CN109636613A - Medical data abnormality recognition method, device, terminal and storage medium - Google Patents
Medical data abnormality recognition method, device, terminal and storage medium Download PDFInfo
- Publication number
- CN109636613A CN109636613A CN201811226809.0A CN201811226809A CN109636613A CN 109636613 A CN109636613 A CN 109636613A CN 201811226809 A CN201811226809 A CN 201811226809A CN 109636613 A CN109636613 A CN 109636613A
- Authority
- CN
- China
- Prior art keywords
- medical data
- disease
- medical
- subclass
- detected
- 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
- 230000005856 abnormality Effects 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000002159 abnormal effect Effects 0.000 claims abstract description 60
- 238000001514 detection method Methods 0.000 claims abstract description 30
- 201000010099 disease Diseases 0.000 claims description 230
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 230
- 238000003745 diagnosis Methods 0.000 claims description 99
- 230000002547 anomalous effect Effects 0.000 claims description 45
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000000354 decomposition reaction Methods 0.000 abstract description 6
- 206010012601 diabetes mellitus Diseases 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000006399 behavior Effects 0.000 description 7
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 7
- 239000010931 gold Substances 0.000 description 7
- 229910052737 gold Inorganic materials 0.000 description 7
- 241001269238 Data Species 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 239000000284 extract Substances 0.000 description 5
- 208000006443 lactic acidosis Diseases 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012163 sequencing technique Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 239000002253 acid Substances 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- JVTAAEKCZFNVCJ-UHFFFAOYSA-N lactic acid Chemical compound CC(O)C(O)=O JVTAAEKCZFNVCJ-UHFFFAOYSA-N 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 231100000572 poisoning Toxicity 0.000 description 2
- 230000000607 poisoning effect Effects 0.000 description 2
- JVTAAEKCZFNVCJ-UHFFFAOYSA-M Lactate Chemical compound CC(O)C([O-])=O JVTAAEKCZFNVCJ-UHFFFAOYSA-M 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000014655 lactic acid Nutrition 0.000 description 1
- 239000004310 lactic acid Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000003607 modifier Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 201000001119 neuropathy Diseases 0.000 description 1
- 230000007823 neuropathy Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 208000033808 peripheral neuropathy Diseases 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 231100001028 renal lesion Toxicity 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/06—Asset management; Financial planning or analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Finance (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Game Theory and Decision Science (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present invention provides a kind of medical data abnormality recognition method, device, terminal and storage medium based on big data, this method comprises: obtaining the medical data to be detected that the abnormality detection requests corresponding objective hospital and the objective hospital when detecting abnormality detection request；The medical data to be detected is grouped according to default dimension, the subclass after being grouped；Each subclass is analyzed based on the corresponding preset rules of default dimension, is judged in the medical data to be detected with the presence or absence of the target medical data for meeting off-note；If there is the target medical data for meeting off-note in the medical data to be detected, the target medical data is subjected to abnormal marking.The present invention can accurately identify decomposition behavior in hospital, effectively prevent insurance fraud behavior occur.
Description
Technical field
The present invention relates to medical data processing technology more particularly to a kind of medical data abnormality recognition method, device,
Terminal and storage medium.
Background technique
Medical insurance system is that country is prevention and a kind of mandatory guarantor of society for sharing medical expense brought by disease
Danger, expense are paid jointly by employing unit and individual, and medical insurance gold is paid by Medical Insurance Organizations, to solve labourer because suffering from
Disease or the bring medical-risk that is hurt.
Medical insurance department can examine hospital, wherein single length of stay is that the important examination of Hospital medical efficiency refers to
Mark, average single length of stay is more, and Hospital medical efficiency is lower, and therefore, in actual operation, hospital is to promote its examination to comment
Point, primary be hospitalized is decomposed into Multiple Hospitalization, or handle falseness and be hospitalized, to reduce single length of stay.This is clearly to violate doctor
Protect the way of regulation, damage public interest.Therefore, it is necessary to a kind of methods for identifying abnormal medical data.
Summary of the invention
The main purpose of the present invention is to provide a kind of medical data abnormality recognition methods, it is intended to which solution can not detect exception
The technical issues of brush medical insurance card behavioral data.
To achieve the above object, the present invention provides a kind of medical data abnormality recognition method, which is characterized in that the medical treatment
Data exception recognition methods the following steps are included:
When detecting abnormality detection request, obtains the abnormality detection and request corresponding objective hospital and the objective hospital
Medical data to be detected；
The medical data to be detected is grouped according to default dimension, the subclass after being grouped；
Analyze each subclass based on the corresponding preset rules of default dimension, judge be in the medical data to be detected
It is no to there is the target medical data for meeting off-note；
If there is the target medical data for meeting off-note in the medical data to be detected, by the target medical treatment
Data carry out abnormal marking.
Optionally, when the default dimension is hospital's dimension, the basis presets dimension for the medical number to be detected
According to being grouped, the step of subclass after being grouped, includes:
The medical data to be detected is grouped according to disease, obtains corresponding first subclass of each disease；
It is described to analyze each subclass based on the corresponding preset rules of default dimension, judge the medical data to be detected
In with the presence or absence of meet off-note target medical data the step of include:
From corresponding first subclass of each disease, obtain each disease total hospitalizations and total length of stay, be based on institute
The total hospitalizations and total length of stay for stating each disease calculate the average single length of stay for obtaining each disease；
The average single length of stay of each disease preset number of days corresponding with each disease is compared, each disease is obtained
The average single length of stay and the number of days of corresponding preset number of days of kind are poor, and the number of days difference for calculating each disease presets day with corresponding
Several ratio obtains the corresponding number of days Outlier factor of each disease；
When the corresponding number of days Outlier factor of any disease is greater than the first preset threshold, corresponding first subclass of the disease
Meet off-note.
Optionally, described to analyze each subclass based on the corresponding preset rules of default dimension, judge described to be detected
Include: with the presence or absence of the step of target medical data for meeting off-note in medical data
From corresponding first subclass of each disease, obtain each disease insured number and total hospitalizations, be based on each disease
The insured number of kind and total hospitalizations, which calculate, obtains everyone corresponding average time in hospital number of each disease；
Everyone corresponding average time in hospital number of each disease is compared with corresponding preset times, obtains the every of each disease
Hospitalizations and the number of corresponding preset times are poor for each person, calculate the number difference of each disease and the ratio of corresponding preset times
Value, obtains the corresponding frequency abnormality factor of each disease；
When the corresponding frequency abnormality factor of any disease is greater than the second preset threshold, corresponding first subclass of the disease
Meet off-note.
Optionally, described to analyze each subclass based on the corresponding preset rules of default dimension, judge described to be detected
Include: with the presence or absence of the step of target medical data for meeting off-note in medical data
From corresponding first subclass of each disease, total hospitalizations of each disease and the amount of money of being always hospitalized are obtained, institute is based on
The total hospitalizations and total amount of money of being hospitalized for stating each disease, which calculate, to be obtained the average single of each disease and is hospitalized the amount of money；
The average single of each disease amount of money of being hospitalized is compared with corresponding preset cost, the flat of each disease is obtained
The equal single amount of money of being hospitalized is poor with the amount of money of corresponding preset cost, calculates the amount of money difference of each disease and the ratio of correspondence preset cost
Value, obtains the corresponding amount of money Outlier factor of each disease；
When the corresponding amount of money Outlier factor of any disease is greater than third predetermined threshold value, corresponding first subclass of the disease
Meet off-note.
Optionally, when the default dimension is personal factor, the basis presets dimension for the medical number to be detected
According to being grouped, the step of subclass after being grouped, includes:
The medical data to be detected is grouped according to insured people, the corresponding second subset of each insured people is obtained and closes；
It is described to analyze each subclass based on the corresponding preset rules of default dimension, judge the medical data to be detected
In with the presence or absence of meet off-note target medical data the step of include:
The corresponding consultation time of every medical data in the second subset conjunction is extracted, calculates and obtains the second subset conjunction
In adjacent medical data adjacent medical interval；
Preset time interval threshold value is obtained, the time at the adjacent medical interval and the time interval threshold value is calculated
Difference calculates the ratio of the time difference Yu the time interval threshold value, obtains every medical data pair in the second subset conjunction
The interval Outlier factor answered；
When the corresponding interval Outlier factor of medical data in second subset conjunction is greater than four preset thresholds, the doctor
It treats data and meets off-note.
Optionally, described to obtain preset time interval threshold value, calculate the adjacent medical interval and the time interval
The time difference of threshold value calculates the ratio of the time difference Yu the time interval threshold value, obtains in the second subset conjunction every
Include: after the step of medical data corresponding interval Outlier factor
The corresponding diagnostic result of every medical data in the second subset conjunction is extracted, judges that the second subset is in closing
The no adjacent medical data that there is corresponding similar diagnosis result；
If the second subset has the adjacent medical data for corresponding to similar diagnosis result in closing, the adjacent doctor is obtained
The time difference for treating the adjacent medical interval and the time interval threshold value of data, calculate the time difference and the time interval threshold
The ratio of value obtains the corresponding interval Outlier factor of the adjacent medical data；
When the corresponding interval Outlier factor of the adjacent medical data is greater than five preset thresholds, which meets
Off-note.
Optionally, described to analyze each subclass based on the corresponding preset rules of default dimension, judge described to be detected
Include: with the presence or absence of the step of target medical data for meeting off-note in medical data
The corresponding diagnosis and treatment item of every medical data in the second subset conjunction is extracted, by adjacent corresponding examine of going to a doctor
Treatment project with compared respectively with reference to diagnosis and treatment item set, obtain it is adjacent go to a doctor corresponding diagnosis and treatment item with reference to diagnosis and treatment
The coincidence ratio of project set, and all adjacent medical average coincidence ratios are calculated, obtain adjacent medical diagnosis and treatment item
Continuation degree；
When the continuation degree is greater than six preset thresholds, corresponding medical data meets off-note.
In addition, to achieve the above object, the present invention also provides a kind of medical data anomalous identification device, the medical datas
Anomalous identification device includes:
Module is obtained, for the abnormality detection being obtained and requesting corresponding objective hospital when detecting abnormality detection request
With the medical data to be detected of the objective hospital；
Grouping module, for the medical data to be detected to be grouped according to default dimension, the son after being grouped
Set；
Anomaly analysis module, for analyzing each subclass based on the corresponding preset rules of default dimension, described in judgement
With the presence or absence of the target medical data for meeting off-note in medical data to be detected；
Abnormal marking module, if for there is the target medical treatment number for meeting off-note in the medical data to be detected
According to then by target medical data progress abnormal marking.
In addition, to achieve the above object, the present invention also provides a kind of medical data anomalous identification terminal, the medical datas
Anomalous identification terminal includes processor, memory and is stored in the doctor that can be executed on the memory and by the processor
Data exception recognizer is treated, wherein realizing when the medical data anomalous identification program is executed by the processor as above-mentioned
Medical data abnormality recognition method the step of.
In addition, to achieve the above object, the present invention also provides a kind of storage medium, being stored with medical treatment on the storage medium
Data exception recognizer, wherein realizing such as above-mentioned medical treatment when the medical data anomalous identification program is executed by processor
The step of data exception recognition methods.
The embodiment of the present invention can be detected effectively by being grouped analysis to medical data to be detected according to default dimension
Abnormal medical data；To there is the target medical data for meeting off-note in medical data to be detected and carries out abnormal marking, it can
Abnormal medical data is distinguished, specific aim processing, the reimbursement of specification medical expense are carried out to abnormal medical data so as to subsequent
The normal operation of system.
Detailed description of the invention
Fig. 1 is the medical data anomalous identification terminal structure signal for the hardware running environment that the embodiment of the present invention is related to
Figure；
Fig. 2 is the flow diagram of medical data abnormality recognition method first embodiment of the present invention；
Fig. 3 is the flow diagram of medical data abnormality recognition method second embodiment of the present invention；
Fig. 4 is the flow diagram of the 5th embodiment of medical data abnormality recognition method of the present invention；
Fig. 5 is the functional block diagram of medical data anomalous identification device first embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Referring to Figure 1, Fig. 1 is the hardware structural diagram of medical data anomalous identification terminal provided by the present invention.
The medical data anomalous identification terminal can be PC, be also possible to smart phone, tablet computer, portable calculating
The device end having a display function such as machine, desktop computer, optionally, the medical data anomalous identification terminal can be clothes
Business device equipment, there are the rear end management system of medical data anomalous identification, user is by the rear end management system to medical number
It is managed according to anomalous identification terminal.
The medical data anomalous identification terminal may include: the components such as processor 101 and memory 201.Described
In medical data anomalous identification terminal, the processor 101 is connect with the memory 201, is stored on the memory 201
Medical data anomalous identification program, processor 101 can call the medical data anomalous identification program stored in memory 201,
And the step of realizing embodiment as each such as following medical data abnormality recognition methods.
The memory 201 can be used for storing software program and various data.Memory 201 can mainly include storage
Program area and storage data area, wherein storing program area can application program needed for storage program area, at least one function
(such as medical data anomalous identification program) etc.；Storage data area may include database, such as the present invention need to inquire acquisition effectively
Historical record etc..In addition, memory 201 may include high-speed random access memory, it can also include nonvolatile memory,
A for example, at least disk memory, flush memory device or other volatile solid-state parts.
Processor 101 is the control centre of medical data anomalous identification terminal, entire using various interfaces and connection
The various pieces of medical data anomalous identification terminal, by run or execute the software program being stored in memory 201 and/or
Module, and the data being stored in memory 201 are called, execute the various functions and processing of medical data anomalous identification terminal
Data, to carry out integral monitoring to medical data anomalous identification terminal.Processor 101 may include that one or more processing are single
Member；Optionally, processor 101 can integrate application processor and modem processor, wherein application processor mainly handles behaviour
Make system, user interface and application program etc., modem processor mainly handles wireless communication.It is understood that above-mentioned
Modem processor can not also be integrated into processor 101.
It will be understood by those skilled in the art that medical data anomalous identification terminal structure shown in Fig. 1 is not constituted pair
The restriction of medical data anomalous identification terminal may include components more more or fewer than diagram, or combine certain components, or
The different component layout of person.
Based on above-mentioned hardware configuration, each embodiment of the method for the present invention is proposed, identification terminal hereinafter is medical number
According to the abbreviation of anomalous identification terminal.
The present invention provides a kind of medical data abnormality recognition method.
It is the flow diagram of medical data abnormality recognition method first embodiment of the present invention referring to Fig. 2, Fig. 2.
In the present embodiment, the medical data abnormality recognition method the following steps are included:
Step S10 obtains the abnormality detection and requests corresponding objective hospital and described when detecting abnormality detection request
The medical data to be detected of objective hospital；
Abnormality detection request is that triggering medical data method for detecting abnormality of the present invention corresponds to medical data abnormality detecting program
The instruction of starting can trigger abnormality detection request by a preset external event/page request, as user input it is different
The clicking operation often detected；It can also be by using the mode of timing JOB (task) as medical data abnormality detecting program
Start interface, triggering is carried out at regular intervals to the abnormality detection of medical data, medical number to be detected can also received
It carries out abnormality detection, can also be carried out after the data volume of received medical data to be detected reaches certain threshold value different immediately after
Often detection.Optionally, the present invention can be one or more of medical data auditing flow to the abnormality detection of medical data
Flow nodes, can be by the abnormality detecting program of the circulation automatic trigger medical data of flow nodes.
In the present embodiment, to identify using hospital as the abnormal medical data of violation main body main body, with target to be detected doctor
The medical data of institute is the basis of anomalous identification.It include the mark of corresponding objective hospital in abnormality detection request.Doctor to be detected
Treat the specific medical data of objective hospital that the request of data, that is, abnormality detection is directed toward, request detection, medical data include but
It is not limited to: visit type, diagnosis and treatment item mark, consultation time, diagnostic result, medicining condition, the diagnosis and treatment amount of money, charges for drug etc..
After insured people's brush medical insurance card, medical data can be uploaded to hospital terminal by card swiping device, then by hospital terminal
It is sent to identification terminal or is uploaded directly into identification terminal, can directly be stored in the memory of identification terminal, it is different in progress
Often when detection, identification terminal directly obtains medical data to be detected according to abnormality detection request from local storage, can also be with
It is stored in data storage terminal/cloud of medical data, so that identification terminal is when carrying out the detection of abnormal medical data, according to
Abnormality detection request obtains medical data to be detected from data storage terminal/cloud.
The medical data to be detected is grouped, the subclass after being grouped by step S20 according to default dimension；
Default dimension includes at least one of hospital's dimension and personal factor, and hospital's dimension is i.e. based on the complete of objective hospital
Portion's Analysis of Medical Treatment Data judges whether there is abnormal medical data, and personal factor is that the Analysis of Medical Treatment Data based on each insured people is sentenced
It is disconnected to whether there is abnormal medical data.Each default dimension can be set to that control can be clicked, and is determined and is used according to user's clicking operation
The default dimension of the selected analysis in family, can also directly obtain the default dimension of default.
Default dimension is different, and the packet mode being grouped to medical data to be detected is different.It is hospital in default dimension
When, medical data to be detected is grouped by identification terminal according to disease, can get the corresponding subclass of each disease；In default dimension
When degree is personal, medical data to be detected is grouped by identification terminal according to insured people, can get the corresponding son of each insured people
Set.
Optionally, identification terminal can analyze the medical number to be detected according to hospital's dimension and personal factor simultaneously, respectively
According to identifying abnormal data.
Step S30 analyzes each subclass based on the corresponding preset rules of default dimension, judges the medical treatment to be detected
With the presence or absence of the target medical data for meeting off-note in data；
Each default dimension has corresponding preset rules, and each subset is combined into the analysis object of preset rules.Specifically
In one embodiment, when the default dimension is hospital's dimension, the medical data to be detected is divided according to disease for ground
Group obtains corresponding first subclass of each disease, and it is corresponding average to calculate each disease based on corresponding first subclass of each disease
Single length of stay and/or everyone average time in hospital number and/or average single are hospitalized the amount of money, and each disease is corresponding average single
Secondary length of stay and/or everyone average time in hospital number and/or average single be hospitalized amount of money preset value corresponding with each disease into
Row compares, and judges whether corresponding first subclass of each disease meets off-note.
Off-note refers to the preset feature for being used to limit abnormal medical data, for example, each disease is corresponding flat
Equal single length of stay preset value corresponding with each disease is compared, if the corresponding average single length of stay of any disease is big
In the corresponding preset value of each disease, then corresponding first subclass of the disease meets off-note.
Step S40 will be described if there is the target medical data for meeting off-note in the medical data to be detected
Target medical data carries out abnormal marking.
If medical data to be detected has abnormal target medical data, target medical data is subjected to abnormal marking,
And output abnormality testing result, to prompt user verify under line.If there is no abnormal target doctors for medical data to be detected
Data are treated, then are prompted without exception.
Further, preferably to carry out abnormal marking, so that identification terminal and/or user select target medical data
Corresponding processing mode is handled, and the Exception Type of target medical data is obtained；It is obtained according to the Exception Type corresponding different
Often mark carries out abnormal marking to medical data to be detected.Further, the amendment inputted after being verified in the case where detecting user's line
When instruction, the status modifier by target medical data is normal.
The present embodiment by when detecting abnormality detection request, obtain the abnormality detection request corresponding objective hospital and
The medical data to be detected of the objective hospital；The medical data to be detected is grouped according to default dimension, is divided
Subclass after group；Each subclass is analyzed based on the corresponding preset rules of default dimension, judges the medical number to be detected
With the presence or absence of the target medical data for meeting off-note in；If existing in the medical data to be detected and meeting off-note
Target medical data, then by the target medical data carry out abnormal marking.By with the mesh of pending/progress anomalous identification
Whether mark hospital is screening conditions, obtain the medical data to be detected of objective hospital, can targetedly violate to objective hospital
Medical insurance specification carries out analysis identification；By being grouped analysis to medical data to be detected according to default dimension, can effectively detect
Abnormal medical data out；To there is the target medical data for meeting off-note in medical data to be detected and carry out abnormal marking,
Abnormal medical data can be distinguished, specific aim processing, specification medical expense report are carried out to abnormal medical data so as to subsequent
The normal operation of pin system.
Further, based on the above embodiment, such as Fig. 3, in medical data abnormality recognition method second embodiment of the present invention
In, when the default dimension is hospital's dimension, the step S20 includes:
The medical data to be detected is grouped according to disease, obtains corresponding first subset of each disease by step S21
It closes；
Disease, including main diagnosis (such as diabetes) and secondary diagnosis/complication (renal lesions, eyeground pathological changes, neuropathy
Deng), it in grouping, diagnoses, i.e., medical data is grouped according to subdivision disease according to main diagnosis+pair.
The step S30 includes:
Step S310 obtains total hospitalizations of each disease and day of being always hospitalized from corresponding first subclass of each disease
Number, total hospitalizations and total length of stay based on each disease calculate the average single length of stay for obtaining each disease；
Corresponding first subclass of each disease, includes the corresponding all medical datas of each disease of objective hospital, not with insured
It is artificial to distinguish standard.From corresponding first subclass of each disease, counts each disease and correspond to total hospitalizations and total be hospitalized
Number of days, average single length of stay can be obtained by total length of stay divided by total hospitalizations.For example, diabetes complicated lactic acid
Property acid poisoning, total degree is 20 times in hospital, total length of stay 140 days, then the average single length of stay of the disease be 140/20
=7 days/time.
The average single length of stay of each disease refers to the average time that the single of certain disease is hospitalized.The average list of different diseases
Secondary length of stay is different, therefore, calculates the corresponding average single of each disease based on corresponding first subclass of each disease and is hospitalized day
Number.
It optionally, can be first corresponding by the disease to guarantee that data are more accurate and guaranteeing to accidental unexpected compatibility
In first subclass, single length of stay is marked above/below the medical data of the first preset value, is further verified, together
When, the average single length of stay of the disease is calculated according to remaining medical data.It in example as above, is hospitalized at 20 times, firmly
Institute's number of days is between 4-10 days mostly, but has once been hospitalized 2 days, has once been hospitalized 14 days, then by the two exceptions
Data markers do further verification, and by the two abnormal data eliminations, calculate average single based on remaining 18 data and live
Institute's number of days.
Step S311 compares the average single length of stay of each disease preset number of days corresponding with each disease
Compared with average single length of stay and the number of days of corresponding preset number of days for obtaining each disease are poor, and the number of days for calculating each disease is poor
With the ratio of corresponding preset number of days, the corresponding number of days Outlier factor of each disease is obtained；
Preset number of days according to Jing Guo Shen He same level and generic medical institutions in average single under same disease
Length of stay, which calculates, to be obtained, and same level refers to that general Hospital Grade, classifications such as " Grade A hospital " " diformazan hospitals " refer to that training is identical, example
Children's training is such as belonged to, same disease includes main diagnosis and secondary diagnosis.Preset number of days can be a range.
Number of days Outlier factor refers to the parameter for identifying average single length of stay intensity of anomaly.It, will be each in the present embodiment
The average single length of stay and the number of days of corresponding preset number of days of disease are poor, with the ratio of corresponding preset number of days as each disease pair
The number of days Outlier factor answered.For example, if the preset number of days of diabetes complicated lactic acidosis is 15 days, and objective hospital
Average single length of stay is 7 days, then differs 8 days, then calculate day according to preset number of days, average single length of stay, number of days difference
Outlier factor is counted, the number of days Outlier factor in above-mentioned example is (15-7)/15.
When number of days Outlier factor=0, i.e., average single length of stay is 0 with the number of days difference of corresponding preset number of days, explanation
The average single length of stay of corresponding disease is without exception；When number of days Outlier factor < 0, i.e. average single length of stay is greater than pair
Preset number of days is answered, illustrates that corresponding disease can not exist to decompose and is hospitalized, is i.e. average single length of stay is without exception；When number of days exception
When the factor > 0, i.e., average single length of stay is less than corresponding preset number of days, illustrates that the average single length of stay of corresponding disease has
Abnormal, numerical values recited identifies intensity of anomaly.
Step S312, when the corresponding number of days Outlier factor of any disease is greater than the first preset threshold, the disease is corresponding
First subclass meets off-note.
First preset threshold refers to the value by user preset, can be adjusted according to the actual situation.
It optionally, can be different according to the corresponding number of days of each disease after the corresponding number of days Outlier factor of each disease is calculated
Constant factor calculates the abnormal scoring of objective hospital.It is when carrying out whole examination to objective hospital, the corresponding number of days of each disease is different
Constant factor integrates the abnormal scoring for calculating objective hospital, it is alternatively possible to regard the sum of all number of days Outlier factors as mesh
The abnormal scoring of hospital is marked, a reference value can also be distributed according to the unstable degree (the biggish disease of individual difference) of each disease,
It scores using the corresponding a reference value of each disease and the sum of products of number of days Outlier factor as the exception of objective hospital.
The present embodiment obtains the average single length of stay of each disease, and putting down each disease of objective hospital by calculating
Average single length of stay (the i.e. default day of equal single length of stay and same level and the same disease of generic medical institutions
Number) it compares, it identifies in the medical data to be detected of objective hospital and does not meet the medical data of usual situation, and pass through day
Number Outlier factor characterizes intensity of anomaly, is greater than the first preset threshold in number of days Outlier factor, by corresponding first subclass of disease
It is judged to meeting off-note, realizes the identification to abnormal medical data.
Further, based on the above embodiment, described in medical data abnormality recognition method 3rd embodiment of the present invention
Step S30 includes:
Step S313, from corresponding first subclass of each disease, obtain each disease insured number and total hospitalizations,
Insured number and total hospitalizations based on each disease, which calculate, obtains everyone corresponding average time in hospital number of each disease；
Corresponding first subclass of each disease includes the corresponding all medical datas of each disease of objective hospital.From each disease
It in corresponding first subclass, counts each disease and corresponds to insured number and total hospitalizations, everyone can lead to average time in hospital number
Total hospitalizations are crossed to obtain divided by insured number.For example, diabetes complicated lactic acidosis, total hospitalizations are 100 times, ginseng
Guarantor's number 14, then everyone corresponding average time in hospital number of the disease is 100/14.
Everyone corresponding average time in hospital number of each disease is compared by step S314 with corresponding preset times, is obtained
Everyone average time in hospital number and the number of corresponding preset times of each disease are poor, calculate the number difference of each disease with it is corresponding pre-
If the ratio of number, the corresponding frequency abnormality factor of each disease is obtained；
Preset times according to Jing Guo Shen He same level and generic medical institutions under same disease same diagnosis it is (main
Diagnosis/pair diagnosis) average time in hospital number calculate and obtain, preset times can be equal to all same levels and generic medical institutions
In the same diagnosis of same disease (main diagnosis/pair diagnoses) average time in hospital number.
The frequency abnormality factor refers to the parameter for identifying everyone average time in hospital frequency abnormality degree.It, will be each in the present embodiment
Everyone average time in hospital number and the number of corresponding preset times of disease are poor, with the ratio of corresponding preset times as each disease pair
The frequency abnormality factor answered.For example, if everyone average time in hospital number of diabetes complicated lactic acidosis is 10 times, and mesh
Everyone the average time in hospital number for marking hospital is 17 times, is differed 7 times, then poor according to preset times, everyone average time in hospital number, number
Calculation times Outlier factor, the frequency abnormality factor in above-mentioned example are (17-10)/10.
When the frequency abnormality factor=0, i.e., everyone average time in hospital number is 0 with the number difference of corresponding preset times, explanation
Everyone average time in hospital number of corresponding disease is without exception；When the frequency abnormality factor < 0, i.e. everyone average time in hospital number is less than pair
Preset times are answered, illustrates that corresponding disease can not exist to decompose and is hospitalized, i.e., everyone average time in hospital number is without exception；Work as frequency abnormality
When the factor > 0, i.e., everyone average time in hospital number is greater than corresponding preset times, illustrates that everyone average time in hospital number of corresponding disease has
Abnormal, numerical values recited identifies intensity of anomaly, and numerical value is bigger, and intensity of anomaly is bigger.
Step S315, when the corresponding frequency abnormality factor of any disease is greater than the second preset threshold, the disease is corresponding
First subclass meets off-note.
Second preset threshold refers to the value by user preset, can be adjusted according to the actual situation.
It optionally, can be different according to the corresponding number of each disease after the corresponding frequency abnormality factor of each disease is calculated
Constant factor calculates the abnormal scoring of objective hospital.It is when carrying out whole examination to objective hospital, the corresponding number of each disease is different
Constant factor integrates the abnormal scoring for calculating objective hospital, it is alternatively possible to regard the sum of all frequency abnormality factors as mesh
Mark the abnormal scoring of hospital；A reference value can also be distributed according to the unstable degree (the biggish disease of individual difference) of each disease,
It scores the sum of products of the corresponding a reference value of each disease and the frequency abnormality factor as the exception of objective hospital；It can also will be each
The corresponding number of days Outlier factor of disease and the frequency abnormality factor integrate the whole abnormal scoring for calculating objective hospital, participate in dimension
Degree is more, and abnormal scoring is more accurate.
The present embodiment obtains everyone average time in hospital number of each disease by calculating, and by the every of each disease of objective hospital
Everyone average time in hospital number of hospitalizations and same level and the same disease of generic medical institutions is (i.e. default secondary for each person
Number) it compares, identify the medical data that usual situation is not met in the medical data to be detected of objective hospital, and by secondary
Number Outlier factor characterizes intensity of anomaly, is greater than the second preset threshold in the frequency abnormality factor, by corresponding first subclass of disease
It is judged to meeting off-note, realizes the identification to abnormal medical data.
Further, based on the above embodiment, described in medical data abnormality recognition method fourth embodiment of the present invention
Step S30 includes:
Step S316 obtains the total hospitalizations and total gold of being hospitalized of each disease from corresponding first subclass of each disease
Volume, total hospitalizations based on each disease and total amount of money of being hospitalized, which calculate, to be obtained the average single of each disease and is hospitalized the amount of money；
From corresponding first subclass of each disease, counts each disease and corresponds to total hospitalizations and total amount of money of being hospitalized,
The average single amount of money of being hospitalized can be obtained by the amount of money of being always hospitalized divided by total hospitalizations.For example, diabetes complicated lactate acid
Poisoning, total amount is 100,000 in hospital, total hospitalizations 20 times, then the average single of the disease is hospitalized the amount of money as 100000/20=
5000 yuan/time.
The average single of each disease be hospitalized the amount of money refer to certain disease single be hospitalized average amount.The average list of different diseases
The secondary amount of money of being hospitalized is different, therefore, calculates the corresponding average single of each disease gold in hospital based on corresponding first subclass of each disease
Volume.
It optionally, can be first corresponding by the disease to guarantee that data are more accurate and guaranteeing to accidental unexpected compatibility
In first subclass, the single amount of money of being hospitalized is marked above/below the medical data of the first preset value, is further verified, together
When, it is hospitalized the amount of money according to the average single that remaining medical data calculates the disease.
The average single of each disease amount of money of being hospitalized is compared with corresponding preset cost, obtains by step S317
The average single of each disease be hospitalized the amount of money and the amount of money of corresponding preset cost it is poor, calculate the amount of money difference of each disease with it is corresponding pre-
If the ratio of the amount of money, the corresponding amount of money Outlier factor of each disease is obtained；
Preset cost according to Jing Guo Shen He same level and generic medical institutions in average single under same disease
The amount of money, which calculates, in hospital obtains, and same level refers to that general Hospital Grade, classifications such as " Grade A hospital " " diformazan hospitals " refer to that training is identical, example
Children's training is such as belonged to, same disease includes main diagnosis and secondary diagnosis.Preset cost can be a numberical range.
Amount of money Outlier factor refers to and is hospitalized the parameter of amount of money intensity of anomaly for identifying average single.It, will be each in the present embodiment
The average single of disease be hospitalized the amount of money and the amount of money of corresponding preset cost it is poor, with the ratio of corresponding preset cost as each disease
The corresponding amount of money Outlier factor of kind.If for example, the preset cost of diabetes complicated lactic acidosis be 10000 yuan/time, and
The average single of objective hospital is hospitalized the amount of money as 5000 yuan/time, then differs 5000 yuan/time, then according to preset cost, average single
The amount of money, amount of money difference calculate amount of money Outlier factor in hospital, and the amount of money Outlier factor in above-mentioned example is (10000-5000)/10000.
If average single is hospitalized, the amount of money is fewer, then illustrates to be shared on being in addition once hospitalized, that is, have what decomposition was hospitalized
It may.
When amount of money Outlier factor < 0 or=1, illustrate that the average single of the corresponding disease amount of money of being hospitalized is without exception, when the amount of money is different
When constant factor < 1, illustrate there is exception, numerical value is smaller, and intensity of anomaly is bigger.
Step S318, when the corresponding amount of money Outlier factor of any disease is greater than third predetermined threshold value, the disease is corresponding
First subclass meets off-note.
Third predetermined threshold value refers to the value by user preset, can be adjusted according to the actual situation.
It optionally, can be different according to the corresponding amount of money of each disease after the corresponding amount of money Outlier factor of each disease is calculated
Constant factor calculates the abnormal scoring of objective hospital.It is when carrying out whole examination to objective hospital, the corresponding amount of money of each disease is different
Constant factor integrates the abnormal scoring for calculating objective hospital, it is alternatively possible to regard the sum of all amount of money Outlier factors as mesh
Mark the abnormal scoring of hospital；A reference value can also be distributed according to the unstable degree (the biggish disease of individual difference) of each disease,
It scores using the corresponding a reference value of each disease and the sum of products of amount of money Outlier factor as the exception of objective hospital；It may be based on gold
Volume Outlier factor, number of days Outlier factor and the frequency abnormality factor calculate the abnormal scoring of objective hospital.
The present embodiment obtains the average single of each disease and is hospitalized the amount of money by calculating, and by the flat of each disease of objective hospital
The be hospitalized average single of same disease of the amount of money and same level and generic medical institutions of equal single is hospitalized the amount of money (i.e. default gold
Volume) it compares, it identifies in the medical data to be detected of objective hospital and does not meet the medical data of usual situation, and pass through gold
Volume Outlier factor characterizes intensity of anomaly, is greater than third predetermined threshold value in amount of money Outlier factor, by corresponding first subclass of disease
It is judged to meeting off-note, realizes the identification to abnormal medical data.
Further, based on the above embodiment, such as Fig. 4, in the 5th embodiment of medical data abnormality recognition method of the present invention
In, when the default dimension is personal factor, the step S20 includes:
The medical data to be detected is grouped by step S22 according to insured people, obtains each insured people corresponding second
Subclass；
With insured artificial unit, the corresponding medical data to be detected of each insured people in medical data to be detected is obtained, including
Insured people swipe the card every time medical visit type, diagnosis and treatment item mark, consultation time, diagnostic result, medicining condition, diagnosis and treatment gold
Volume, charges for drug etc..The corresponding medical data collection to be detected of each insured people is combined into second subset conjunction.
The step S30 includes:
Step S320 extracts the corresponding consultation time of every medical data in the second subset conjunction, calculates described in obtaining
The adjacent medical interval of adjacent medical data in second subset conjunction；
Identification terminal can obtain consultation time medical every time from medical data, wherein medical every time all to generate one
Corresponding medical data, is arranged according to consultation time sequencing, then the adjacent medical data of calculated permutations it is adjacent it is medical between
Every.
In one embodiment, identification terminal obtains the adjacent medical interval of all adjacent medical datas in second subset conjunction；Separately
In one embodiment, second subset is closed and is grouped again according to disease, obtains third subclass, it will be each in third subclass
Medical data is arranged according to consultation time sequencing, calculate identical disease it is adjacent be hospitalized it is medical it is adjacent it is medical between
Every.
Step S321 obtains preset time interval threshold value, calculates the adjacent medical interval and the time interval threshold
The time difference of value calculates the ratio of the time difference Yu the time interval threshold value, obtains every doctor in the second subset conjunction
Treat the corresponding interval Outlier factor of data；
Preset time interval threshold value is obtained, judges whether the corresponding adjacent medical interval of each medical data is less than the time
Interval threshold illustrates that this medical data is hospitalized behavior there may be suspicious decomposition if being less than.In one embodiment,
Identification terminal can calculate the difference at adjacent medical interval and time interval threshold value, determine that every medical data is corresponding according to the difference
Interval Outlier factor, when it is adjacent it is medical interval be less than time interval threshold value when, the difference of the two is bigger, interval Outlier factor get over
Greatly, when adjacent medical interval is greater than or equal to time interval threshold value, the difference of the two is bigger, and interval Outlier factor is smaller.
Because adjacent medical interval is referred at least to two medical datas, in the present embodiment, every medical data is corresponding extremely
A few interval Outlier factor, for example, [1,2], [2,3] etc., wherein number refers to the serial number of the hospitalizations of single insured people,
It arranges sequentially in time, then for the 2nd corresponding medical data of being hospitalized, at least corresponds to two interval Outlier factors, respectively
For the 2nd be hospitalized the interval Outlier factor being hospitalized with the 1st time and the interval Outlier factor being hospitalized with the 3rd time of being hospitalized for the 2nd time.
Step S322, the corresponding interval Outlier factor of medical data in second subset conjunction are greater than the 4th default threshold
When value, which meets off-note.
4th preset threshold refers to the value by user preset, can be adjusted according to the actual situation.
The present embodiment calculates by extracting the corresponding consultation time of every medical data in the second subset conjunction and obtains institute
State the adjacent medical interval of adjacent medical data in second subset conjunction；Preset time interval threshold value is obtained, is calculated described adjacent
The time difference at medical interval and the time interval threshold value, the ratio of the time difference Yu the time interval threshold value are calculated, is obtained
The corresponding interval Outlier factor of every medical data into second subset conjunction；Medical data in second subset conjunction
When corresponding interval Outlier factor is greater than four preset thresholds, which meets off-note, because for the same ginseng
Guarantor is then likely that there are the suspicion decomposed and be hospitalized if the adjacent interval duration being hospitalized is too short, by calculating adjacent medical number
According to adjacent medical interval, and the big medical number of suspicious degree is filtered out based on the corresponding interval Outlier factor of every medical data
According to whether being hospitalized behavior comprising suspicious decomposition in accurate judgement medical data, effectively prevent insurance fraud behavior occur, reduce public
Many interests losses.
Further, include: after the step of step S321
Step S323 extracts the corresponding diagnostic result of every medical data in the second subset conjunction, judges described second
With the presence or absence of the adjacent medical data of corresponding similar diagnosis result in subclass；
Diagnostic result refer to doctor for insured people's disease by judgement, for example, be diagnosed as diabetes complicated lactic acidosis,
Diagnostic result includes main diagnosis and secondary diagnosis.
Similar diagnosis result in the present embodiment includes same diagnostic result and similar diagnosis as a result, diagnosis can be pre-established
The different same diagnostic result of title and similar diagnostic result are stored in diagnostic result library, are judging adjacent doctor by results repository
When whether the diagnostic result for treating data is similar diagnosis result, pass through the judgement of inquiry diagnostic result library.
Step S323 is obtained if there is the adjacent medical data of corresponding similar diagnosis result in second subset conjunction
Adjacent medical interval and the time difference of the time interval threshold value of the adjacent medical data, calculate the time difference with it is described
The ratio of time interval threshold value obtains the corresponding interval Outlier factor of the adjacent medical data；
If the adjacent medical data of certain two or more pieces correspond to similar diagnosis as a result, if the adjacent medical data of the two or more pieces
Have very big decomposition be hospitalized may, then further obtain the common corresponding interval exception of the adjacent medical data of the two or more pieces because
Son judges the consultation time of the adjacent medical data of the two or more pieces with the presence or absence of abnormal (abnormal if it exists, then intensity of anomaly
Size is how many).
Step S324, when the corresponding interval Outlier factor of the adjacent medical data is greater than five preset thresholds, the doctor
It treats data and meets off-note.
When the corresponding interval Outlier factor of adjacent medical data is greater than five preset thresholds, adjacent doctor is further increased
Treating data is to decompose the possibility of inpatient medical data.Because if the corresponding consultation time of adjacent medical data is closely spaced, together
When diagnostic result it is same or similar, then the adjacent medical data be possible for decompose be hospitalized data.
The present embodiment by extract the second subset close in the corresponding diagnostic result of every medical data, judge described the
With the presence or absence of the adjacent medical data of corresponding similar diagnosis result in two subclass；If there is corresponding phase in closing in the second subset
Like the adjacent medical data of diagnostic result, then obtain the adjacent medical data adjacent medical interval and the time interval threshold
The time difference of value calculates the ratio of the time difference Yu the time interval threshold value, and it is corresponding to obtain the adjacent medical data
It is spaced Outlier factor；When the corresponding interval Outlier factor of the adjacent medical data is greater than five preset thresholds, the medical treatment number
According to meeting off-note, it may be assumed that by the diagnostic result of adjacent medical data and adjacent medical interval as judging adjacent medical data
Whether meet the basic parameter of off-note, can more precisely identify decomposition behavior in hospital, effectively prevent insurance fraud row occur
To reduce public interest loss.
Further, based on the above embodiment, described in medical data abnormality recognition method sixth embodiment of the present invention
Step S30 includes:
Step S325 extracts the corresponding diagnosis and treatment item of every medical data in the second subset conjunction, will be adjacent medical each
Self-corresponding diagnosis and treatment item compares respectively with reference to diagnosis and treatment item set, obtains adjacent corresponding diagnosis and treatment item of going to a doctor
It is overlapped ratio with reference diagnosis and treatment item set, and calculates all adjacent medical average coincidence ratios, is obtained adjacent medical
The continuation degree of diagnosis and treatment item；
After identification terminal arranges each going to a doctor in hospital according to consultation time sequencing, adjacent live can be obtained
The diagnosis and treatment item mark for including during institute is medical, and adjacent corresponding diagnosis and treatment item of going to a doctor is divided with reference to diagnosis and treatment item set
It does not compare, obtains adjacent corresponding diagnosis and treatment item of going to a doctor with reference to diagnosis and treatment item set and be overlapped ratio, and calculate
All adjacent medical average coincidence ratios, obtain the continuation degree of adjacent medical diagnosis and treatment item.It can pre-establish and be examined with various
The disconnected corresponding diagnosis and treatment normative database of result includes the corresponding diagnosis and treatment item of each diagnostic result (mark), because sick to one
Treatment, often follows certain clinic/treatment path, therefore, the corresponding diagnosis and treatment item of each diagnostic result is faced based on certain
Bed/treatment path carries out sequencing arrangement.
Adjacent corresponding diagnosis and treatment item of going to a doctor is being compared respectively with reference to diagnosis and treatment item set, is being obtained adjacent
Corresponding diagnosis and treatment item of going to a doctor with reference to diagnosis and treatment item set is overlapped ratio, and calculates all adjacent medical average weights
Composition and division in a proportion example when obtaining the continuation degree of adjacent medical diagnosis and treatment item, obtains adjacent corresponding targeted diagnostics of going to a doctor as a result, inquiry is examined
Normative database is treated, the corresponding reference diagnosis and treatment item set having of targeted diagnostics result is obtained from normative database.By phase
Neighbour go to a doctor corresponding diagnosis and treatment item with compared respectively with reference to diagnosis and treatment item set, obtain it is adjacent go to a doctor it is corresponding
Diagnosis and treatment item is overlapped ratio with reference to diagnosis and treatment item set, i.e., has much ratios in adjacent corresponding diagnosis and treatment item of going to a doctor
Diagnosis and treatment item in reference diagnosis and treatment item set, specifically: belong to accounted for reference to the diagnosis and treatment item in diagnosis and treatment item set it is adjacent
The ratio of medical diagnosis and treatment item, and all adjacent medical average coincidence ratios are calculated, it is adjacent medical that this, which is averagely overlapped ratio,
Diagnosis and treatment item continuation degree.Gone to a doctor in (n+1)th time for example, n-th is gone to a doctor to be adjacent medical, the diagnosis and treatment item of n-th be A,
B, C, (n+1)th medical diagnosis and treatment item are D, E, F, G, H, are combined into [A, B, C, D, E, F, G] with reference to diagnosis and treatment item collection, then n-th
Secondary corresponding diagnosis and treatment item of going to a doctor is 100% with the ratio that is overlapped with reference to diagnosis and treatment item set, (n+1)th corresponding diagnosis and treatment of going to a doctor
Project is 80% with the ratio that is overlapped with reference to diagnosis and treatment item set, then all adjacent medical average coincidence ratios are (100%+
80%)/2=90%.
Step S326, when the continuation degree is greater than six preset thresholds, corresponding medical data meets off-note.
6th preset threshold refers to the value by user preset, can be adjusted according to the actual situation, in the above-described example, the
Six preset thresholds can be 45%.
Optionally, may also include before step S325 in the present embodiment adjacent medical interval based on adjacent medical data and
The judgement of the similitude of diagnostic result, specifically includes:
The corresponding diagnostic result of every medical data in the second subset conjunction is extracted, judges that the second subset is in closing
The no adjacent medical data that there is corresponding similar diagnosis result；If there is corresponding similar diagnosis result in closing in the second subset
Adjacent medical data then obtains the time difference at the adjacent medical interval and the time interval threshold value of the adjacent medical data,
The ratio for calculating the time difference Yu the time interval threshold value, obtain the corresponding interval of the adjacent medical data it is abnormal because
Son；If the corresponding interval Outlier factor of the adjacent medical data is greater than the 5th preset threshold, extracts the second subset and close
In the corresponding diagnosis and treatment item of every medical data, by the adjacent corresponding diagnosis and treatment item and with reference to diagnosis and treatment item set point of going to a doctor
It does not compare, obtains adjacent corresponding diagnosis and treatment item of going to a doctor with reference to diagnosis and treatment item set and be overlapped ratio, and calculate
All adjacent medical average coincidence ratios, obtain the continuation degree of adjacent medical diagnosis and treatment item；It is greater than the in the continuation degree
When six preset thresholds, corresponding medical data meets off-note.Wherein, the adjacent medical interval of adjacent medical data and diagnosis
As a result the associated description of similitude describes in the 5th embodiment, does not repeat herein.Based on the adjacent of adjacent medical data
Medical interval, the similitude of diagnostic result and adjacent medical diagnosis and treatment item continuation degree, abnormal medical data identification can be promoted
Accuracy, reduce wrong report, promote the flexibility of abnormality detection.
The present embodiment is greater than the 6th default threshold in the continuation degree by the continuation degree of the adjacent medical diagnosis and treatment item of analysis
When value, corresponding medical data meets off-note, and the judgement to the behavior of being hospitalized is decomposed can be realized based on clinical treatment path, is had
Effect prevents insurance fraud behavior, reduces public interest loss.
In addition, the present invention also provides a kind of medical data corresponding with above-mentioned each step of medical data abnormality recognition method is different
Normal identification device.
It is the functional block diagram of medical data anomalous identification device first embodiment of the present invention referring to Fig. 5, Fig. 5.
In the present embodiment, medical data anomalous identification device of the present invention includes:
Module 10 is obtained, requests corresponding target doctor for when detecting abnormality detection request, obtaining the abnormality detection
The medical data to be detected of institute and the objective hospital；
Grouping module 20, for being grouped the medical data to be detected according to default dimension, after being grouped
Subclass；
Anomaly analysis module 30 judges institute for analyzing each subclass based on the corresponding preset rules of default dimension
It states in medical data to be detected with the presence or absence of the target medical data for meeting off-note；
Abnormal marking module 40, if for there is the target medical treatment number for meeting off-note in the medical data to be detected
According to then by target medical data progress abnormal marking.
Further, the medical data anomalous identification device includes:
The grouping module 20 is also used to when the default dimension is hospital's dimension, by the medical data to be detected
It is grouped according to disease, obtains corresponding first subclass of each disease；
First computing module, for from corresponding first subclass of each disease, obtain each disease total hospitalizations and
Total length of stay, the average single that total hospitalizations and total length of stay based on each disease calculate each disease of acquisition are hospitalized
Number of days；The average single length of stay of each disease preset number of days corresponding with each disease is compared, each disease is obtained
Average single length of stay and the number of days of corresponding preset number of days it is poor, calculate the number of days difference of each disease and corresponding preset number of days
Ratio, obtain the corresponding number of days Outlier factor of each disease；
First abnormal determination module is used for when the corresponding number of days Outlier factor of any disease is greater than the first preset threshold,
Corresponding first subclass of the disease meets off-note.
Further, the medical data anomalous identification device includes:
Second computing module obtains the insured number of each disease and total for from corresponding first subclass of each disease
Hospitalizations, insured number and total hospitalizations based on each disease, which calculate, obtains corresponding everyone average time in hospital time of each disease
Number；Everyone corresponding average time in hospital number of each disease is compared with corresponding preset times, everyone for obtaining each disease is flat
Equal hospitalizations and the number of corresponding preset times are poor, calculate the ratio of the number difference and corresponding preset times of each disease,
Obtain the corresponding frequency abnormality factor of each disease；
Second abnormal determination module is used for when the corresponding frequency abnormality factor of any disease is greater than the second preset threshold,
Corresponding first subclass of the disease meets off-note.
Further, the medical data anomalous identification device includes:
Third computing module, for from corresponding first subclass of each disease, obtain each disease total hospitalizations and
Always the amount of money, the average single that total hospitalizations and total amount of money of being hospitalized based on each disease calculate each disease of acquisition are hospitalized in hospital
The amount of money；The average single of each disease amount of money of being hospitalized is compared with corresponding preset cost, being averaged for each disease is obtained
Single be hospitalized the amount of money and the amount of money of corresponding preset cost it is poor, calculate the amount of money difference of each disease and the ratio of corresponding preset cost
Value, obtains the corresponding amount of money Outlier factor of each disease；
Third abnormal determination module is used for when the corresponding amount of money Outlier factor of any disease is greater than third predetermined threshold value,
Corresponding first subclass of the disease meets off-note.
Further, the medical data anomalous identification device includes:
The grouping module 20 is also used to when the default dimension is personal factor, will be described to be checked according to insured people
It surveys medical data to be grouped, obtains the corresponding second subset of each insured people and close；
4th computing module is calculated for extracting the corresponding consultation time of every medical data in the second subset conjunction
Obtain the adjacent medical interval of adjacent medical data in the second subset conjunction；Preset time interval threshold value is obtained, institute is calculated
The time difference for stating adjacent medical interval and the time interval threshold value, calculate the ratio of the time difference and the time interval threshold value
Value obtains the corresponding interval Outlier factor of every medical data in the second subset conjunction；
4th abnormal determination module is big for the corresponding interval Outlier factor of medical data in second subset conjunction
When four preset thresholds, which meets off-note.
Further, the medical data anomalous identification device includes:
Similar judgment module, for extracting the corresponding diagnostic result of every medical data in the second subset conjunction, judgement
The second subset whether there is the adjacent medical data for corresponding to similar diagnosis result in closing；
First obtains module, if the adjacent medical number for there is corresponding similar diagnosis result in second subset conjunction
According to adjacent medical interval and the time difference of the time interval threshold value of the adjacent medical data then being obtained, when calculating described
Between difference and the ratio of the time interval threshold value, obtain the corresponding interval Outlier factor of the adjacent medical data；
5th abnormal determination module is preset for being greater than the 5th in the corresponding interval Outlier factor of the adjacent medical data
When threshold value, which meets off-note.
Further, the medical data anomalous identification device includes:
5th computing module, for extracting the corresponding diagnosis and treatment item of every medical data in the second subset conjunction, by phase
Neighbour go to a doctor corresponding diagnosis and treatment item with compared respectively with reference to diagnosis and treatment item set, obtain it is adjacent go to a doctor it is corresponding
Diagnosis and treatment item is overlapped ratio with reference to diagnosis and treatment item set, and calculates all adjacent medical average coincidence ratios, obtains phase
The continuation degree of the medical diagnosis and treatment item of neighbour；
6th abnormal determination module, for when the continuation degree is greater than six preset thresholds, corresponding medical data to be full
Sufficient off-note.
The present invention also proposes a kind of storage medium, is stored thereon with computer program.The storage medium can be Fig. 1's
Memory 201 in medical data anomalous identification terminal, be also possible to as ROM (Read-Only Memory, read-only memory)/
At least one of RAM (Random Access Memory, random access memory), magnetic disk, CD, the storage medium packet
Some instructions are included to use so that a terminal device with processor (can be mobile phone, computer, server, the network equipment
Or medical data anomalous identification terminal in the embodiment of the present invention etc.) execute method described in each embodiment of the present invention.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the server-side that include a series of elements not only include those elements,
And including other elements that are not explicitly listed, or including for this process, method, article or server-side institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or server-side.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of medical data abnormality recognition method, which is characterized in that the medical data abnormality recognition method includes following step
It is rapid:
When detecting abnormality detection request, obtain the abnormality detection request corresponding objective hospital and the objective hospital to
Detect medical data；
The medical data to be detected is grouped according to default dimension, the subclass after being grouped；
Each subclass is analyzed based on the corresponding preset rules of default dimension, judges whether deposit in the medical data to be detected
In the target medical data for meeting off-note；
If there is the target medical data for meeting off-note in the medical data to be detected, by the target medical data
Carry out abnormal marking.
2. medical data abnormality recognition method as described in claim 1, which is characterized in that the default dimension be hospital dimension
When spending, the step of basis presets dimension and is grouped the medical data to be detected, subclass after being grouped, is wrapped
It includes:
The medical data to be detected is grouped according to disease, obtains corresponding first subclass of each disease；
It is described to analyze each subclass based on the corresponding preset rules of default dimension, judge be in the medical data to be detected
It is no to include: in the presence of the step of target medical data for meeting off-note
From corresponding first subclass of each disease, obtain each disease total hospitalizations and total length of stay, based on described each
Total hospitalizations of disease and total length of stay calculate the average single length of stay for obtaining each disease；
The average single length of stay of each disease preset number of days corresponding with each disease is compared, each disease is obtained
Average single length of stay and the number of days of corresponding preset number of days are poor, calculate the number of days difference of each disease and corresponding preset number of days
Ratio obtains the corresponding number of days Outlier factor of each disease；
When the corresponding number of days Outlier factor of any disease is greater than the first preset threshold, corresponding first subclass of the disease meets
Off-note.
3. medical data abnormality recognition method as claimed in claim 2, which is characterized in that described corresponding based on default dimension
Preset rules analyze each subclass, judge in the medical data to be detected with the presence or absence of the target doctor for meeting off-note
Treat data the step of include:
From corresponding first subclass of each disease, obtain each disease insured number and total hospitalizations, based on each disease
Insured number and total hospitalizations, which calculate, obtains everyone corresponding average time in hospital number of each disease；
Everyone corresponding average time in hospital number of each disease is compared with corresponding preset times, everyone for obtaining each disease is flat
Equal hospitalizations and the number of corresponding preset times are poor, calculate the ratio of the number difference and corresponding preset times of each disease,
Obtain the corresponding frequency abnormality factor of each disease；
When the corresponding frequency abnormality factor of any disease is greater than the second preset threshold, corresponding first subclass of the disease meets
Off-note.
4. medical data abnormality recognition method as claimed in claim 2, which is characterized in that described corresponding based on default dimension
Preset rules analyze each subclass, judge in the medical data to be detected with the presence or absence of the target doctor for meeting off-note
Treat data the step of include:
From corresponding first subclass of each disease, total hospitalizations of each disease and the amount of money of being always hospitalized are obtained, based on described each
Total hospitalizations of disease and total amount of money of being hospitalized, which calculate, to be obtained the average single of each disease and is hospitalized the amount of money；
The average single of each disease amount of money of being hospitalized is compared with corresponding preset cost, the average list of each disease is obtained
The secondary amount of money and the amount of money of corresponding preset cost in hospital are poor, calculate the ratio of the amount of money difference and corresponding preset cost of each disease,
Obtain the corresponding amount of money Outlier factor of each disease；
When the corresponding amount of money Outlier factor of any disease is greater than third predetermined threshold value, corresponding first subclass of the disease meets
Off-note.
5. medical data abnormality recognition method as described in claim 1, which is characterized in that tieed up in the default dimension to be personal
When spending, the step of basis presets dimension and is grouped the medical data to be detected, subclass after being grouped, is wrapped
It includes:
The medical data to be detected is grouped according to insured people, the corresponding second subset of each insured people is obtained and closes；
It is described to analyze each subclass based on the corresponding preset rules of default dimension, judge be in the medical data to be detected
It is no to include: in the presence of the step of target medical data for meeting off-note
The corresponding consultation time of every medical data in the second subset conjunction is extracted, calculates and obtains phase in the second subset conjunction
The adjacent medical interval of adjacent medical data；
Preset time interval threshold value is obtained, the time difference at the adjacent medical interval and the time interval threshold value, meter are calculated
The ratio for calculating the time difference Yu the time interval threshold value obtains in the second subset conjunction every medical data corresponding
Every Outlier factor；
When the corresponding interval Outlier factor of medical data in second subset conjunction is greater than four preset thresholds, the medical treatment number
According to meeting off-note.
6. medical data abnormality recognition method as claimed in claim 5, which is characterized in that described to obtain preset time interval
Threshold value calculates the time difference at the adjacent medical interval and the time interval threshold value, calculates the time difference and the time
The ratio of interval threshold, obtain the second subset close in every medical data corresponding interval Outlier factor the step of after packet
It includes:
The corresponding diagnostic result of every medical data in the second subset conjunction is extracted, judges whether the second subset deposits in closing
In the adjacent medical data of corresponding similar diagnosis result；
If the second subset has the adjacent medical data for corresponding to similar diagnosis result in closing, the adjacent medical number is obtained
According to adjacent medical interval and time difference of the time interval threshold value, calculate the time difference and the time interval threshold value
Ratio obtains the corresponding interval Outlier factor of the adjacent medical data；
When the corresponding interval Outlier factor of the adjacent medical data is greater than five preset thresholds, which meets abnormal
Feature.
7. medical data abnormality recognition method as claimed in claim 5, which is characterized in that described corresponding based on default dimension
Preset rules analyze each subclass, judge in the medical data to be detected with the presence or absence of the target doctor for meeting off-note
Treat data the step of include:
The corresponding diagnosis and treatment item of every medical data in the second subset conjunction is extracted, by adjacent corresponding diagnosis and treatment item of going to a doctor
Mesh with compared respectively with reference to diagnosis and treatment item set, obtain it is adjacent go to a doctor corresponding diagnosis and treatment item with reference to diagnosis and treatment item
The coincidence ratio of set, and all adjacent medical average coincidence ratios are calculated, obtain the continuous of adjacent medical diagnosis and treatment item
Degree；
When the continuation degree is greater than six preset thresholds, corresponding medical data meets off-note.
8. a kind of medical data anomalous identification device, which is characterized in that the medical data anomalous identification device includes:
Module is obtained, for the abnormality detection being obtained and requesting corresponding objective hospital and institute when detecting abnormality detection request
State the medical data to be detected of objective hospital；
Grouping module, for the medical data to be detected to be grouped according to default dimension, the subclass after being grouped；
Anomaly analysis module judges described to be checked for analyzing each subclass based on the corresponding preset rules of default dimension
It surveys in medical data with the presence or absence of the target medical data for meeting off-note；
Abnormal marking module, if for there is the target medical data for meeting off-note in the medical data to be detected,
The target medical data is subjected to abnormal marking.
9. a kind of medical data anomalous identification terminal, which is characterized in that the medical data anomalous identification terminal include processor,
Memory and be stored on the memory and can by the medical data anomalous identification program that the processor executes, wherein
When the medical data anomalous identification program is executed by the processor, the doctor as described in any one of claims 1 to 7 is realized
The step for the treatment of data exception recognition methods.
10. a kind of storage medium, which is characterized in that medical data anomalous identification program is stored on the storage medium, wherein
When the medical data anomalous identification program is executed by processor, the medical number as described in any one of claims 1 to 7 is realized
The step of according to abnormality recognition method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811226809.0A CN109636613A (en) | 2018-10-19 | 2018-10-19 | Medical data abnormality recognition method, device, terminal and storage medium |
| PCT/CN2019/096631 WO2020078058A1 (en) | 2018-10-19 | 2019-07-19 | Medical data abnormality identification method and device, terminal, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811226809.0A CN109636613A (en) | 2018-10-19 | 2018-10-19 | Medical data abnormality recognition method, device, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109636613A true CN109636613A (en) | 2019-04-16 |
Family
ID=66066479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811226809.0A Pending CN109636613A (en) | 2018-10-19 | 2018-10-19 | Medical data abnormality recognition method, device, terminal and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109636613A (en) |
| WO (1) | WO2020078058A1 (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020078058A1 (en) * | 2018-10-19 | 2020-04-23 | 平安医疗健康管理股份有限公司 | Medical data abnormality identification method and device, terminal, and storage medium |
| CN111242793A (en) * | 2020-01-16 | 2020-06-05 | 上海金仕达卫宁软件科技有限公司 | Method and device for detecting medical insurance data abnormity |
| CN111340641A (en) * | 2020-05-22 | 2020-06-26 | 浙江工业大学 | Abnormal hospitalizing behavior detection method |
| CN111445343A (en) * | 2020-03-23 | 2020-07-24 | 平安医疗健康管理股份有限公司 | Risk medical data identification method and device, computer equipment and storage medium |
| CN112305290A (en) * | 2019-07-25 | 2021-02-02 | 比亚迪股份有限公司 | Current detection device |
| CN112447286A (en) * | 2019-09-05 | 2021-03-05 | 天津新开心生活科技有限公司 | Method and device for automatically distinguishing treatment line |
| CN112445846A (en) * | 2020-12-01 | 2021-03-05 | 平安医疗健康管理股份有限公司 | Medical item identification method, device, equipment and computer readable storage medium |
| CN113130053A (en) * | 2021-03-24 | 2021-07-16 | 大光惠世医疗科技有限公司 | Self-service registration method and device based on user activity data |
| CN113822365A (en) * | 2021-09-28 | 2021-12-21 | 刘玉棚 | Medical data storage and big data mining method and system based on block chain technology |
| CN113838579A (en) * | 2021-09-29 | 2021-12-24 | 平安医疗健康管理股份有限公司 | Medical data anomaly detection method, device, equipment and storage medium |
| CN114912804A (en) * | 2022-05-17 | 2022-08-16 | 四川大学华西医院 | Scientific research data related property control method and system |
| CN115267094A (en) * | 2022-09-29 | 2022-11-01 | 南通迈伦机电科技有限公司 | Exhaust emission monitoring and detecting method |
| CN112445846B (en) * | 2020-12-01 | 2024-07-05 | 深圳平安医疗健康科技服务有限公司 | Medical item identification method, device, equipment and computer readable storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5508912A (en) * | 1989-01-23 | 1996-04-16 | Barry Schneiderman | Clinical database of classified out-patients for tracking primary care outcome |
| JP2006018532A (en) * | 2004-06-30 | 2006-01-19 | Toshiba Corp | System and method for analyzing hospital management information, computer readable storage medium and computer program |
| CN101986333A (en) * | 2010-12-01 | 2011-03-16 | 福州维胜信息技术有限公司 | Auxiliary decision supporting system of hospital |
| CN106021967A (en) * | 2016-07-12 | 2016-10-12 | 上海市浦东新区周浦医院 | Data processing method for clinical pathway quality evaluation |
| CN106408141A (en) * | 2015-07-28 | 2017-02-15 | 平安科技（深圳）有限公司 | Abnormal expense automatic extraction system and method |
| CN107609980A (en) * | 2017-09-07 | 2018-01-19 | 平安医疗健康管理股份有限公司 | Medical data processing method, device, computer equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514576A (en) * | 2013-09-06 | 2014-01-15 | 深圳民太安信息技术有限公司 | Screening method for illegal cashing of social security treatment |
| CN107194184B (en) * | 2017-05-31 | 2020-11-17 | 成都数联易康科技有限公司 | Method and system for detecting abnormality of people in hospital based on time sequence similarity analysis |
| CN109636613A (en) * | 2018-10-19 | 2019-04-16 | 平安医疗健康管理股份有限公司 | Medical data abnormality recognition method, device, terminal and storage medium |
-
2018
- 2018-10-19 CN CN201811226809.0A patent/CN109636613A/en active Pending
-
2019
- 2019-07-19 WO PCT/CN2019/096631 patent/WO2020078058A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5508912A (en) * | 1989-01-23 | 1996-04-16 | Barry Schneiderman | Clinical database of classified out-patients for tracking primary care outcome |
| JP2006018532A (en) * | 2004-06-30 | 2006-01-19 | Toshiba Corp | System and method for analyzing hospital management information, computer readable storage medium and computer program |
| CN101986333A (en) * | 2010-12-01 | 2011-03-16 | 福州维胜信息技术有限公司 | Auxiliary decision supporting system of hospital |
| CN106408141A (en) * | 2015-07-28 | 2017-02-15 | 平安科技（深圳）有限公司 | Abnormal expense automatic extraction system and method |
| CN106021967A (en) * | 2016-07-12 | 2016-10-12 | 上海市浦东新区周浦医院 | Data processing method for clinical pathway quality evaluation |
| CN107609980A (en) * | 2017-09-07 | 2018-01-19 | 平安医疗健康管理股份有限公司 | Medical data processing method, device, computer equipment and storage medium |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020078058A1 (en) * | 2018-10-19 | 2020-04-23 | 平安医疗健康管理股份有限公司 | Medical data abnormality identification method and device, terminal, and storage medium |
| CN112305290A (en) * | 2019-07-25 | 2021-02-02 | 比亚迪股份有限公司 | Current detection device |
| CN112305290B (en) * | 2019-07-25 | 2023-11-24 | 比亚迪半导体股份有限公司 | Current detection device |
| CN112447286B (en) * | 2019-09-05 | 2023-02-03 | 天津新开心生活科技有限公司 | Method and device for automatically distinguishing treatment line |
| CN112447286A (en) * | 2019-09-05 | 2021-03-05 | 天津新开心生活科技有限公司 | Method and device for automatically distinguishing treatment line |
| CN111242793A (en) * | 2020-01-16 | 2020-06-05 | 上海金仕达卫宁软件科技有限公司 | Method and device for detecting medical insurance data abnormity |
| CN111242793B (en) * | 2020-01-16 | 2024-02-06 | 上海金仕达卫宁软件科技有限公司 | Medical insurance data abnormality detection method and device |
| CN111445343A (en) * | 2020-03-23 | 2020-07-24 | 平安医疗健康管理股份有限公司 | Risk medical data identification method and device, computer equipment and storage medium |
| CN111340641A (en) * | 2020-05-22 | 2020-06-26 | 浙江工业大学 | Abnormal hospitalizing behavior detection method |
| CN111340641B (en) * | 2020-05-22 | 2020-11-13 | 浙江工业大学 | Abnormal hospitalizing behavior detection method |
| CN112445846A (en) * | 2020-12-01 | 2021-03-05 | 平安医疗健康管理股份有限公司 | Medical item identification method, device, equipment and computer readable storage medium |
| CN112445846B (en) * | 2020-12-01 | 2024-07-05 | 深圳平安医疗健康科技服务有限公司 | Medical item identification method, device, equipment and computer readable storage medium |
| CN113130053A (en) * | 2021-03-24 | 2021-07-16 | 大光惠世医疗科技有限公司 | Self-service registration method and device based on user activity data |
| CN113822365B (en) * | 2021-09-28 | 2023-09-05 | 北京恒生芸泰网络科技有限公司 | Medical data storage and big data mining method and system based on block chain technology |
| CN113822365A (en) * | 2021-09-28 | 2021-12-21 | 刘玉棚 | Medical data storage and big data mining method and system based on block chain technology |
| CN113838579A (en) * | 2021-09-29 | 2021-12-24 | 平安医疗健康管理股份有限公司 | Medical data anomaly detection method, device, equipment and storage medium |
| CN114912804A (en) * | 2022-05-17 | 2022-08-16 | 四川大学华西医院 | Scientific research data related property control method and system |
| CN115267094A (en) * | 2022-09-29 | 2022-11-01 | 南通迈伦机电科技有限公司 | Exhaust emission monitoring and detecting method |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020078058A1 (en) | 2020-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109636613A (en) | Medical data abnormality recognition method, device, terminal and storage medium | |
| CN109636623A (en) | Medical data method for detecting abnormality, device, equipment and storage medium | |
| Wen et al. | MVS-GCN: A prior brain structure learning-guided multi-view graph convolution network for autism spectrum disorder diagnosis | |
| Saini et al. | Efficiency in functional analysis of problem behavior: A quantitative and qualitative review | |
| Mathes et al. | Pay for performance for hospitals | |
| Elbers et al. | Interventions for fatigue in Parkinson's disease | |
| Nikolakopoulou et al. | Demystifying fixed and random effects meta-analysis | |
| US10140422B2 (en) | Progression analytics system | |
| Archer et al. | Collaborative care for depression and anxiety problems | |
| Hu et al. | Automated detection of postoperative surgical site infections using supervised methods with electronic health record data | |
| CN107785057B (en) | Medical data processing method, device, storage medium and computer equipment | |
| CN109545317A (en) | The method and Related product of behavior in hospital are determined based on prediction model in hospital | |
| Mills et al. | An analysis of general medical and specialist journals that endorse CONSORT found that reporting was not enforced consistently | |
| CN106599553A (en) | Disease early-warning method and device | |
| CN113657548A (en) | Medical insurance abnormity detection method and device, computer equipment and storage medium | |
| CN109523399A (en) | A kind of medical data processing method, device, equipment and storage medium | |
| US20070150314A1 (en) | Method for carrying out quality control of medical data records collected from different but comparable patient collectives within the bounds of a medical plan | |
| CN110729054B (en) | Abnormal diagnosis behavior detection method and device, computer equipment and storage medium | |
| Garnaat et al. | Factor structure and measurement invariance of the Yale-Brown Obsessive Compulsive Scale across four racial/ethnic groups | |
| CN109934723B (en) | Medical insurance fraud behavior identification method, device and equipment | |
| CN109119137A (en) | A kind of method for detecting abnormality, device, server and storage medium | |
| CN109767067A (en) | Method and Related product based on more evaluative dimensions evaluation hospital | |
| CN109544364A (en) | Unlawful practice detection method, device and terminal based on data analysis | |
| US20220005565A1 (en) | System with retroactive discrepancy flagging and methods for use therewith | |
| CN109377207A (en) | The abnormal method and Related product that behavior determines of being hospitalized |
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: 20190416 |