CN110415832A - Infection control management system and method based on artificial intelligence - Google Patents
Infection control management system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN110415832A CN110415832A CN201910699679.0A CN201910699679A CN110415832A CN 110415832 A CN110415832 A CN 110415832A CN 201910699679 A CN201910699679 A CN 201910699679A CN 110415832 A CN110415832 A CN 110415832A
- Authority
- CN
- China
- Prior art keywords
- infection
- module
- antibiotic
- prediction
- bacterium
- 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
- 208000015181 infectious disease Diseases 0.000 title claims abstract description 113
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 title claims abstract description 13
- 230000003115 biocidal effect Effects 0.000 claims abstract description 48
- 241000894006 Bacteria Species 0.000 claims abstract description 46
- 230000001580 bacterial effect Effects 0.000 claims abstract description 32
- 238000010801 machine learning Methods 0.000 claims abstract description 23
- 238000013528 artificial neural network Methods 0.000 claims abstract description 18
- 238000013507 mapping Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000007726 management method Methods 0.000 claims description 31
- 230000004044 response Effects 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 12
- 238000013523 data management Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 5
- 241000894007 species Species 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 230000036541 health Effects 0.000 claims description 4
- 238000011461 current therapy Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000004069 differentiation Effects 0.000 claims description 3
- 230000002458 infectious effect Effects 0.000 claims description 3
- 238000013439 planning Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 206010011409 Cross infection Diseases 0.000 description 12
- 206010059866 Drug resistance Diseases 0.000 description 10
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 201000010099 disease Diseases 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 206010029803 Nosocomial infection Diseases 0.000 description 4
- 241000191967 Staphylococcus aureus Species 0.000 description 4
- 238000007418 data mining Methods 0.000 description 4
- 229930182555 Penicillin Natural products 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000002906 microbiologic effect Effects 0.000 description 3
- UWYHMGVUTGAWSP-JKIFEVAISA-N oxacillin Chemical compound N([C@@H]1C(N2[C@H](C(C)(C)S[C@@H]21)C(O)=O)=O)C(=O)C1=C(C)ON=C1C1=CC=CC=C1 UWYHMGVUTGAWSP-JKIFEVAISA-N 0.000 description 3
- 229960001019 oxacillin Drugs 0.000 description 3
- 229940049954 penicillin Drugs 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 102000004190 Enzymes Human genes 0.000 description 2
- 108090000790 Enzymes Proteins 0.000 description 2
- JGSARLDLIJGVTE-MBNYWOFBSA-N Penicillin G Chemical compound N([C@H]1[C@H]2SC([C@@H](N2C1=O)C(O)=O)(C)C)C(=O)CC1=CC=CC=C1 JGSARLDLIJGVTE-MBNYWOFBSA-N 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 239000012620 biological material Substances 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- FFRBMBIXVSCUFS-UHFFFAOYSA-N 2,4-dinitro-1-naphthol Chemical compound C1=CC=C2C(O)=C([N+]([O-])=O)C=C([N+]([O-])=O)C2=C1 FFRBMBIXVSCUFS-UHFFFAOYSA-N 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 208000032163 Emerging Communicable disease Diseases 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 239000003242 anti bacterial agent Substances 0.000 description 1
- 229940088710 antibiotic agent Drugs 0.000 description 1
- 230000003796 beauty Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012252 genetic analysis Methods 0.000 description 1
- 230000003455 independent Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000000926 neurological effect Effects 0.000 description 1
- -1 penicillin Amide Chemical class 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000005180 public health Effects 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 150000003952 β-lactams Chemical class 0.000 description 1
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
-
- 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
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present invention discloses a kind of infection control management system and method based on artificial intelligence, and wherein system includes middleware module, clinical data library module, bacterial strain identification module, expert system module, machine learning neural network and prediction and warning module；Wherein expert system module is for the bacterium newly separated and relevant antibiotic knowledge mapping, for verifying the cultivation results of antibiotic, selecting suitable antibiotic inventory and sound an alarm to the new isolated bacterium；Machine learning neural network is used to learn and establish the correlation rule of bacterium, antibiotic and infection site, and the relationship of three is established specific knowledge mapping and is integrated into expert system module；Prediction and warning module, establishes prediction model, and for predicting the Infection trend in infection data and particular range, when the predicted value of the event of infection is higher than the threshold value of setting, prediction and warning module issues early warning to medical worker and/or community.The present invention can be realized automatic monitoring, analysis and the real-time early warning of infection conditions in cases of infection and community.
Description
Technical field
The present invention relates to field of artificial intelligence more particularly to a kind of infection control management systems based on artificial intelligence
And method.
Background technique
Currently, the control and management for infecting disease have received significant attention, and hospital information and community information words construction are existing
The big element in current development is had become, each hospital and each community have increasingly paid attention to information-based construction, hospital
It integrated, optimized, analyzed, counted by the information system of hospital or community with every resource of community, be hospital or society
The decision in area provides strong argument, while can reduce cost, improves efficiency, hospital in line with patient-centered health care original
Then, community increasingly payes attention to public health, by Hospital Informatization and community see informatization improve hospital and
The quality of community's respective services pushes ahead the reform of hospital and community, and the paces of the development of hospital and community are getting faster.
Traditional Hospital Infection and Nosocomial Infections monitoring broad covered area, process complexity, inefficiency, need to collect,
It arranges, analysis mass data, it is difficult to ensure that the accuracy and timeliness of monitoring.With in hospital and community informatization it is fast
Speed development, prevention and control nosocomial infection face more and more challenges, and emerging infectious diseases continuously emerge, multi-drug resistant bacteria
Infection is increasing etc. to infecting professional in the manager of vast hospital and community, hospital and community and medical worker proposes
New, higher requirement.In order to meet the needs of hospital, community and patient, hospital needs continuous innovation and perfect, therefore,
The use of carry out system has become a kind of inevitable development trend, and hospital and Nosocomial Infections monitor in real time, nosocomial infection is real-time
Prediction and early warning by information at the first time be transmitted in doctor and related management personnel's hand, make related personnel adopt an effective measure into
Row is intervened, and is following inevitable development trend to improve hospital and Nosocomial Infections control quality.
Although currently having information management and communication system to have been supplied in hospital, it cannot achieve cases of infection and community
Automatic monitoring, analysis and the real-time early warning of interior infection conditions；Infection can not efficiently be assessed and to future of infected zone
Prediction；The real-time update of database knowledge can not be carried out, the efficiency and accuracy rate of infection management control are low.
Summary of the invention
The main purpose of the present invention is to provide a kind of infection control management system and method based on artificial intelligence, in conjunction with
Artificial intelligence and machine learning means realize the real time monitoring and Accurate Prediction of infection disease.
To achieve the goals above, according to an aspect of the invention, there is provided a kind of infection control based on artificial intelligence
Management system processed, including a kind of infection control management system based on artificial intelligence, including middleware module, clinical database mould
Block, bacterial strain identification module, expert system module, machine learning neural network and prediction and warning module；The middleware module packet
Include new infections planning system and other hospital data management systems, for complete institute sense clinical data conversion, calculate and
Match, and is transmitted to clinical data library module；The clinical data library module is used for the expert system module and the bacterial strain
Identification module provides newest clinical data；The bacterial strain identification module, for classifying to the infectious bacteria in bacterial strain, and point
Not from different infection event correlations；The expert system module, for the bacterium newly separated and relevant antibiotic knowledge graph
Spectrum, for verifying the cultivation results of antibiotic, selecting suitable antibiotic inventory and sounded an alarm to the new isolated bacterium；
The machine learning neural network, for learning and establishing bacterium, antibiotic and the correlation rule of infection site, by the pass of three
System establishes specific knowledge mapping and is integrated into the expert system module；The prediction and warning module, establishes prediction model,
For predicting the Infection trend in infection data and particular range, when the predicted value of the event of infection is higher than the threshold value of setting, institute
It states prediction and warning module and issues early warning to medical worker and/or community.
Further, other hospital data management systems that the middleware module includes further include hospital information system, reality
Test room information management system, Picture Archive and communication system and electronic health record.
Further, the process for establishing prediction model is as follows: obtaining variable from the expert system module, and is divided into
Response variable and independent variable then screen the variable, obtain the pass between the independent variable and the response variable
System, identifies the candidate independent variable that can preferably predict the response variable, using the candidate independent variable as the prediction mould
The input of type.
Further, the response variable is divided into two kinds, and a kind of expression infection event generation number, another kind indicates that case is total
Number.
Further, which further includes statistical report module, is connect with the machine learning neural network, to each infection
Event is analyzed, and generates statistical report, and infection details are presented to ward, hospital and/or community.
According to another aspect of the present invention, a kind of infection control management method based on artificial intelligence is provided, comprising: pass through
Clinical database obtains the data containing infection event；Bacterial strain identification module analyzes the infection event, from bacterium kind
Class and antibioticogram start, and are corresponding bacterial strain by division bacteria；Machine learning neural network learning simultaneously establishes bacterium, antibiotic
With the correlation rule of infection site, the relationship of three is established into specific knowledge mapping and is integrated into the expert system module
In；Whether the bacterium and relevant antibiotic knowledge mapping that expert system module analysis newly separates, check relevant antibiotic
By test, whether test result meets international standard；The state of an illness of patient and the differentiation of infection are detected, whether judges current therapy
Properly；Whether there is the infection of the state of an illness in detection hospital, if it is sounds an alarm；By the machine learning nerve net
Network repetition training and verifying obtain optimum prediction model, and prediction result is output in prediction and warning module；Update the expert
The correlation rule and knowledge mapping are stored in the expert system module by system module.
Further, the data in the clinical database are from new infections system and other hospital data management systems
System.
Further, the machine learning neural network learning and the correlation rule of bacterium, antibiotic and infection site is established
After further include that statistical report module analyzes each infection event, and generates statistical report, to ward, hospital and/or society
Infection details are presented in area.
The invention has the following advantages:
(1) middleware module and clinical data library module real-time loading institute can feel clinical data automatically, improve sensing control
Real-time.
(2) relevance between the relevance of machine learning analysis high risk factor and various infection pathogenies is introduced, it is last automatic
New and useful Knowledge Aggregation into intelligent expert system knowledge base, is improved the accuracy and accuracy of judgement by ground.
(3) can high quality complete touching prediction, verified, obtained by the repetition training of artificial neural network and machine learning
To optimum prediction model, finally by prediction result and predictive information with visual interface display in prediction and warning module.When
When predicted value is higher than the early warning value of setting, prediction and warning module issues alarm, reminds medical worker to specified region or specifying part
Position carries out science, accurately control.
Detailed description of the invention
Fig. 1 is infection control schematic diagram of management system structure of the one embodiment of the invention based on artificial intelligence；
Fig. 2 is the flow diagram that prediction model is established using Logistic regression technique.
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.
As shown in Figure 1, a kind of infection control management system based on artificial intelligence, the knowledge data which considers has micro-
Biological data, bacterial strain data, antibiotic data, admission discharge table, community personnel registration form and international Experiment on Microbiology
Room guide etc., the structure is by middleware module, clinical data library module, bacterial strain identification module, expert system module, statistical report
Module and prediction and warning module composition, wherein be introduced into machine learning data mining technology to knowledge in intelligent expert system module into
Row real-time update simultaneously makes corresponding intellectual analysis modeling to statistical report and prediction and warning.
Centre modeling block: middleware module mainly includes HIS (Hospital Information System, information for hospital
System), LIS (Laboratory Information Management System, Laboratory Information Management System), PACS
(Picture archiving and communication systems, Picture Archiving and Communication System), EMR
Hospitals' existing systems such as (Electronic Medical Record, electronic health records) and EIP new infections planning system, are based on
The infection control management system of artificial intelligence can be realized docks in real time with the data of existing system, is automatically performed and nosocomial infection
The load of relevant infection element information, and it is automatically performed the conversion, calculating and matching work of every institute's sense clinical data, finally
Constitute clinical data library module.
Bacterial strain identification module: realizing the classification of infectious bacteria in bacterial strain, so as to by visibly different infection event correlation
Get up, one plant of bacterial strain is a kind of combination of the particular result of bacterium and one group of associated antibiotic map.It is more effective in order to calculate
Microbe quantity it was reported that, each infection event only needs to calculate once.Therefore, to the same cause of disease of same patient acquisition
Learning the separating again for same bacterium that material is carried out with identical or closely similar drug resistance in the short time is considered as Simple infection
Event will not repeat in statistical report.Under default situations, the configuration of similarity function and filtration parameter meets NCCLS (beauty
The Clinical Laboratory Standard committee, state) guide that provides, but it can also be matched according to specific hospital epidemiology guide
It sets.The intelligence bacterial strain identification module can easily modify, to adapt to through genetic analysis come the case where identifying bacterial strain.
The quality of this alarm for generating the system of raising and statistical data.
Expert system module: a given bacterium newly separated and relevant antibiotic knowledge mapping, expert system solve
Three short term problems:
1) verify cultivation results: system discovery after tested but required antibiotic, does not determine the impossible of particular species
Antibiotic is as a result, and test the common relation between antibiotic result；
2) intelligence reports most suitable antibiotic inventory: in view of the cost of antibiotic, infection place, bacterial species and doctor
The features such as institute ward, obtains the intelligence report of most suitable antibiotic inventory；
3) sound an alarm to the bacterium newly separated: system provides the information in relation to the bacterium, and such as dangerous drug resistance is (to spy
The drug resistance of fixed previous generation antibiotic) and multi-drug resistant (to the drug resistance of more than one antibiotic).
Prediction and warning module: the infections relating data that prediction is obtained by system obtain the infection at the desired moment
Prediction data, prediction and warning module cooperate down some reports of generation of intelligence jointly with expert system module, and to ward and whole
The Infection trend of user and entire community in a hospital or community, pass through the repetition training of artificial neural network and machine learning
Verifying, obtain optimum prediction model, finally by prediction result and predictive information with visual interface display in prediction and warning mould
In block.When predicted value is higher than the early warning value of setting, prediction and warning module issues alarm, reminds medical worker and administrative staff couple
Specified region or appointed part carry out science, accurately control.
Expert system can provide a kind of novel separation of bacterial and antibiotic for prediction and warning module, solve two mid-terms
Problem:
1) issue the alarm of single patient: system is sounded an alarm according to the infection history of patient, comprising:
A) on multiple populations: if two of same sample material continuous different time point discoveries two or more
Bacterium；
B) drug resistance obtain: if newly identified bacterium than same species previous bacterium to more antibiotics resistances.
In the latter case, clinician can determine modification therapeutic scheme.
2) alarm: whether systems inspection finds in same ward or different wards (the same area or different zones in community)
Whether identical bacterial strain has occurred bacterium infection, is accordingly alarmed.
Whether the systems inspection has found identical bacterium bacterial strain in same ward or same community unit, or not
Same bacterial strain；Whether have discovered that new bacterium bacterial strain.In view of the bacterium collection currently isolated, expert system module is solved
The long-standing problem of identification infection outburst.But the bacterium initially occurred causes many new infection events to be significantly greater than positive reason
Condition (the use statistical technique of normal value prediction).If it find that infection quantity be higher than prediction infection quantity, then can issue quick-fried
Activating alarm simultaneously notifies epidemiologist.
Prediction policy: user customized can add for certain institute sense disease in special time period or certain time section
It is interior, report warning information；
Prediction policy association doctor and related management personnel: user can define after warning information generates and send warning information
To doctor and related management personnel；
Warning information details: doctor or related management personnel can check details, including infection event by warning information
Information, infection reason infect variation tendency, the information such as infection time.
Statistical report module: first, it can be used to monitor the diffusion of hospital or Nosocomial Infections at any time, and by diffusion data
It generates tendency chart and is convenient for medical worker and related management personnel's observation analysis；Second, it is effective that clinician is helped to carry out first
Secondary diagnosis provides reference frame for clinician；Clinical statistics and reporting modules are very rich and general: user can pass through choosing
It selects target data and freely writes explanation, and determine how to polymerize these information, and how to indicate on text and/or figure
These information.
1) illness rate: what it was measured is the number influenced in a certain crowd by a certain specified disease in a certain specific period.
Illness rate for damage caused by a certain disease in estimation crowd, realize the difficulty for the treatment of plan or before coming into effect the plan
Cost/benefit ratio be useful.
2) biomaterial or carry out source distribution/bacterium: each biomaterial or source represent point of the bacterium separated thereon
Cloth.
3) bacterium/antibiotic: for the bacterium of each separation, it all represents antibiotic test result obtained
The distribution situation of (sensitive, medium, drug resistance).
Machine learning neural network: NCCLS outline is on the basis of in view of the data in many laboratories all over the world
It establishes, therefore it contains some general guilding principles, these guilding principles completely and may not be able to be explained correctly
The infection occurred in particular hospital environment.In order to solve these problems, data mining technology is applied to infection in the present embodiment
In data.In particular, being extracted correlation rule from data.Correlation rule is mutual by the neurological susceptibility of different antibiotic or drug resistance
Connect, verify certain NCCLS rule and by " it was found that " the new rule that not yet considers is come in terms of extending these rules
It is proved to be very effective.In addition, because new rule is more suitable for hospital and society in view of the history in laboratory after consideration
The case where area.
The correlation rule that will be seen that is converted to alarm rule (passing through syntax conversion), after expert confirms, in intelligent expert
Data verification is used in system.Correlation rule describes the correlation between event and can be considered the rule of probability.If event is frequent
It is observed simultaneously, they will be remembered.
Machine learning data mining technology is applied to one group of infection event, and by learning specific correlation rule,
Learn and analyze the relevance between the relevance of high risk factor and various infection pathogenies, the correlation rule of study includes to come from level
The item of structure different stage.Using the hierarchical relationship between bacterium, antibiotic and infection site, study is established bacterial families, is resisted
The social activity rule of relationship, establishes specific knowledge map for the relationship between three, most between Sheng Su family and infection site family
Afterwards automatically by new and useful Knowledge Aggregation into expert system knowledge base, expert system knowledge base and sense are updated in real time
Knowledge mapping is controlled, the accuracy and accuracy of judgement are improved.
Infection event in infection control management system based on artificial intelligence generally undergoes 5 stages.
First stage: the data containing infection event, the infection control based on artificial intelligence are obtained by clinical database
Management system carries out data with existing system and docks, and is automatically performed the load of infection element information relevant to nosocomial infection, and
It is automatically performed the conversion, calculating and matching work of every institute's sense clinical data, finally constitutes clinical data library module.
Second stage: infection event is analyzed by bacterial strain identification module, which, will since bacterial species and antibioticogram
Division bacteria is corresponding bacterial strain.After classifying to bacterial strain, expert system will analyze each infection event, execute
Short-term and long-term test.For infecting event, is assessed by expert system, the doctor of Microbiological Lab is shown to, by curing
It is raw to determine that any correct bacteria antibiotic figure submit to the definition that clinician treats and which is sent out by expert system
Alarm out.In order to reduce the mistake of artificial and machine to the greatest extent, the doctor of each Microbiological Lab changes the knot of antibiotic map
Fruit, antibiotic map can all be handled again by expert system.The bacterium and relevant antibiosis that expert system module analysis newly separates
Plain knowledge mapping, checks whether relevant antibiotic has been subjected to test, and whether test result meets international standard；Detect patient's
The differentiation of the state of an illness and infection judges whether current therapy is suitable；Whether there is the infection of the state of an illness in detection hospital, if it is
Then sound an alarm.
Phase III: each infection event (related to final antibiotic map) is divided by statistical report module
Analysis, which generates some reports, and makes accordingly to the infection of resident and entire community in ward and entire hospital or community
Report and infection details.Infection event is then reprocessed repeatedly, and is tested for a long time.
Fourth stage: it is verified by the repetition training of artificial neural network and machine learning, obtains optimum prediction model, most
Prediction result and predictive information are with visual interface display in prediction and warning module at last.When predicted value is higher than the pre- of setting
When alert value, the sending alarm of prediction and warning module reminds medical worker to carry out specified region or appointed part science, accurate
Control.
5th stage: machine learning neural network learning and the correlation rule for establishing bacterium, antibiotic and infection site, it will
The relationship of three is established specific knowledge mapping and is integrated into the expert system module.Periodically by machine learning data mining
Technology is applied to one group of infection event, and by learning specific correlation rule, learns and analyze the relevance of high risk factor
New, is finally automatically only integrated into expert system knowledge base, in real time more by the relevance between various infection pathogenies
New expert system knowledge base and sensing control knowledge mapping, improve the accuracy and accuracy of judgement.
Establishing prediction model using regression calculation technology is a kind of typical data digging method.Establish the mistake of prediction model
Journey is as follows: obtaining variable from expert system module, and is divided into response variable and independent variable, then screen to variable, obtains
To the relationship between independent variable and response variable, the candidate independent variable that can preferably predict response variable is identified, certainly by candidate
Input of the variable as prediction model.
In some embodiments, response variable is divided into two kinds, and number, another expression case occur for a kind of expression infection event
Example sum.
As shown in Fig. 2, logistic regression analysis is to simulate the common statistics skill of relationship between dependent variable and one group of independent variable
Art.The criterion of modeling is the functional form of correct preference pattern variable and model.Therefore, it first has to carry out modeling data whole
Reason obtains suitable variable by expert system, response variable and a series of independents variable is arranged.Response variable is divided into two kinds of shapes
Formula.The first type application is in two classification response variables of case rank.1 mark of coding in the response variable of i.e. each infection event
Knowledge event occurs, 0 indicates not occur.Second is applied to two points of response variables for summarizing rank, it is obtained from packet data
Take the information whether occurred about event.Two variables are needed in infection event every group of event is respectively set, number and case occurs
Number of cases, secondly automatically by the relationship between analysis independent variable and response variable, identify preferably to predict selection variables
The candidate independent variable of response variable, and its whole is included in prediction model.Finally, establishing model and carrying out goodness inspection to model
It tests and compares.Common computer model includes integral modeling methodology and gradually modelling, and all possible model all should be according to can
The usability principles that explanatory, terseness, variable are easy to incorporate initial model are tested, and are carried out by addition variable to model
It extends, adjust, compare and tests repeatedly, available optimal models.
For example, carrying out auto-associating Rule Extraction to staphylococcus aureus below.
Staphylococcus aureus: the staphylococcus aureus data set considered includes 7009 records, wherein there is 41 kinds
Different antibiotic.Data set is filtered by removing useless antibiotic, bacterium is (green to the antibiotic of every kind of antibiotic sensitive
Except mycin), staphylococcus aureus to penicillin-susceptible, is known as resistance to mould sometimes sometimes.By filtering data
Collection is reduced to 3734 records.Then Apriori algorithm is used, minimum is supported to be equal to 0.1, and minimum, which is trusted, to be reduced, and is found most
General rule is about 6500.In these rules, 10 are in NCCLS report and expert system knowledge base about golden yellow
10 in staphylococcic 27 rule of color.Two class antibiotic Oxacillin and penicillin are finally obtained (when a bacterium pair
When Oxacillin has drug resistance, it also must be known as drug resistance to any mould) result and to Oxacillin and
The relationship between drug resistance result that the beta-lactam enzyme of penicillin inhibits, wherein must also have in anti-B- to any penicillin
Amide enzyme inhibition.
The technical means disclosed in the embodiments of the present invention is not limited only to technological means disclosed in above embodiment, further includes
Technical solution consisting of any combination of the above technical features.
Claims (8)
1. a kind of infection control management system based on artificial intelligence, it is characterised in that: including middleware module, clinical database
Module, bacterial strain identification module, expert system module, machine learning neural network and prediction and warning module；
The middleware module includes new infections planning system and other hospital data management systems, for completing institute's sense clinic
Conversion, calculating and the matching of data, and it is transmitted to clinical data library module；
The clinical data library module, for providing newest clinic to the expert system module and the bacterial strain identification module
Data；
The bacterial strain identification module, for in bacterial strain infectious bacteria classify, and respectively from different infection event correlations；
The expert system module, for the bacterium newly separated and relevant antibiotic knowledge mapping, for verifying antibiotic
Cultivation results are selected suitable antibiotic inventory and are sounded an alarm to the new isolated bacterium；
The machine learning neural network, for learning and establishing bacterium, antibiotic and the correlation rule of infection site, by three
Relationship establish specific knowledge mapping and be integrated into the expert system module；
The prediction and warning module, establishes prediction model, for predicting the Infection trend in infection data and particular range, works as sense
When the predicted value of dye event is higher than the threshold value of setting, the prediction and warning module issues early warning to medical worker and/or community.
2. infection control management system according to claim 1, it is characterised in that: the middleware module include other
Hospital data management system further includes hospital information system, Laboratory Information Management System, Picture Archive and communication system
And electronic health record.
3. infection control management system according to claim 1, which is characterized in that the process for establishing prediction model is such as
Under: variable is obtained from the expert system module, and is divided into response variable and independent variable, and then the variable is sieved
Choosing, obtains the relationship between the independent variable and the response variable, identifies the time that can preferably predict the response variable
Selected from variable, using the candidate independent variable as the input of the prediction model.
4. infection control management system according to claim 3, which is characterized in that the response variable is divided into two kinds, and one
Kind indicates that number occurs for infection event, and another kind indicates that case is total.
5. infection control management system according to claim 1, it is characterised in that: the system further includes statistical report mould
Block is connect with the machine learning neural network, is analyzed each infection event, and generate statistical report, to ward, doctor
Infection details are presented in institute and/or community.
6. a kind of infection control management method based on artificial intelligence, which is characterized in that including
The data containing infection event are obtained by clinical database；
Bacterial strain identification module analyzes the infection event, since bacterial species and antibioticogram, is by division bacteria
Corresponding bacterial strain；
Machine learning neural network learning and the correlation rule for establishing bacterium, antibiotic and infection site, the relationship of three is built
It founds specific knowledge mapping and is integrated into the expert system module；
Whether the bacterium and relevant antibiotic knowledge mapping that expert system module analysis newly separates, check relevant antibiotic
By test, whether test result meets international standard；The state of an illness of patient and the differentiation of infection are detected, whether judges current therapy
Properly；Whether there is the infection of the state of an illness in detection hospital, if it is sounds an alarm；
By the machine learning neural network repetition training and verifying, optimum prediction model is obtained, prediction result is output to
In prediction and warning module；
The expert system module is updated, the correlation rule and knowledge mapping are stored in the expert system module.
7. the infection control management method according to claim 6 based on artificial intelligence, which is characterized in that the clinic number
According to the data in library from new infections system and other hospital data management systems.
8. the infection control management method according to claim 6 based on artificial intelligence, which is characterized in that the engineering
Practising neural network learning and establishing after the correlation rule of bacterium, antibiotic and infection site further includes that statistical report module is to every
A infection event is analyzed, and generates statistical report, and infection details are presented to ward, hospital and/or community.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910699679.0A CN110415832A (en) | 2019-07-31 | 2019-07-31 | Infection control management system and method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910699679.0A CN110415832A (en) | 2019-07-31 | 2019-07-31 | Infection control management system and method based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110415832A true CN110415832A (en) | 2019-11-05 |
Family
ID=68364556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910699679.0A Pending CN110415832A (en) | 2019-07-31 | 2019-07-31 | Infection control management system and method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110415832A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111128391A (en) * | 2019-12-24 | 2020-05-08 | 北京推想科技有限公司 | Information processing apparatus, method and storage medium |
| CN111816245A (en) * | 2020-07-20 | 2020-10-23 | 成都博欣医学检验实验室有限公司 | Pathogenic microorganism detection method and system combining mNGS and clinical knowledge base |
| CN113130085A (en) * | 2021-03-25 | 2021-07-16 | 边缘智能研究院南京有限公司 | 5G intelligent sensing control prediction system based on big data |
| CN113355851A (en) * | 2020-03-05 | 2021-09-07 | 青岛海尔洗衣机有限公司 | Sterilizing control method for clothes treating apparatus and clothes treating apparatus |
| CN113848821A (en) * | 2020-10-27 | 2021-12-28 | 上海飞机制造有限公司 | Quality detection method, device, equipment and medium based on industrial neural network |
| CN117133459A (en) * | 2023-09-12 | 2023-11-28 | 江苏省人民医院（南京医科大学第一附属医院） | Machine learning-based postoperative intracranial infection prediction method and system |
| TWI826332B (en) * | 2023-06-08 | 2023-12-11 | 宏碁股份有限公司 | Method and system for establishing disease prediction model |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120179491A1 (en) * | 2011-01-07 | 2012-07-12 | Taipei Medical University | High performance and integrated nosocomial infection surveillance and early detection system and method thereof |
| CN202472664U (en) * | 2012-03-01 | 2012-10-03 | 刘晞照 | Hospital infection comprehensive prediction device |
| CN103605911A (en) * | 2013-12-11 | 2014-02-26 | 东南大学 | Intelligent community health care system control method based on Internet of things |
| CN105893725A (en) * | 2014-11-13 | 2016-08-24 | 北京众智汇医科技有限公司 | Management system for an entire process of hospital infection prevention and control, and method thereof |
| CN108231202A (en) * | 2016-12-11 | 2018-06-29 | 哈尔滨光凯科技开发有限公司 | A kind of nosocomial infection approaches to IM based on internet medical services |
| CN109065178A (en) * | 2018-07-11 | 2018-12-21 | 南京医科大学第二附属医院 | A kind of nosocomial infection early warning construction method based on hospital information system |
-
2019
- 2019-07-31 CN CN201910699679.0A patent/CN110415832A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120179491A1 (en) * | 2011-01-07 | 2012-07-12 | Taipei Medical University | High performance and integrated nosocomial infection surveillance and early detection system and method thereof |
| CN202472664U (en) * | 2012-03-01 | 2012-10-03 | 刘晞照 | Hospital infection comprehensive prediction device |
| CN103605911A (en) * | 2013-12-11 | 2014-02-26 | 东南大学 | Intelligent community health care system control method based on Internet of things |
| CN105893725A (en) * | 2014-11-13 | 2016-08-24 | 北京众智汇医科技有限公司 | Management system for an entire process of hospital infection prevention and control, and method thereof |
| CN108231202A (en) * | 2016-12-11 | 2018-06-29 | 哈尔滨光凯科技开发有限公司 | A kind of nosocomial infection approaches to IM based on internet medical services |
| CN109065178A (en) * | 2018-07-11 | 2018-12-21 | 南京医科大学第二附属医院 | A kind of nosocomial infection early warning construction method based on hospital information system |
Non-Patent Citations (10)
| Title |
|---|
| EVELINA LAMMA等: "Artificial Intelligence Techniques for Monitoring", 《IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE》 * |
| EVELINA LAMMA等: "Artificial Intelligence Techniques for Monitoring", 《IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE》, vol. 10, no. 1, 31 January 2006 (2006-01-31), pages 143 - 154 * |
| 张廷轩: "医院感染预警信息的应用价值", 《中国民康医学》 * |
| 张廷轩: "医院感染预警信息的应用价值", 《中国民康医学》, no. 23, 10 December 2016 (2016-12-10) * |
| 林雪峰等: "专家系统在临床微生物实验室信息节点中的应用", 《中国高等医学教育》 * |
| 林雪峰等: "专家系统在临床微生物实验室信息节点中的应用", 《中国高等医学教育》, no. 01, 15 January 2011 (2011-01-15) * |
| 章锐锋等: "社区获得性MRSA感染的临床特征和耐药性分析", 《中华微生物学和免疫学杂志》 * |
| 章锐锋等: "社区获得性MRSA感染的临床特征和耐药性分析", 《中华微生物学和免疫学杂志》, no. 07, 30 July 2004 (2004-07-30) * |
| 胡良平等: "临床科研设计与统计分析", vol. 1, 中国中医药出版社, pages: 275 * |
| 鲁怀伟等: "抗生素敏感试验专家指导系统研究", 《临床输血与检验》, no. 04, 30 December 2004 (2004-12-30) * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111128391A (en) * | 2019-12-24 | 2020-05-08 | 北京推想科技有限公司 | Information processing apparatus, method and storage medium |
| CN113355851A (en) * | 2020-03-05 | 2021-09-07 | 青岛海尔洗衣机有限公司 | Sterilizing control method for clothes treating apparatus and clothes treating apparatus |
| CN111816245A (en) * | 2020-07-20 | 2020-10-23 | 成都博欣医学检验实验室有限公司 | Pathogenic microorganism detection method and system combining mNGS and clinical knowledge base |
| CN111816245B (en) * | 2020-07-20 | 2022-09-09 | 成都博欣医学检验实验室有限公司 | Pathogenic microorganism detection method and system combining mNGS and clinical knowledge base |
| CN113848821A (en) * | 2020-10-27 | 2021-12-28 | 上海飞机制造有限公司 | Quality detection method, device, equipment and medium based on industrial neural network |
| CN113130085A (en) * | 2021-03-25 | 2021-07-16 | 边缘智能研究院南京有限公司 | 5G intelligent sensing control prediction system based on big data |
| TWI826332B (en) * | 2023-06-08 | 2023-12-11 | 宏碁股份有限公司 | Method and system for establishing disease prediction model |
| CN117133459A (en) * | 2023-09-12 | 2023-11-28 | 江苏省人民医院（南京医科大学第一附属医院） | Machine learning-based postoperative intracranial infection prediction method and system |
| CN117133459B (en) * | 2023-09-12 | 2024-04-09 | 江苏省人民医院（南京医科大学第一附属医院） | Machine learning-based postoperative intracranial infection prediction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110415832A (en) | Infection control management system and method based on artificial intelligence | |
| Khalifa et al. | Utilizing health analytics in improving the performance of healthcare services: A case study on a tertiary care hospital | |
| EP2229587B1 (en) | Genome identification system | |
| CN109920508A (en) | prescription auditing method and system | |
| KR101611838B1 (en) | Method and system for predicting high risk based on patient safety | |
| Mohanty et al. | Machine learning for predicting readmission risk among the frail: Explainable AI for healthcare | |
| US20120179491A1 (en) | High performance and integrated nosocomial infection surveillance and early detection system and method thereof | |
| CN105096225A (en) | Analysis system, apparatus and method for assisting disease diagnosis and treatment | |
| CN104246781A (en) | System and method for improving neurologist's workflow on alzheimer's disease | |
| KR20190069687A (en) | Multi questionnaire analysis system for depression diagnosis and the method thereof | |
| US11177024B2 (en) | Identifying and indexing discriminative features for disease progression in observational data | |
| CN111081379B (en) | Disease probability decision method and system thereof | |
| Merone et al. | A decision support system for tele-monitoring COPD-related worrisome events | |
| CN115798734B (en) | New burst infectious disease prevention and control method and device based on big data and storage medium | |
| WO2022211385A1 (en) | Health care consultation system using distribution of disease prediction values | |
| EP4070216A1 (en) | Enhanced data security and access control using machine learning | |
| Vilhena et al. | A case-based reasoning view of thrombophilia risk | |
| Sujatha et al. | A survey of health care prediction using data mining | |
| Blessie et al. | Novel approach for psychiatric patient detection and prediction using data mining techniques | |
| Baiju et al. | Diabetes Retinopathy Prediction Using Multi-model Hyper Tuned Machine Learning | |
| Cabot et al. | Evaluating prediction model performance | |
| Amalarathinam et al. | A New Monotony Advanced Decision Tree Using Graft Algorithm to Predict the Diagnosis of Diabetes Mellitus | |
| Hossain et al. | Applications of artificial intelligence technologies on mental health research during COVID-19 | |
| CN103354929A (en) | System for differentiation and analysis of recorded clinical data | |
| Mathew et al. | A web based decision support system driven for the neurological disorders |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |