that the current point morbidity statistics data are by data
Processing system is determined as doubtful abnormal data.
Implement the present embodiment, abnormal too high or too low doubtful abnormal number can be identified from present illness statistical data
According to so as to provide effective reference data.
Example IV:
The present embodiment is on the basis of embodiment three, it is further provided following content:
As shown in fig. 7, step S103 is specifically included:
Step S701 obtains multiple history diseases of discrete distribution in corresponding with specified section at least two corresponding sections
In statistical data, the first history portion morbidity statistics corresponding with the doubtful abnormal data position of the current point in doubtful abnormal data
Data.
Specifically, needing to take multiple history contemporaneous datas (such as: the first three years history contemporaneous data), thus from each history
In contemporaneous data interception obtain morbidity statistics data in part therein (such as: containing corresponding to the front and back 5 weeks including current point in time
Data), as the first history portion morbidity statistics data, when interception needs to meet: history portion morbidity statistics data are corresponding
Position in entire history morbidity statistics data, it is corresponding with position of the doubtful abnormal data of current point in doubtful abnormal data.
The doubtful abnormal data of current point typically refers to the targeted doubtful abnormal data of point of data processing instantly.
Step S702 calculates the first history portion morbidity statistics data z in multiple history morbidity statistics data1,
z2,...,zrThird variances sigma3And third average valueWherein, r is the point disease in the first history portion morbidity statistics data
Statistical data quantity:
Step S703, with third average valueWith the second float value ε2Sum as the second threshold data C, the second float value
ε2For third variances sigma3Reasonable multiple.
Such as: ε2=2 σ3Or ε2=-2 σ3, then, the second threshold data C can be with are as follows:Alternatively,WhenWhen, the second threshold data C corresponds to Upper threshold, whenWhen, the second threshold number
Lower Threshold is corresponded to according to C.
Certainly, in other embodiments, reasonable multiple can also use other multiple values.
It usually requires to meet corresponding relationship between second threshold data C and above-mentioned first threshold data B, such as: second
The Upper threshold of threshold data is higher than the Upper threshold of the first threshold data, and the Lower Threshold of the second threshold data is lower than the first threshold data
Lower Threshold.
Step S704, it is doubtful different with current point when the doubtful abnormal data of current point, which is less than the second threshold data, to be required
Regular data is as non-abnormal data.
Such as: the doubtful abnormal data of current point is higher than the Upper threshold of the first threshold data but not higher than the second threshold data
Upper threshold, then the doubtful abnormal data of current point is as non-abnormal data；The doubtful abnormal data of current point is lower than the first threshold data
Lower Threshold but be not below the Lower Threshold of the second threshold data, then the doubtful abnormal data of current point is also used as non-abnormal data.
Implement the present embodiment, periodic breaks and non-real abnormal non-abnormal number can be identified from doubtful abnormal data
According to so as to provide effective reference data.
Embodiment five:
The present embodiment is on the basis of embodiment three or four, it is further provided following content:
One present illness statistical data counts gained by a data statistics station.
As shown in figure 8, step S103 is specific further include:
Step S801 obtains at least two data statistics stations out of the same area and counts resulting, at least two parts of current diseases
Sick statistical data.
Step S802, when by part part or whole part present illness statistical data via disease surveillance data exception detection side
When method acquired results are same or like, confirm using doubtful abnormal data as abnormal data.
Specifically, counting to obtain the morbidity statistics of the same period for data statistics multiple in the same area station in the present embodiment
For data, if there is substantially simultaneous, similar catastrophe in each data statistics station statistics the data obtained, and
The abnormal extreme value of normal range (NR) is occurred more than, but since these catastrophes really occur, the abnormal extreme value accordingly generated
It should be excluded from doubtful abnormal data and should be considered being non-abnormal data, correspondingly, other are remaining, cannot obtain
Other data statistics stations statistics gained corresponding data proves that the doubtful abnormal data of not abnormal data then accordingly can be considered as
Real abnormal data.
Implement the present embodiment, different data statistics station in one region of reflection can be identified from doubtful abnormal data while uniting
The similar catastrophe and non-real abnormal non-abnormal data counted, so as to provide effective reference data.
Embodiment six:
Fig. 9 shows the structure of the disease surveillance data exception detection system of the offer of the embodiment of the present invention six, for the ease of
Illustrate, only parts related to embodiments of the present invention are shown.
Above system includes:
Acquiring unit 901, for obtaining the present illness statistical data of the discrete distribution in specified section.
First screening unit 902 is screened from present illness statistical data for utilizing doubtful anomaly data detection strategy
Obtain doubtful abnormal data.And
Second screening unit 903, for utilizing correction strategy, screening obtains non-abnormal data from doubtful abnormal data,
With finally screening obtains abnormal data or normal data from present illness statistical data.
In the present embodiment, the function and above-mentioned each method that each unit is realized in disease surveillance data exception detection system
Corresponding part content is consistent in embodiment, and details are not described herein again.
In the present embodiment, each unit of disease surveillance data exception detection system can be by corresponding hardware or software unit
It realizes, each unit can be independent soft and hardware unit, also can integrate as a soft and hardware unit, herein not to limit
The present invention.
Embodiment seven:
Figure 10 show the embodiment of the present invention seven offer calculatings equipment structure, for ease of description, illustrate only and
The relevant part of the embodiment of the present invention.
The calculating equipment of the embodiment of the present invention includes processor 1001 and memory 1002, and processor 1001 executes memory
When the computer program 1003 stored in 1002, the step in above-mentioned each embodiment of the method, such as step shown in FIG. 1 are realized
S101 to S103.Alternatively, realizing the function of each unit in above-mentioned each Installation practice when processor 801 executes computer program 803
Can, such as the function of unit 901 to 903 shown in Fig. 9.
The calculating equipment of the embodiment of the present invention can be the robot entity with human-computer interaction module.Human-computer interaction module
It may include touch screen, microphone, loudspeaker etc..In the device, can with corresponding configuration other function module, such as: net
Network module etc..When processor 1001 executes computer program 1003 in the equipment, the step of realizing when realizing above-mentioned each method, can
With reference to the description of preceding method embodiment, details are not described herein.
Embodiment eight:
In embodiments of the present invention, a kind of computer readable storage medium is provided, which deposits
Computer program is contained, when which is executed by processor, realizes the step in above-mentioned each method embodiment, for example,
Step S101 to S103 shown in FIG. 1.
Execution shown in Fig. 9 or 10 is realized by hardware component, herein for setting for Fig. 1 to Fig. 8 operation described
Standby, unit, module, device and other assemblies.The example of hardware component includes controller, sensor, generator, driver, deposits
Reservoir, comparator, arithmetic logic unit, adder, subtracter, multiplier, divider, integrator, processor and this field
Other any electronic building bricks known to a person of ordinary skill in the art for being configured as executing operation described in this application.In an example
In, hardware component is realized by one or more processors or computer.It (such as, is patrolled by one or more processing elements
It collects array, controller and the arithmetic logic unit of door, digital signal processor, microcomputer, programmable logic controller (PLC), show
It can be with restriction known to field programmable gate array, programmable logic array, microprocessor or those of ordinary skill in the art
Mode is responded and is executed instruction to obtain other any devices of expected result or the combination of device) Lai Shixian processor or calculating
Machine.
In one example, processor or computer include or are connected to storage by processor or the instruction of computer execution
Or one or more memories of software.Executed instruction by processor or computer implemented hardware component or software (such as,
Operating system OS and the one or more software applications run on OS), to execute the operation described herein for Fig. 1 to Fig. 8.
Hardware component is additionally in response to the execution of instruction or software and accesses, manipulates, handling, creating and storing data.For simplicity, singular
Term " processor " or " computer " can be used for exemplary description described herein, but in other examples, multiple processors or
Computer is used or processor or computer include multiple processing elements or a plurality of types of processing elements or both.?
In one example, hardware component includes multiple processors, and in another example, hardware component includes processor and controller.Firmly
Part component has any one or more different processing configurations, and example includes single processor, independent processor, parallel place
Manage device, SISD single instruction single data SISD multiprocessing, single-instruction multiple-data SIMD multiprocessing, multiple instruction single data MISD multiprocessing and more
Instruction multiple is according to MIMD multiprocessing.
By being implemented as executing instruction as described above or software described in this application is executed by method to execute
Computing hardware (for example, passing through one or more processors or computer) Lai Zhihang of operation is executed shown in Fig. 1 to Fig. 8
The method of operation described in this application.For example, single processor or two or more processors or place can be passed through
Device and controller are managed to execute single operation or two or more operations.One or more processors or place can be passed through
Reason device and controller execute one or more operations, can by other one or more processors or another processor and
Another controller operates to execute one or more other.One or more processors or processor and controller are executable
Single operation or two or more operations.Hardware component is realized for control processor or computer and is executed as above
The instruction or software of the method for description can be written as computer program, code segment, instruction or any combination of them, with individually
Or jointly instruction or configuration processor or computer be used as executes the operation executed by hardware component with as described above
The machine or special purpose computer of method are operated.In one example, instruction or software includes directly by processor or calculating
The machine code that machine executes, such as, the machine code generated by compiler.In another example, instruction or software include by
Manage the more advanced code that device or computer use interpreter to execute.The common programming personnel in this field can be based on disclosing for executing
The operation executed by hardware component and the block diagram and flow chart shown in the accompanying drawings and explanation of the algorithm of method as described above
Corresponding description in book, easily writes instruction or software.
Hardware component is realized for control processor or computer and executes the instruction or software of method as described above
And any associated data, data file and data structure can be recorded, store or be fixed on it is one or more it is non-temporarily
In when property computer readable storage medium, or it is recorded, stores or is fixed on that one or more non-transitories are computer-readable deposits
On storage media.The example of non-transitory computer-readable storage media includes: read only memory ROM, random access memory
RAM, flash memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-
It is RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, tape, floppy disk, magneto-optic data storage device, optical data storage device, hard
Any device known to disk, solid-state disk and those of ordinary skill in the art, any device can be in a manner of non-transitories
Store instruction or software and any associated data, data file and data structure, and by instruction or software and any
Associated data, data file and data structure are supplied to processor or computer, so that processor or computer capacity are held
Row instruction.In one example, instruction or software and any associated data, data file and data structure are distributed
In the computer system of networking, make call instruction and software and any associated data, data file and data structure
By processor or computer in a distributed fashion by storage, access and execution.
Although the disclosure includes particular example, those of ordinary skill in the art are being obtained disclosed in this subject application
It will be clear that after comprehensive understanding:, can in the case where not departing from the spirit and scope of claim and their equivalent
These examples are carried out various changes of form and details.Example described herein should only consider in descriptive sense, rather than
For the purpose of limitation.It will be considered as can be applied in description of the features or aspects in each of the examples similar in other examples
Features or aspect.If being executed in different order the technology of description, and/or if system, framework, device or the circuit described
In component be combined in different ways and/or substituted or supplemented by other assemblies or its equivalent, then suitable knot can be achieved
Fruit.Therefore, the scope of the present disclosure is not limited by specific implementation, but is limited by claim and their equivalent
It is fixed, and all changes in the range of equivalent of the claim with them are to be interpreted as including in the disclosure.
Claims (10)
1. a kind of disease surveillance data exception detection method, which is characterized in that the described method includes:
Obtain the present illness statistical data of the discrete distribution in specified section；
Using doubtful anomaly data detection strategy, screening obtains doubtful abnormal data from the present illness statistical data；
Using correction strategy, screening obtains non-abnormal data from the doubtful abnormal data, finally from the present illness
Screening obtains abnormal data or normal data in statistical data.
2. the method as described in claim 1, which is characterized in that doubtful anomaly data detection strategy is utilized, from the current disease
Screening obtains doubtful abnormal data in sick statistical data, specifically includes:
According to the current portions morbidity statistics data in the present illness statistical data in specified range, obtain for being reflected in
Actually become in the current portions morbidity statistics data, between a current point morbidity statistics data and consecutive points morbidity statistics data
The current variation degree designation date of change degree；
When the current variation degree designation date is less than preset threshold, using the current point morbidity statistics data described in
Doubtful abnormal data.
3. method according to claim 2, which is characterized in that according in the present illness statistical data in specified range
Current portions morbidity statistics data are obtained for being reflected in the current portions morbidity statistics data, current point disease system
The current variation degree designation date for counting the actual change degree between consecutive points morbidity statistics data, specifically includes:
First variance and the first average value are acquired to the current portions morbidity statistics data；
Using the quotient of the first variance and first average value as the current variation degree designation date.
4. method according to claim 1 or 2, which is characterized in that doubtful anomaly data detection strategy is utilized, from described current
Screening obtains doubtful abnormal data in morbidity statistics data, specifically includes:
According to the history morbidity statistics data for corresponding to discrete distribution in section with the specified section corresponding one, obtain for sentencing
The first whether too high or too low threshold data of current point morbidity statistics data in the present illness statistical data of breaking；
When the current point morbidity statistics data are more than that first threshold data requires, with the current point morbidity statistics number
According to as the doubtful abnormal data.
5. method as claimed in claim 4, which is characterized in that according in corresponding with the specified section one corresponding section from
The history morbidity statistics data for dissipating distribution, obtain for judging current point morbidity statistics data in the present illness statistical data
Whether the first too high or too low threshold data, specifically include:
To in the history morbidity statistics data, the first history portion disease corresponding with the current point morbidity statistics Data Position
Sick statistical data assigns relatively high weight, unites in the history morbidity statistics data, far from the first history portion disease
The the second history portion morbidity statistics data counted assign relatively low weight, obtain intermediate data；
Calculate the second variance and the second average value of the intermediate data；
Using the sum of second average value and the first float value as first threshold data, first float value is described
The reasonable multiple of second variance.
6. method as claimed in claim 4, which is characterized in that utilize correction strategy, screened from the doubtful abnormal data
Non- abnormal data is obtained, with finally screening obtains abnormal data or normal data from the present illness statistical data, specifically
Include:
Obtain multiple history morbidity statistics data that discrete distribution in section is corresponded to the specified section corresponding at least two
In, the first history portion morbidity statistics number corresponding with the doubtful abnormal data position of current point in the doubtful abnormal data
According to；
Calculate the first history portion morbidity statistics data in multiple history morbidity statistics data third variance and
Third average value；
Using the sum of the third average value and the second float value as the second threshold data, second float value is the third
The reasonable multiple of variance；
When the doubtful abnormal data of the current point, which is less than second threshold data, to be required, with the doubtful exception of the current point
Data are as the non-abnormal data.
7. method as claimed in claim 4, which is characterized in that a present illness statistical data is united by a data statistics station
Meter gained, using correction strategy, screening obtains non-abnormal data from the doubtful abnormal data, finally from the current disease
Screening obtains abnormal data or normal data in sick statistical data, specifically includes:
It obtains at least two data statistics stations out of the same area and counts resulting, at least two parts of present illness statistics
Data；
When by part part or whole part present illness statistical data via knot obtained by the disease data method for detecting abnormality
When fruit is same or like, confirm using the doubtful abnormal data as the abnormal data.
8. a kind of disease surveillance data exception detection system, which is characterized in that the system comprises:
Acquiring unit, for obtaining the present illness statistical data of the discrete distribution in specified section；
First screening unit is screened from the present illness statistical data for utilizing doubtful anomaly data detection strategy
To doubtful abnormal data；And
Second screening unit, for utilizing correction strategy, screening obtains non-abnormal data from the doubtful abnormal data, with most
Screening obtains abnormal data or normal data from the present illness statistical data eventually.
9. a kind of calculating equipment, including memory and processor, which is characterized in that the processor executes to be deposited in the memory
It realizes when the computer program of storage such as the step in any one of claim 1 to 7 the method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realization such as the step in any one of claim 1 to 7 the method when the computer program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910430065.2A CN110299208A (en) | 2019-05-22 | 2019-05-22 | Disease surveillance data exception detection method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910430065.2A CN110299208A (en) | 2019-05-22 | 2019-05-22 | Disease surveillance data exception detection method, system, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110299208A true CN110299208A (en) | 2019-10-01 |
Family
ID=68027082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910430065.2A Pending CN110299208A (en) | 2019-05-22 | 2019-05-22 | Disease surveillance data exception detection method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110299208A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652212A (en) * | 2020-05-21 | 2020-09-11 | 哈尔滨市科佳通用机电股份有限公司 | Method for detecting loss fault of fastening bolt at end part of cross rod based on deep learning |
| CN114304033A (en) * | 2022-01-25 | 2022-04-12 | 海南大学 | Method for screening fish function indicating species in river ecosystem |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103856366A (en) * | 2012-12-06 | 2014-06-11 | 腾讯科技（深圳）有限公司 | Method and system for monitoring platform data |
| CN107871190A (en) * | 2016-09-23 | 2018-04-03 | 阿里巴巴集团控股有限公司 | A kind of operational indicator monitoring method and device |
| CN108491875A (en) * | 2018-03-19 | 2018-09-04 | 深圳乐信软件技术有限公司 | A kind of data exception detection method, device, equipment and medium |
| CN108597616A (en) * | 2018-04-11 | 2018-09-28 | 平安科技（深圳）有限公司 | Disease abnormal deviation data examination method and device, computer installation and storage medium |
| CN108647249A (en) * | 2018-04-18 | 2018-10-12 | 平安科技（深圳）有限公司 | Public sentiment data prediction technique, device, terminal and storage medium |
-
2019
- 2019-05-22 CN CN201910430065.2A patent/CN110299208A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103856366A (en) * | 2012-12-06 | 2014-06-11 | 腾讯科技（深圳）有限公司 | Method and system for monitoring platform data |
| CN107871190A (en) * | 2016-09-23 | 2018-04-03 | 阿里巴巴集团控股有限公司 | A kind of operational indicator monitoring method and device |
| CN108491875A (en) * | 2018-03-19 | 2018-09-04 | 深圳乐信软件技术有限公司 | A kind of data exception detection method, device, equipment and medium |
| CN108597616A (en) * | 2018-04-11 | 2018-09-28 | 平安科技（深圳）有限公司 | Disease abnormal deviation data examination method and device, computer installation and storage medium |
| CN108647249A (en) * | 2018-04-18 | 2018-10-12 | 平安科技（深圳）有限公司 | Public sentiment data prediction technique, device, terminal and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652212A (en) * | 2020-05-21 | 2020-09-11 | 哈尔滨市科佳通用机电股份有限公司 | Method for detecting loss fault of fastening bolt at end part of cross rod based on deep learning |
| CN114304033A (en) * | 2022-01-25 | 2022-04-12 | 海南大学 | Method for screening fish function indicating species in river ecosystem |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jackson et al. | A simulation study comparing aberration detection algorithms for syndromic surveillance | |
| CN110753971B (en) | Systems and methods for dynamically monitoring patient condition and predicting adverse events | |
| Andersson et al. | Syndromic surveillance for local outbreak detection and awareness: evaluating outbreak signals of acute gastroenteritis in telephone triage, web-based queries and over-the-counter pharmacy sales | |
| CN107871190A (en) | A kind of operational indicator monitoring method and device | |
| WO2014078563A1 (en) | Method and apparatus for remote site monitoring | |
| CN110299208A (en) | Disease surveillance data exception detection method, system, equipment and storage medium | |
| WO2016073776A1 (en) | System for management of health resources | |
| CN112633542A (en) | System performance index prediction method, device, server and storage medium | |
| US20210383926A1 (en) | Mental health assessment system and method | |
| JP5387779B2 (en) | Operation management apparatus, operation management method, and program | |
| Aghsami et al. | A meta-heuristic optimization for a novel mathematical model for minimizing costs and maximizing donor satisfaction in blood supply chains with finite capacity queueing systems | |
| Barnum et al. | Improving the efficiency of distributive and clinical services in hospital pharmacy | |
| Mancini et al. | Marked point process models for the admissions of heart failure patients | |
| US20190051411A1 (en) | Decision making platform | |
| Patwari et al. | Understanding the spatial characteristics of DRAM errors in HPC clusters | |
| CN117827593A (en) | Method and system for determining abnormality cause based on abnormality recognition result | |
| Fu et al. | Utilizing timestamps of longitudinal electronic health record data to classify clinical deterioration events | |
| CN111260142A (en) | Commodity index data prediction method and device, storage medium and electronic equipment | |
| Cheng et al. | Structural models used in real-time biosurveillance outbreak detection and outbreak curve isolation from noisy background morbidity levels | |
| CN114663232A (en) | Abnormal transaction behavior monitoring method and device and computer program product | |
| JP2015011391A (en) | Change detector, change detection system, change detection method, and program | |
| Braeye et al. | Incidence estimation from sentinel surveillance data; a simulation study and application to data from the Belgian laboratory sentinel surveillance | |
| US20210019676A1 (en) | Identifying and managing enterprise product availability | |
| EP3929932A1 (en) | Method and system for frailty detection | |
| CN111949503B (en) | Database management method, device, computing equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191001 |
|
| RJ01 | Rejection of invention patent application after publication |