CN104951894B - Hospital's disease control intellectual analysis and assessment system - Google Patents
Hospital's disease control intellectual analysis and assessment system Download PDFInfo
- Publication number
- CN104951894B CN104951894B CN201510359039.7A CN201510359039A CN104951894B CN 104951894 B CN104951894 B CN 104951894B CN 201510359039 A CN201510359039 A CN 201510359039A CN 104951894 B CN104951894 B CN 104951894B
- Authority
- CN
- China
- Prior art keywords
- hospital
- disease
- data
- patient
- model
- 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.)
- Active
Links
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a kind of hospital's disease control intellectual analysis and assessment systems, user inquires terminal for the assessment result according to quality of hospital management evaluation module, comparative evaluation between realization quality of medical care, the Yuan Nei hospitals lean management of four dimension of medical efficiency, Medical Benefit and patient satisfaction and hospital.The present invention does not realize a variety of screenings between different dimensions not only, independent assortment and 360 degrees omnidirection dynamic contrast, also achieve a variety of screenings between dimension of the same race, combination and comprehensive comparison, by in hospital's big data specialty analysis, Decision-making of Hospital Management is supported and hospital management different dimensions compare, the problem of clear and definite hospital management and ranking, positioning, realization includes ICD diagnosis, medical diagnosis on disease associated packets DRG diseases, clinician, clinical department, the global alignment and ranking waited between ground district hospital, clearly compare strengths and weaknesses of the object between industry, it can clearly position, it pushes hospital's lean management and promotes comprehensive hospital competitiveness.
Description
Technical field
The present invention relates to lean hospital quality management field, medical big data analysis and Decision-making of Hospital Management to support field,
More particularly to a kind of hospital's disease control intellectual analysis and assessment system.
Background technology
Compared with external advanced hospital management, domestic hospital management not only lacks substantially or in the starting stage
High-quality managerial talent more lacks scientific quality and efficiency evaluation standard, leads to the planless enlarging of hospital and doctor
Treat the huge waste of resource.In recent years due to the fast development of domestic hospitals IT technologies, patient and disease are preliminarily completed
The initial data accumulation of disease, however suffers from no methodology, it is impossible to these data are effectively refined as tutorial message and
The decision-making foundation of hospital management causes most data that can only be stored in the data warehouse of hospital, wastes resource.If
Successful pattern and outstanding methodology of the U.S. government to hospital management can be fully used for reference, then is subject to localization improvement, no
But domestic medical management agency can be allowed to increase effective monitoring approach and means, and hospital can also be promoted to accelerate from extensive
The paces to make the transition to fine management model.
Recent government actively advocates and encourages transition of the traditional industries to " internet+", and digital technology is made full use of to improve
The quality of medical care of hospital, operation efficiency and reduction medical resource waste have become the tendency of the day, seize the opportunity, use for reference advanced warp
It tests, establishes mode standard, it will occupy first-strike advantage, lead the reform tide of industry.
Clinical medical multidisciplinary and disease complexity increases data depth analysis and purification is management decision-making support
The difficulty of foundation, compared with the data of other industry, medical data have nonadditivity (such as financial data) and it is indirect can
Than property (such as size of data) feature, due to each hospital admissions patient crowd and disease degree difference, by directly using death
Performance comparative assessment of the data such as rate, length of stay and cost between disease, doctor, section office and hospital is unreasonable.Example
As said, due to receiving largely to transfer from one hospital to another patient and accepting the more serious patient crowd of the state of an illness for medical treatment, Sichuan West China Hospital cannot be direct
Simple performance evaluation is carried out with some County Hospital.
Have to effectively solve one of the predicament of clinical data injustice, the evaluation profile of hospital's generally use and used with resource
For the disease group inductive method of standard, such as all kinds of DRG and DCG, the medical treatment cost that then will be used in treatment, through excessive
Analysis obtains the case complexity index method (CMI) of disease group.Hospital Disease group is retrodicted out by medical resource cost service condition
The severity extent of group, so as to fulfill the assessment of hospital and section office in same system.However it is being evaluated in the CMI methods calculated
Quality of medical care, operation efficiency and Rational drugs use etc. have its congenital deficiency, and pattern this first does not consider disease
The sick characteristic of itself and other clinical correlation influence factors, do not meet medical rule；Secondly excessive imaging causes with treating
Virtual height cost treatment can also increase model unstability in itself, so as to cause the deviation of judging result.
Existing hospital management system can not be realized to be supported by professional medical big data analysis auxiliary Decision-making of Hospital Management,
It can not realize in lean hospital management between different dimensions or medical control data comparison between dimension of the same race, hospital's pipe can not be specified
Ranking and positioning in reason, including the comparison object present position between disease, between doctor, between section office, between hospital and ranking etc., nothing
Method compares strengths and weaknesses of the object between industry, is unfavorable for giving priority to Priority Department, makes up short slab subject.In addition, for hospital
Manage regulatory agency, can not in region self-defined hospital, clinical speciality, discharge section office, disease DRG, discharge time section, master
Secondary diagnosis or operation, patient age, gender, classification and aggregation patient satisfaction etc. carry out dynamic queries, can not be by across doctor
The compound comparison query of the different condition combination of institute, hospital management supervision are difficult to science, effectively carry out.
Invention content
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of novel hospital's disease control intellectual analysis
And assessment system, full disease hospital quality management assessment is realized based on the adjustment of inpatient's disease risks, model passes through certain
A hospital or the historical data of a certain all In-patients in area, by patient admission when adjoint complication/concurrent
Disease, individual patient speciality (such as gender, age, survival condition) and state source etc. of being admitted to hospital are integrated into the shadow of disease treatment
Ring Variable Factors, by disease associated group (DRG) classification and the final treatment information of these patients, establish respectively the death rate,
The correlation statistically regression model of length of stay and in hospital cost.Then the algorithm obtained again by these models shows hospital
There is patient precisely to be predicted, calculate each patient in the death rate, length of stay and the desired value of cost in hospital.
The purpose of the present invention is achieved through the following technical solutions：Hospital's disease control intellectual analysis and assessment are
System inquires terminal including patient terminal, evaluating server and user, and patient terminal and user inquire terminal and pass through communication network respectively
Network is connected with evaluating server；
Collection disease follow-up APP is provided on the patient terminal, the satisfaction information of patient after leaving hospital for dynamic acquisition；
The evaluating server is looked into including clinical data introducting interface module, quality of hospital management evaluation module and user
Ask interface module：
Clinical data introducting interface module imports clinical data for hospital management for connecting hospital clinical data center
Quality assessment modules carry out the assessment of medical control quality；
Quality of hospital management evaluation module be used for according to hospital clinical data realize to each inpatient the death rate, live
The risk profile of being admitted to hospital of institute's number of days and medical treatment cost, by being found out in the historical data of each disease management of hospital to influencing most
The universal law of whole treatment results and can quantization factor, thus it is speculated that go out the dead of the current patient for having similar disease degree and a similar features
Die the prediction occurrence value of rate, length of stay and medical treatment cost；
User's query interface module is used to inquire terminal offer user interface to user；
The user inquires terminal for the assessment result according to quality of hospital management evaluation module, realizes different dimensions
Between a variety of screenings, independent assortment and comprehensive comparison, while realize a variety of screenings between dimension of the same race, combination and comprehensive right
Than；User inquires terminal can also inquire the satisfaction assessment that discharged patient feeds back by evaluating server, and realize same
The analysis and comparison of patient satisfaction in standard.
The quality of hospital management evaluation module includes a historical data screening and modeling unit, a current number
According to screening and pre-value computing unit：
Historical data is screened includes historical data import modul, data scrubbing module, medical diagnosis on disease phase with modeling unit
Close grouping DRG and model classifications module, conjunction complication and its dependent variable sorts out collection modules, statistics of variable is examined and screening mould
Block, statistical models establish module, model quality authentication module.Historical data import modul is used to import from hospital database
Historic discharged patient's data；Data scrubbing module filters out bad data and extreme value data for completing data discriminating and cleaning
And it is deleted；Medical diagnosis on disease associated packets DRG and model classifications module are used to complete medical diagnosis on disease associated packets DRG and model
Classification, realize the classification set to medical diagnosis on disease associated packets, category of model, number sort out set；Close complication and other
Variable sort out collection modules for complete be admitted to hospital when diseases and related health problems international statistical classification ICD close complication and its
The classification set of dependent variable realizes the classification to inpatient's complication and complication and its dependent variable；The statistical check of variable
The screening to the variable with statistically significant meaning is used to implement with screening module；Statistical models establish module for complete system
Meter learns the foundation of model, and patient death rate data use Logic Regression Models, and the data of length of stay and cost are then using polynary
Linear regression model (LRM) forms the quantitative formula of predicted value by modeling；Model quality authentication module is used for using in statistics
C-Index is examined and the R-square methods of inspection calculate model in sample population and non-sample crowd, according to corresponding
As a result it is evaluated.
Current data is screened includes current data import modul, data scrubbing module, medical diagnosis on disease with pre-value computing unit
Associated packets DRG and model classifications module close complication and its dependent variable classification collection modules, risk profile value of being admitted to hospital calculating mould
Block.Current data import modul is used to import current discharged patient's data from hospital database；Data scrubbing module has been used for
Differentiate into data and clear up, filter out bad data and deleted；Medical diagnosis on disease associated packets DRG and model classifications module are used for
The classification of medical diagnosis on disease associated packets DRG and model is completed, realizes the classification set to medical diagnosis on disease associated packets, model point
Class, number sort out set；It closes complication and its dependent variable is sorted out disease and related health when collection modules are admitted to hospital for completion and asked
The international statistical classification ICD of topic closes complication and its classification set of dependent variable, realizes to inpatient's complication and complication
And its classification of dependent variable；Risk profile value computing module of being admitted to hospital is used to implement to each inpatient in the death rate, day of being hospitalized
The risk profile of being admitted to hospital of number and medical treatment cost；The risk profile of disease refers to by the historical data of each disease management of hospital
Find out universal law to influencing final treatment results and can quantization factor, thus it is speculated that going out currently has similar disease degree and similar spy
The prediction occurrence value of the death rate of the patient of sign, length of stay and medical treatment cost；
The algorithmic formula of predicted value is as follows：
Expected mortalityWherein, biRepresent significantly correlated property coefficient, b0Represent model intercept, n tables
Show the significant correlation variable number of patient；
Length of stay and medical treatment costWherein, b0Represent model intercept, MSE
Represent the square error of model, biRepresent significantly correlated property coefficient, 0.5 is statistic bias correction value.
Quality of hospital management inquiry dimension includes depth, time, type, patient's information and hospital management：Depth includes row
Name, 10 percentiles, 25 percentiles, median；Time includes past and present, annual, season, monthly；Type includes disease
Disease and international statistical classification ICD, DRG in relation to health problem, clinical sub- subject, clinical speciality, hospital, area, the country, state
Border；Patient's information includes population information, way of paying, admission information, discharge information；Hospital management includes quality of medical care, medical treatment
Efficiency, Medical Benefit, patient satisfaction.
The user that the user inquires terminal includes clinical users, hospital management regulatory agency user：
For clinical users：The clinical medicine that clinical users formulate hospital, DMIAES by hospital's classification Risk Adjusted model
Section, discharge section office, all kinds of doctors, disease DRG, discharge time section, primary and secondary will diagnose or perform the operation, patient enters discharge situation, patient
Age, query result and the regional anonymous comparison of other hospitals progress of gender, patient satisfaction of all categories, realize various differences
It the compound query of conditional combination and actually occurs value, predicted value and O/E indexes and (actually occurs value/desired value, O/E indexes<1：
Illustrate disease risks height, but that treats lapses to, if case fatality rate, length of stay or cost control are less than expection；O/E indexes>1：It says
Bright disease risks are low, but that treats lapses to, if case fatality rate, length of stay or cost control are higher than expection) result displaying；
For hospital management regulatory agency user：The inquiry of hospital management regulatory agency user is superior to clinical users,
Hospital real name can be used to realize to compare, user according to hospital's classification Risk Adjusted model in region self-defined hospital,
Clinical speciality that DMIAES is formulated, discharge section office, disease DRG, discharge time section, primary and secondary will diagnose or perform the operation, patient age, property
Not, classification and aggregation patient satisfaction are inquired, and pass through the compound comparison query combined across the different condition of hospital, inquiry knot
Fruit is arranged according to user-defined sequence.
Hospital's disease control intellectual analysis and appraisal procedure：System is using big data analysis and advanced modeling method to doctor
The medical data of institute realizes the Risk Adjusted and information integration of disease, and the pattern of decision support foundation is converted into data, is disease
Sick Case management, Decision-making of Hospital Management provide effective analysis and assessment approach.
Advanced medical control decision model：It is fully taken into account again based on the external advanced Decision Support Platform of medical treatment various
Under variable, such as classification of diseases model, disease risks index, complication and crisis sign, pass through data analysis and mathematical model
As a result to the effective supplementary means of administrators of the hospital's offer, screening in time is predicted and handles what is occurred in possible therapeutic process
Fortuitous event ensures quality of medical care.The management decision-making support platform that foreign hospital advanced management experience is built is then by setting
A series of core index system (KPIs) is counted to models such as quality of medical care, flow path efficiency, patient safety, patient satisfaction analyses
It helps each operation system of solution hospital of hospital structural and couples sex chromosome mosaicism, improve quality of medical care and efficiency of operation, increase patient
Satisfaction reduces the wasting of resources.In addition for the requirement of quality of medical care, efficiency and patient safety, the assessment mark of whole world hospital
Standard is the same；But the experience of core index (KPIs) the Ze Hui foreigns hospital for hospital financial and performance management, is adopted
With the feasibility model of the administrative standard of suitable domestic hospitals, the localization of product is completed.
The data integrated solution of the full course of disease：Using cell phone end data drainage pattern, doctor is applied by App
The feedback informations such as symptomatic reaction, sign, satisfaction of patient, the data for integrating the excavation in institute have been formed after dynamic acquisition discharge
Clinical research, clinical test and the chronic disease management water of objective hospital is greatly improved in whole full course of disease treatment and rehabilitation data-link
Flat, this integrated data mining, acquisition and integration mode are in the leading level in the world.
The beneficial effects of the invention are as follows：
(1) Introduced From Abroad advanced management experience and localization are integrated, and are reformed and improved, form profession, science,
Practical management assessment method, the analysis and assessment system established based on hospital's this medical treatment of bulk sample big data analysis can be abundant
Consider the influence degree of disease risks etc., realize to quality of medical care, hospital efficiency, medical treatment cost control and patient satisfaction
Comprehensive assessment, can promote to Comprehensive medical quality in hospital management level, and the medical lean of promotion develops and is comprehensive hospital
Assessment, science positioning, Priority Department is established, performance evaluation and comprehensive hospital comprehensive management provide scientific basis and reference.
Hospital's disease control intellectual analysis and assessment system use for reference international advanced medical control experience, with informationization as holding
It carries, science, accurate, comprehensive assessment medical quality managent, efficiency of operation, cost control and patient satisfaction.As quality of medical care
The quantification tool of lean management will push medical control to develop to lean management direction, lead hospital's lean management new direction.
DMIAES efficiently solves quality of medical care in hospital management and, not than problem, doctor is provided for administration office of the hospital
Quality management evaluation criteria and decision-making foundation are treated, standard and judgment is provided for hospital's scientific management, performance evaluation.
Cost and cost model in hospital's disease control intellectual analysis and assessment system, realized full disease disease cost accounting and
Fee calculating, and fully consider the risk factors of various disease, scientific forecasting cost of illness and expense effectively control medical resource
Waste, the popularization and application to be grouped DRG diseases medical treatment payment by disease provide reference frame.
(2) full disease control assessment is realized based on the adjustment of inpatient's disease risks, model passes through some hospital
Or the historical data of a certain all In-patients in area, by patient admission when adjoint complication/complication, patient
Personal attributes (such as gender, age, survival condition) and state source etc. of being admitted to hospital be integrated into the variation of disease treatment because
Element by disease associated group (DRG) classification and the final treatment information of these patients, establishes the death rate, length of stay respectively
The correlation statistically regression model of cost in hospital.Then again by the algorithm that these models obtain to the existing patient of hospital into
Row precisely prediction, calculate each patient the death rate, length of stay and in hospital the desired value of cost with inpatient's disease wind
Realize the assessment of full disease control based on the adjustment of danger, model passes through some hospital or a certain all In-patients in area
Historical data, by patient admission when adjoint complication/complication, individual patient speciality (such as gender, age, survival condition
Deng) and state source etc. of being admitted to hospital be integrated into the variation factor of disease treatment, by disease associated group (DRG) classification
The treatment information final with these patients establishes the death rate, length of stay and the correlation statistically of cost returns in hospital respectively
Model.Then the existing patient of hospital is precisely predicted by the algorithm that these models obtain again, each patient can be calculated
In the death rate, length of stay and the desired value of cost in hospital.The methods of by big data analysis, mathematical statistics and machine learning
Effective conversion of the medical data from data to solution is realized, realizes data value.
(3) concept of the conjunction complication variable of disease is added, according to the natural law of clinical treatment, patient's the past state of an illness
Very big influence is centainly had to the treatment results implemented, the disease risks adjustment symbol based on closing complication
Close clinical medicine rule.
(4) the identical illness speciality of same ethnic population and medical treatment normal process the features such as can meet statistics
The requirement of upper homoplasy sample, usage history data carry out modeling and forecasting and also comply with mathematical law.Disease based on modeling
Sick Risk Adjusted successfully solves the bottleneck of medical data incommensurability, is done step-by-step in the Deng developed countries of USA and Europe
It is widely popularized and applies.
(5) it using O/E index manner of comparison, solves incomparable problem between medical data, can not only realize disease
Quality of medical care assessment between kind, can also be achieved between doctor, between hospital department, between hospital in inpatient's disease treatment
The performance Rationality Assessment of management is health authorities to one of effective regulatory measure of subordinate hospital.
(6) in addition to statistical test, the application model and United States Hospital alliance hospital medical quality managent are assessed
The Risk Adjusted model of system is compared, while is compared with actually occurring value progress O/E indexes, analyzes the conclusion compared
Be the application model is above same class model in the precision and grade of fit in the sample of prediction.
(7) a variety of screenings, independent assortment and the 360 degrees omnidirection comparison between different dimensions are not realized not only, are realized yet
A variety of screenings, combination and comprehensive comparison between dimension of the same race, by being compared in data analysis and different O＆Ms, clear and definite
Ranking and positioning in hospital management, including the comparison object present position between disease, between doctor, between section office, between hospital and row
Name, you can be perfectly clear the strengths and weaknesses for comparing object between industry, maximizes favourable factors and minimizes unfavourable ones, gives priority to Priority Department, makes up short slab
Section, the quick synthesized competitiveness for improving hospital.
As IT system tool, it can be achieved that faster, comprehensive comprehensive, various dimensions and the medical quality in hospital pair of various visual angles
Than assessment.
Description of the drawings
Fig. 1 is present system structural schematic block diagram；
Fig. 2 is DMIAES system 360 degrees omnidirection contrast schematic diagrams；
Fig. 3 adjusts model flow figure for disease risks of the present invention.
Specific embodiment
Technical scheme of the present invention is described in further detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to
It is as described below.
As described in Figure 1, hospital's disease control intellectual analysis and assessment system (DMIAES systems), including patient terminal, are commented
Estimate server and user inquires terminal, patient terminal and user inquire terminal and pass through communication network and evaluating server phase respectively
Even；
Collection disease follow-up APP is provided on the patient terminal, the satisfaction information of patient after leaving hospital for dynamic acquisition；
The evaluating server is looked into including clinical data introducting interface module, quality of hospital management evaluation module and user
Ask interface module：
Clinical data introducting interface module imports clinical data for hospital management for connecting hospital clinical data center
Quality assessment modules carry out the assessment of medical control quality；
Quality of hospital management evaluation module be used for according to hospital clinical data realize to each inpatient the death rate, live
The risk profile of being admitted to hospital of institute's number of days and medical treatment cost, by being found out in the historical data of each disease management of hospital to influencing most
The universal law of whole treatment results and can quantization factor, thus it is speculated that go out the dead of the current patient for having similar disease degree and a similar features
Die the prediction occurrence value of rate, length of stay and medical treatment cost；
User's query interface module is used to inquire terminal offer user interface to user；
The user inquires terminal for the assessment result according to quality of hospital management evaluation module, realizes different dimensions
Between a variety of screenings, independent assortment and comprehensive comparison, while realize a variety of screenings between dimension of the same race, combination and comprehensive right
Than；User inquires terminal can also inquire the satisfaction assessment that discharged patient feeds back by evaluating server, and realize same
The analysis and comparison of patient satisfaction in standard.
The quality of hospital management evaluation module includes a historical data screening and modeling unit, a current number
According to screening and pre-value computing unit：
Historical data is screened includes historical data import modul, data scrubbing module, medical diagnosis on disease phase with modeling unit
Close grouping DRG and model classifications module, conjunction complication and its dependent variable sorts out collection modules, statistics of variable is examined and screening mould
Block, statistical models establish module, model quality authentication module：Historical data import modul is used to import from hospital database
Historic discharged patient's data；Data scrubbing module filters out bad data and extreme value data for completing data discriminating and cleaning
And it is deleted；Medical diagnosis on disease associated packets DRG and model classifications module are used to complete medical diagnosis on disease associated packets DRG and model
Classification, realize the classification set to medical diagnosis on disease associated packets, category of model, number sort out set；Close complication and other
Variable sort out collection modules for complete be admitted to hospital when diseases and related health problems international statistical classification ICD close complication and its
The classification set of dependent variable realizes the classification to inpatient's complication and complication and its dependent variable；Statistics of variable examine and
Screening module, which is used to implement, screens the variable with statistically significant meaning；Statistical models establish module for completing
The foundation of statistical models, patient death rate data use Logic Regression Models, and the data of length of stay and cost are then using more
First linear regression model (LRM) ultimately forms the quantitative formula of prediction by modeling；Model quality authentication module is used for using statistics
In C-Index examine and the R-square methods of inspection calculate model in sample population and non-sample crowd, according to phase
The result answered is evaluated；
Current data is screened includes current data import modul, data scrubbing module, medical diagnosis on disease with pre-value computing unit
Associated packets DRG and model classifications module close complication and its dependent variable classification collection modules, risk profile value of being admitted to hospital calculating mould
Block：Current data import modul is used to import current discharged patient's data from hospital database；Data scrubbing module has been used for
Differentiate into data and clear up, filter out bad data and deleted；Medical diagnosis on disease associated packets DRG and model classifications module are used for
The classification of medical diagnosis on disease associated packets DRG and model is completed, realizes the classification set to medical diagnosis on disease associated packets, model point
Class, number sort out set；It closes complication and its dependent variable is sorted out disease and related health when collection modules are admitted to hospital for completion and asked
The international statistical classification ICD of topic closes complication and its classification set of dependent variable, realizes to inpatient's complication and complication
And its classification of dependent variable；Risk profile value computing module of being admitted to hospital is used to implement to each inpatient in the death rate, day of being hospitalized
The risk profile of being admitted to hospital of number and medical treatment cost；The risk profile of disease refers to by the historical data of each disease management of hospital
Find out universal law to influencing final treatment results and can quantization factor, thus it is speculated that going out currently has similar disease degree and similar spy
The prediction occurrence value of the death rate of the patient of sign, length of stay and medical treatment cost；
The algorithmic formula of predicted value is as follows：
Expected mortalityWherein, biRepresent significantly correlated property coefficient, b0Represent model intercept, n tables
Show the significant correlation variable number of patient；
Length of stay and medical treatment costWherein, b0Represent model intercept, MSE
Represent the square error of model, biRepresent significantly correlated property coefficient, 0.5 is statistic bias correction value.
As shown in Fig. 2, quality of hospital management inquiry dimension includes depth, time, type, patient's information and hospital management：
Depth includes ranking, 10 percentiles, 25 percentiles, median；Time includes past and present, annual, season, monthly；
Type includes ICD, DRG, clinical sub- subject, clinical speciality, hospital, area, the country, the world；Patient's information include population information,
Way of paying, admission information, discharge information；Hospital management includes quality of medical care, medical efficiency, Medical Benefit, patient satisfaction.
The user that the user inquires terminal includes clinical users, hospital management regulatory agency user：
For clinical users：The clinical medicine that clinical users formulate hospital, DMIAES by hospital's classification Risk Adjusted model
Section, discharge section office, all kinds of doctors, disease DRG, discharge time section, primary and secondary will diagnose or perform the operation, patient enters discharge situation, patient
Age, query result and the regional anonymous comparison of other hospitals progress of gender, patient satisfaction of all categories, realize various differences
The compound query of conditional combination is shown with value, predicted value and O/E index results are actually occurred；O/E indexes i.e. actually occur value/in advance
Time value, O/E indexes<1：Illustrate disease risks height, but case fatality rate, length of stay or cost control are less than expection；O/E indexes>1：
Illustrate that disease risks are low, but case fatality rate, length of stay or cost control are higher than expection.
For hospital management regulatory agency user：The inquiry of hospital management regulatory agency user is superior to clinical users,
Hospital real name can be used to realize to compare, user according to hospital's classification Risk Adjusted model in region self-defined hospital,
Clinical speciality that DMIAES is formulated, discharge section office, disease DRG, discharge time section, primary and secondary will diagnose or perform the operation, patient age, property
Not, classification and aggregation patient satisfaction are inquired, and pass through the compound comparison query combined across the different condition of hospital, inquiry knot
Fruit is arranged according to user-defined sequence.
It is arranged below with hospital A in 1/1/2014-12/31/2014 time Nei Ge section office's quality of medical care ranking, operation efficiency
Name, cost control ranking, satisfaction ranking query result for, illustrate this patent 360 degrees omnidirection compare query function：
By clinical speciality (note：The classification of clinical speciality is divided by relevant disease diagnosis group DRG, and non-hospital reality
Section office, the reasonability being advantageously implemented under same medical diagnosis on disease compares) quality of medical care ranking：The O/E indexes of the sick death rate
From small to large
By the medical efficiency ranking of clinical speciality：The O/E indexes of the sick death rate are from small to large
By the Medical Benefit ranking of clinical speciality：Disease calculates the O/E indexes of expense from small to large
By the patient satisfaction ranking of clinical speciality：The value of the total Satisfaction Survey of Patients of hospital is from big to small
Below with quality of medical care ranking of certain ten hospital in certain region within the 1/1/2014-12/31/2014 times, battalion
For the query result for transporting efficiency ranking, cost control ranking, illustrate that the 360 degrees omnidirection of this patent compares query function：
By the quality of medical care ranking of hospital：The O/E indexes of the sick death rate are from small to large
By the medical efficiency ranking of hospital：The O/E indexes of the sick death rate are from small to large
By the Medical Benefit ranking of hospital：Disease calculates the O/E indexes of expense from small to large
The implementation method of hospital's disease control intellectual analysis and assessment system, including a historical data screening and modeling
Step and a current data screening calculate step with pre-value：
S1：Historic discharged patient's data are imported from hospital database；
S2：Data differentiate and cleaning, filter out bad data and extreme value data and are deleted；It is completed using computer programming
The cleaning of bad data and extreme value data.
The definition of bad data：1. in data row space data (such as without patient's essential information, enter information of leaving hospital, disease and
International statistical classification (ICD codings) in relation to health problem etc.)；2. the patient data repeated.
The definition of extreme value data：1. length of stay is the extreme value patient data of 0 day；2. length of stay is outside 99 percentage
Extreme value patient data；3. direct cost of being hospitalized is less than 900 yuan of extreme value patient data；4. after death patient's number of remains donations
According to.
S3：The classification of medical diagnosis on disease associated packets DRG and model, realize the classification set to medical diagnosis on disease associated packets,
Category of model, number sort out set；It is realized using the commercial DRG code machines of 3M and sorts out set.
The definition of DRG：According to patient disease's diagnosis, operation type, complication and complication, discharge situation, gender and year
The sorting and grouping in age etc., there are one DRG by each patient；The classification and assessment of correlation diagnosis are advantageously implemented by DRG, simultaneously
Diagnosis quantity is reduced, improves the relevance grade of model prediction；
The definition of category of model：According to DRG, associated DRG is sorted out, each DRG is incorporated into a model
Number, DMIAES bases DRG is also, by pattern number/DMIAES bases DRG, the classification of correlation DRG is advantageously implemented and comments
Estimate；
S4：The international statistical classification ICD of diseases and related health problems closes the classification of complication and its dependent variable when being admitted to hospital
Set, realizes the classification to inpatient's complication and complication and its dependent variable；Using the international medical diagnosis on disease for closing complication
Criteria for classification completes the classification set of variable using computer programming.
Close the definition of complication variable：According to the international Disease Diagnosis Standard classification of human organ and system to patient admission
When the past complication and complication carry out classification processing (see example one)；It is handled by classification, advantageously reduces medical diagnosis on disease/hand
The quantity of art variable improves the stability of model prediction performance；
The definition of its dependent variable：Its dependent variable includes the letters such as age, gender, social economic environment, situation of being admitted to hospital and source
Breath；Be conducive to the comprehensive sexual factor such as patient and admission information taking into account in Risk Adjusted model that the assessment for risk of being admitted to hospital is more
Completely.
Using international disease close complication collective standard, in same relevant disease group DRG patient ICD diagnosis or
Operation encoding carries out cluster set, is formed and closes complication class variable, as shown in example one；
Example one：The classification variable of disease code and population information
Bacterial endocarditis closes complication group (ICD9 codings)
Patient basic population information and state variable (part) of being admitted to hospital are as follows：
S5：In same DRG groups, have by statistical test method to patient death rate, length of stay and cost
The conjunction complication group of statistically significant meaning and other class variables are pre-processed.
S6：The influence degree for carrying out regression analysis quantization variable the step of establishing statistical models by model；
The description of quantification is carried out to predictive variable using independent variable, in disease risks adjustment, predictive variable is disease
People's death rate, length of stay and cost in hospital, independent variable is closes complication variable and its dependent variable；It is returned by model
The influence degree of analysis quantization variable (see example two).
Patient death rate data use Logic Regression Models modeling method, length of stay and medical treatment cost data using polynary
Linear regression model (LRM) modeling method.Statistical model is established using LASSO homing methods, and carries out ginseng on the training data automatically
Number optimization, establishes optimal model.In the linear regression model (LRM) to length of stay and cost foundation, predictive variable value is carried out
Log transformation so that the predictive variable value after transformation is more in line with the hypothesis of linear regression model (LRM) closer to normal distribution；
Model provides as follows the regression coefficient of independent variable：Closing complication variation coefficient ＞ 0 (increases the death rate, in hospital
The risk of number of days and cost)；Its dependent variable (such as age) coefficient can just be born.In addition, variable symbol must be with just choosing system
The symbol that meter obtains in examining is consistent.
Example two：DMIAES disease mortality models #22：(patient age >=18) acute ischemic stroke and use thrombus
Agent closes complication (MSDRG 61) with serious, closes complication (MSDRG 62), no conjunction complication (MSDRG 63).
Data source：Texas,U.S medical center Herman memorial hospital
Patient's number in modeling sample：996 sample time 7/1/2004-6/30/2014
Model classification：Logic Regression Models
| Close the explanatory variable of complication | Related coefficient |
| Intercept | -4.159 |
| Big brain compression | 2.272 |
| Women, age 75-80 Sui | 1.547 |
| Tracheae built-in pipe | 1.488 |
| Upper lung ventilator in 48 hours when being admitted to hospital | 1.392 |
| It abandons rescuing | 1.388 |
| Epilepsy | 1.344 |
| Acid poisoning | 1.285 |
| Women, 85 years old age or more | 1.092 |
| Brain edema | 0.943 |
| Cardiac arrhythmia | 0.906 |
| Acute respiratory failure | 0.322 |
| Atrial fibrillation | 0.057 |
S7：The model quality verification step includes：
(1) basic comparative analysis：It is analyzed using the comparison with same class model, the data of same test sample is inputted two
Classification, which is compared, to be realized to result after model；Although the mode of selection with the modeling of variable is different, final result still has
(see six chart of example) of comparativity
(2) using statistical testing of business cycles method, including：
1. the inspection of Logic Regression Models：Computation model predicts the coefficient C-Index that is harmonious with actual value；Wherein C-
For Index values closer to 1, the prediction effect of model is better.External comparison model requires the C-Index ＞ 0.7 of model in examining.
(the C test values of example 3 1)
2. the inspection of linear regression model (LRM)：Computation model predicts the fitting coefficient R-square with actual value；R-square
For value closer to 1, the prediction effect of model is better.External comparison model requires the R-square ＞ 0.05 of model in examining.(example three
2,3 R test values)
3. testing in test data, test data is for individually a data or in other conditions (as treated
Approach, means etc.) do not sexually revise at all under the premise of current patient data (reality/predicted value of example 6 1,2,3)；
4. carrying out C and R with class model with other with same test sample data to examine, analysis model quality (example is compared
6 1,2,3 test value).
Example three：The quality verification result of model and comparison are analyzed
1. the sick death rate
Model #22：Acute ischemic stroke DRG 61,62,63
Data source：Texas,U.S medical center Herman memorial hospital
Test sample patient's number：66 discharge time 7/1/2004-6/30/2014
Comparison model：The same class model of our DMIAES models and the U.S. (abbreviation U models)
Model-fitting degree C-Index is examined：DMIAES models in modeling data：0.890, DMIAES moulds in test data
Type：0.964, U model：0.933.
Model #328：Multiple surgery wound DRG 957,958,959
Data source：Texas,U.S medical center Herman memorial hospital
Test sample number：212 discharge time 7/1/2004-6/30/2014
Comparison model：The same class model of our DMIAES models and the U.S. (abbreviation U models)
Model-fitting degree C-Index is examined：DMIAES models in modeling data：0.955, DMIAES moulds in test data
Type：0.987, U model：0.982.
2. length of stay
Model #22：Acute ischemic stroke DRG 61,62,63
Data source：Texas,U.S medical center Herman memorial hospital
Test sample patient's number：66 discharge time 7/1/2004-6/30/2014
Comparison model：The same class model of our DMIAES models and the U.S. (abbreviation U models)
Model-fitting degree R-square is examined：DMIAES models in modeling data：0.244, DMIAES moulds in test data
Type：0.219, U model：0.211.
Model #76：Heart valve and other class cardiothoracic surgeries DRG 219,220,221
Data source：Texas,U.S medical center Herman memorial hospital
Test sample patient's number：199 discharge time 7/1/2004-6/30/2014
Comparison model：The same class model of our DMIAES models and the U.S. (abbreviation U models)
Model-fitting degree R-square is examined：DMIAES models in modeling data：0.256, DMIAES moulds in test data
Type：0.261, U model：0.269.
3. Direct medical cost of being hospitalized
Model #22：Acute ischemic stroke DRG 61,62,63
Data source：Texas,U.S medical center Herman memorial hospital
Test sample patient's number：66 discharge time 7/1/2004-6/30/2014
Comparison model：The same class model of our DMIAES models and the U.S. (abbreviation U models)
Model-fitting degree R-square is examined：DMIAES models in modeling data：0.366, DMIAES moulds in test data
Type：0.217, U model：0.149.
Model #205：Diabetes DRG 637,638,639
Data source：Texas,U.S medical center Herman memorial hospital
Test sample patient's number：119 discharge time 7/1/2004-6/30/2014
Comparison model：The same class model of our DMIAES models and the U.S. (abbreviation U models)
Model-fitting degree R-square is examined：DMIAES models in modeling data：0.383, DMIAES moulds in test data
Type：0.227, U model：0.295.
The current data screening calculates step with pre-value and includes the following steps：
SS1：Current discharged patient's data are imported from hospital database；
SS2：Data differentiate and cleaning, filter out bad data and are deleted；
SS3：The classification of medical diagnosis on disease associated packets DRG and model, realize the classification set to medical diagnosis on disease associated packets,
Category of model, number sort out set；
There are one medical diagnosis on disease associated packets DRG by each patient, and the classification and assessment of correlation diagnosis are realized by DRG；
According to DRG, associated DRG is sorted out, each DRG is incorporated into a pattern number, passes through pattern number reality
The classification and assessment of existing correlation DRG；
SS4：The international statistical classification ICD of diseases and related health problems closes the classification of complication and its dependent variable when being admitted to hospital
Set, realizes the classification to inpatient's complication and complication and its dependent variable；
The complication of the past and conjunction during according to the international Disease Diagnosis Standard classification of human organ and system to patient admission
And disease carries out classification processing；
Its dependent variable includes age, gender, social economic environment, situation of being admitted to hospital and source-information；
SS5：The predicted value of patient admission risk is calculated, is realized to each inpatient in the death rate, length of stay and doctor
Treat the risk profile of being admitted to hospital of cost；
The definition of risk profile value and establishment condition：
Definition：The risk profile of disease refers to final to influencing by being found out in the historical data of each disease management of hospital
The universal law for the treatment of results and can quantization factor, can using the methods of big data analysis, mathematical statistics and machine learning
The prediction of the accurate death rate for deducing the current patient for having similar disease degree and similar features, length of stay and medical treatment cost
Occurrence value；
Establishment condition：Diagnosis coding, diagnosis classifying method, the therapy approach of disease and means and Medical Treatment Price etc. are modeling
Change in current slot that essence does not occur.
Specific method is to calculate predicted value according to all kinds of formula (see example four)：
Example four：DMIAES disease mortality models #22：(patient age >=18) acute ischemic stroke and use thrombus
Agent closes complication (MSDRG 61) with serious, closes complication (MSDRG 62), no conjunction complication (MSDRG 63).
Data source：Texas,U.S medical center Herman memorial hospital
Patient's number in modeling sample：996 sample time 7/1/2004-6/30/2014
Model classification：Logic Regression Models
Degree of fitting in modeling sample：C-Index=0.890
Patient's death average expectancy rate in model：68.4%
Patient's pre-value of dying of illness is 1.54% during no disease variable, has pre-value during multiple disease variables to be raised to 64.14%.
Risk when example one is shown using the disease risks adjustment model acute ischemic stroke patient admission different to two
Prediction due to the age of patient, gender, closes the differences such as complication and disease degree and causes to die of illness the different disease risks systems died
Number.
The algorithmic formula of predicted value is as follows：
Expected mortalityWherein, biRepresent significantly correlated property coefficient, b0Represent model intercept, n tables
Show the significant correlation variable number of patient；
Length of stay and medical treatment costWherein, b0Represent model intercept, MSE
Represent the square error of model, biRepresent significantly correlated property coefficient, 0.5 is statistic bias correction value；
Final use actually occurs and is expected relative value and inpatient's medical control quality is assessed.
The above is only the preferred embodiment of the present invention, it should be understood that the present invention is not limited to described herein
Form is not to be taken as the exclusion to other embodiment, and available for other combinations, modifications, and environments, and can be at this
In the text contemplated scope, modifications can be made through the above teachings or related fields of technology or knowledge.And those skilled in the art institute into
Capable modifications and changes do not depart from the spirit and scope of the present invention, then all should be in the protection domain of appended claims of the present invention
It is interior.
Claims (3)
1. hospital's disease control intellectual analysis and assessment system, which is characterized in that including patient terminal, evaluating server and user
Terminal is inquired, patient terminal and user inquire terminal and be connected respectively by communication network with evaluating server；
Collection disease follow-up APP is provided on the patient terminal, the satisfaction information of patient after leaving hospital for dynamic acquisition；
The evaluating server includes clinical data introducting interface module, quality of hospital management evaluation module and user's inquiry and connects
Mouth mold block：
Clinical data introducting interface module imports clinical data for quality of hospital management for connecting hospital clinical data center
Evaluation module carries out the assessment of medical control quality；
Quality of hospital management evaluation module is used to be realized to each inpatient in the death rate, day of being hospitalized according to hospital clinical data
The risk profile of being admitted to hospital of number and medical treatment cost finally controls influence by being found out in the historical data of each disease management of hospital
Treat result universal law and can quantization factor, thus it is speculated that go out the death of the current patient for having similar disease degree and similar features
The prediction occurrence value of rate, length of stay and medical treatment cost；
User's query interface module is used to inquire terminal offer user interface to user；
The user inquires terminal and is used for assessment result according to quality of hospital management evaluation module, realizes between different dimensions
A variety of screenings, independent assortment and comprehensive comparison, while realize a variety of screenings, combination and comprehensive comparison between dimension of the same race；
User inquires terminal can also inquire the satisfaction assessment that discharged patient feeds back by evaluating server, and realize in same standard
On patient satisfaction analysis and comparison；
The quality of hospital management evaluation module includes a historical data screening and is sieved with modeling unit, a current data
Choosing and pre-value computing unit：
Historical data is screened includes historical data import modul, data scrubbing module, related point of medical diagnosis on disease to modeling unit
Group DRG and model classifications module, conjunction complication are sorted out collection modules, statistics of variable inspection and screening module, statistical models and are built
Formwork erection block, model quality authentication module：Historical data import modul is used to import historic discharged patient from hospital database
Data；Data scrubbing module filters out bad data and extreme value data and is deleted for completing data discriminating and cleaning；Disease
Diagnosis associated packets DRG and model classifications module are used to complete the classification of medical diagnosis on disease associated packets DRG and model, realize to disease
Classification set, category of model, the number of disease diagnosis associated packets sort out set；Close complication sort out collection modules for complete into
The international statistical classification ICD of diseases and related health problems closes the classification set of complication during institute, realizes and inpatient is merged
Disease and the classification of complication；Statistics of variable is examined and screening module is used to implement and the variable with statistically significant meaning is carried out
Screening；Statistical models establish module for completing the foundation of statistical models, and patient death rate data use logistic regression mould
Using multiple linear regression model, the quantitative formula of prediction is then ultimately formed by modeling for the data of type, length of stay and cost；
Model quality authentication module is used to examine with the R-square methods of inspection to model in sample people using the C-Index in statistics
Group and non-sample crowd are calculated, and are evaluated according to corresponding result；
Current data is screened related including current data import modul, data scrubbing module, medical diagnosis on disease to pre-value computing unit
It is grouped DRG and model classifications module, closes complication classification collection modules, risk profile value computing module of being admitted to hospital：Current data is led
Enter module for importing current discharged patient's data from hospital database；Data scrubbing module differentiates for completing data and clear
Reason, filters out bad data and is deleted；Medical diagnosis on disease associated packets DRG and model classifications module are used to complete medical diagnosis on disease phase
The classification of grouping DRG and model is closed, realizes that the classification set to medical diagnosis on disease associated packets, category of model, number sort out set；
The international statistical classification ICD for closing diseases and related health problems when complication sorts out collection modules for completing to be admitted to hospital closes complication
Classification set, realize the classification to inpatient's complication and complication；Risk profile value computing module of being admitted to hospital is used to implement
To each inpatient the death rate, length of stay and medical treatment cost risk profile of being admitted to hospital；The risk profile of disease, which refers to, to be passed through
Found out in the historical data of each disease management of hospital universal law to influencing final treatment results and can quantization factor, push away
Measure the prediction of the death rate of the current patient for having similar disease degree and similar features, length of stay and medical treatment cost
Value；
The algorithmic formula of predicted value is as follows：
Expected mortality is：
Wherein, biRepresent significantly correlated property coefficient, b0Represent model intercept, n represents the significant correlation variable number of patient；
Length of stay and medical treatment cost are：
Wherein, b0Represent model intercept, MSE represents the square error of model, biRepresent significantly correlated property coefficient, 0.5 is inclined for statistics
Poor correction value.
2. hospital's disease control intellectual analysis according to claim 1 and assessment system, which is characterized in that hospital management matter
Amount inquiry dimension includes depth, time, type, patient's information and hospital management：Depth includes ranking, 10 percentiles, 2,500
Quantile, median；Time includes past and present, annual, season, monthly；Type includes ICD, DRG, clinical sub- subject, faces
Bed subject, hospital, area, the country, the world；Patient's information includes population information, way of paying, admission information, discharge information；Doctor
Institute's management includes quality of medical care, medical efficiency, Medical Benefit, patient satisfaction.
3. hospital's disease control intellectual analysis according to claim 1 and assessment system, which is characterized in that the user looks into
The user for asking terminal includes clinical users, hospital management regulatory agency user：
For clinical users：Clinical users by hospital's classification Risk Adjusted model to hospital, hospital's disease control intellectual analysis and
Clinical speciality that assessment system is formulated, discharge section office, all kinds of doctors, disease DRG, discharge time section, primary and secondary will diagnose or perform the operation,
Patient enters discharge situation, patient age, gender, the query result of patient satisfaction of all categories and other hospitals of regionality and hides
Name compares, and realizes the compound query of various different condition combinations and actually occurs value, predicted value and the displaying of O/E index results,
Middle O/E exponential representations actually occur value/desired value, O/E indexes<1：Illustrate disease risks height, but case fatality rate, length of stay or into
This control is less than expection；O/E indexes>1：Illustrate that disease risks are low, but case fatality rate, length of stay or cost control are higher than expection；
For hospital management regulatory agency user：The inquiry of hospital management regulatory agency user is superior to clinical users, can
Realized and compared using hospital real name, user according to hospital's classification Risk Adjusted model in region self-defined hospital, hospital's disease
Disease management intellectual analysis and the clinical speciality of assessment system formulation, discharge section office, disease DRG, discharge time section, primary and secondary will diagnose
Or operation, patient age, gender, classification and aggregation patient satisfaction are inquired, and pass through what is combined across the different condition of hospital
Compound comparison query, query result are arranged according to user-defined sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510359039.7A CN104951894B (en) | 2015-06-25 | 2015-06-25 | Hospital's disease control intellectual analysis and assessment system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510359039.7A CN104951894B (en) | 2015-06-25 | 2015-06-25 | Hospital's disease control intellectual analysis and assessment system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104951894A CN104951894A (en) | 2015-09-30 |
| CN104951894B true CN104951894B (en) | 2018-07-03 |
Family
ID=54166532
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510359039.7A Active CN104951894B (en) | 2015-06-25 | 2015-06-25 | Hospital's disease control intellectual analysis and assessment system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104951894B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3806107A1 (en) * | 2019-10-11 | 2021-04-14 | C The Signs Limited | Diagnostic tool |
Families Citing this family (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303064B (en) * | 2015-12-03 | 2018-10-16 | 万达信息股份有限公司 | A kind of construction method of the medical index system based on semantic model |
| US20190057762A1 (en) * | 2016-02-26 | 2019-02-21 | Toyosaki Accounting Office Co., Ltd. | Information processing device |
| CN105893765B (en) * | 2016-04-06 | 2018-09-11 | 成都数联易康科技有限公司 | A kind of classification diagnosis and treatment analysis and data visualisation system based on Echarts |
| CN105893766B (en) * | 2016-04-06 | 2022-03-18 | 成都数联易康科技有限公司 | Grading diagnosis and treatment evaluation method based on data mining |
| CN106202955B (en) * | 2016-07-19 | 2019-03-01 | 中电科软件信息服务有限公司 | Diagnosis associated packets method and system based on intellectual coded adaptation |
| CN106777898B (en) * | 2016-11-23 | 2019-06-11 | 北京汉博信息技术有限公司 | Medical information method for pushing |
| CN107463770A (en) * | 2017-07-11 | 2017-12-12 | 武汉金豆医疗数据科技有限公司 | A kind of evaluation method and system based on medical diagnosis on disease associated packets |
| CN107610761B (en) * | 2017-09-30 | 2020-06-23 | 电子科技大学 | A clinical pathway analysis method based on medical insurance data |
| CN107657991B (en) * | 2017-11-13 | 2021-01-29 | 医渡云（北京）技术有限公司 | Patient data screening method and device, storage medium and electronic equipment |
| CN107967948B (en) * | 2017-12-07 | 2021-10-08 | 泰康保险集团股份有限公司 | Medical big data analysis method and device |
| WO2019219660A1 (en) * | 2018-05-15 | 2019-11-21 | Koninklijke Philips N.V. | System and method for providing model-based predictions of actively managed patients |
| CN108846503B (en) * | 2018-05-17 | 2022-07-08 | 电子科技大学 | A neural network-based dynamic prediction method for the number of patients with respiratory diseases |
| CN109509549B (en) * | 2018-05-28 | 2024-05-14 | 深圳平安医疗健康科技服务有限公司 | Diagnosis and treatment service provider evaluation method, diagnosis and treatment service provider evaluation device, computer equipment and storage medium |
| CN110610761A (en) * | 2018-06-14 | 2019-12-24 | 胡盛寿 | Hypertension auxiliary diagnosis method and system |
| CN108876636B (en) * | 2018-06-19 | 2023-10-27 | 平安健康保险股份有限公司 | Intelligent air control method, system, computer equipment and storage medium for claim settlement |
| CN109192312B (en) * | 2018-08-20 | 2021-03-09 | 中国人民解放军总医院 | A system and method for intelligent management of adverse events in patients with heart failure |
| CN109036579B (en) * | 2018-08-22 | 2021-11-26 | 泰康保险集团股份有限公司 | Information prediction method, device, medium and electronic equipment based on block chain |
| CN109493975B (en) * | 2018-12-20 | 2021-10-01 | 广州医科大学附属第一医院 | Method, device and computer equipment for chronic disease recurrence prediction based on xgboost model |
| CN109949940B (en) * | 2019-03-11 | 2022-11-29 | 金力 | A big data mining system based on diabetes database |
| CN110051324B (en) * | 2019-03-14 | 2022-06-10 | 深圳大学 | Method and system for predicting death rate of acute respiratory distress syndrome |
| CN110930030B (en) * | 2019-11-22 | 2022-09-02 | 北京雅丁信息技术有限公司 | Doctor skill level rating method |
| CN111180055B (en) * | 2019-12-31 | 2023-06-30 | 重庆亚德科技股份有限公司 | Hospital supervision system and method |
| CN111489821B (en) * | 2020-03-31 | 2021-02-02 | 宜昌市中心人民医院(三峡大学第一临床医学院、三峡大学附属中心人民医院) | Diagnostic group management system |
| CN111667898B (en) * | 2020-06-19 | 2021-08-27 | 重庆中联信息产业有限责任公司 | Classification processing method of medical medication rule information |
| CN111613292B (en) * | 2020-06-19 | 2021-02-09 | 和宇健康科技股份有限公司 | Reasonable medication judgment decision system based on big data |
| CN111816288B (en) * | 2020-07-22 | 2023-10-13 | 望海康信(北京)科技股份公司 | Hospital operation system |
| CN111899867B (en) * | 2020-07-28 | 2022-08-05 | 四川大学华西医院 | Auxiliary decision-making system for prediction and avoidance of surgical complications based on deep learning |
| CN112037918B (en) * | 2020-11-02 | 2021-02-12 | 平安科技（深圳）有限公司 | Chronic disease medical insurance cost prediction method fusing complication risks and related equipment |
| CN112635030A (en) * | 2020-11-30 | 2021-04-09 | 望海康信(北京)科技股份公司 | Automatic generation system and method for fee control prompt, corresponding equipment and storage medium |
| CN113160976B (en) * | 2021-04-29 | 2024-08-16 | 上海商汤善萃医疗科技有限公司 | Medical data processing method and device based on SaaS service and electronic equipment |
| CN114153871B (en) * | 2021-12-07 | 2022-10-11 | 上海柯林布瑞信息技术有限公司 | Data query optimization method and device based on DIP (dual in-line package) grouping |
| CN115185936B (en) * | 2022-07-12 | 2023-02-03 | 曜立科技(北京)有限公司 | Medical clinical data quality analysis system based on big data |
| CN117524454B (en) * | 2024-01-05 | 2024-03-12 | 南京横渡医疗技术有限公司 | Medical data safety monitoring system and method based on Internet |
| CN118822045B (en) * | 2024-09-18 | 2024-12-17 | 安徽医科大学第一附属医院 | Medical insurance DRG grouping prediction method and system based on artificial intelligence technology |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101178804A (en) * | 2007-12-05 | 2008-05-14 | 卫生部医院管理研究所 | Hospital information system construction method taking medical quality as core |
| CN101554322A (en) * | 2008-04-09 | 2009-10-14 | 陈敦金 | System for estimating state of critically ill patient in obstetrical department |
| CN103201743A (en) * | 2010-11-08 | 2013-07-10 | 皇家飞利浦电子股份有限公司 | Method of continuous prediction of patient severity of illness, mortality, and length of stay |
| CN103488856A (en) * | 2012-06-15 | 2014-01-01 | 国家人口计生委科学技术研究所 | Low birth weight risk prediction system and method for constructing low birth weight risk prediction system |
| CN104063579A (en) * | 2014-05-16 | 2014-09-24 | 上海亿保健康管理有限公司 | Health dynamic prediction method and equipment based on multivariate medical consumption data |
-
2015
- 2015-06-25 CN CN201510359039.7A patent/CN104951894B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101178804A (en) * | 2007-12-05 | 2008-05-14 | 卫生部医院管理研究所 | Hospital information system construction method taking medical quality as core |
| CN101554322A (en) * | 2008-04-09 | 2009-10-14 | 陈敦金 | System for estimating state of critically ill patient in obstetrical department |
| CN103201743A (en) * | 2010-11-08 | 2013-07-10 | 皇家飞利浦电子股份有限公司 | Method of continuous prediction of patient severity of illness, mortality, and length of stay |
| CN103488856A (en) * | 2012-06-15 | 2014-01-01 | 国家人口计生委科学技术研究所 | Low birth weight risk prediction system and method for constructing low birth weight risk prediction system |
| CN104063579A (en) * | 2014-05-16 | 2014-09-24 | 上海亿保健康管理有限公司 | Health dynamic prediction method and equipment based on multivariate medical consumption data |
Non-Patent Citations (1)
| Title |
|---|
| "医院信息数据挖掘及实现技术的探索";易静;《中国博士学位论文全文数据库医药卫生科技辑》;20080215(第2期);第E053-2页 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3806107A1 (en) * | 2019-10-11 | 2021-04-14 | C The Signs Limited | Diagnostic tool |
| WO2021069105A1 (en) * | 2019-10-11 | 2021-04-15 | C The Signs Limited | Diagnostic tool |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104951894A (en) | 2015-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104951894B (en) | Hospital's disease control intellectual analysis and assessment system | |
| CN105046406A (en) | Inpatient medical management quality assessment method | |
| CN106202955A (en) | Diagnosis associated packets method and system based on intellectual coded adaptation | |
| CN113808738B (en) | Disease identification system based on self-identification image | |
| CN107463769A (en) | A kind of medical receipt and payment expense management method and system based on medical diagnosis on disease associated packets | |
| CN108898476A (en) | A kind of loan customer credit-graded approach and device | |
| CN106897546A (en) | Medical information aids in doctor's diagnosis and therapy system and method | |
| DE102007007190A1 (en) | Patient classification predicting method for medical treatment, involves predicting classifications of medical treatment for patient based on patient treatment target data and set of predicted modules | |
| CN106202891A (en) | A kind of big data digging method towards Evaluation of Medical Quality | |
| CN107463771A (en) | A kind of method and system of case packet | |
| CN109192316B (en) | A disease subtype prediction system based on gene network analysis | |
| CN112216361A (en) | Follow-up plan list generation method, device, terminal and medium based on artificial intelligence | |
| CN113657805A (en) | Method, device, equipment and storage medium for constructing weights of assessment index system | |
| Westerdijk et al. | Defining care products to finance health care in the Netherlands | |
| CN113053513A (en) | Wisdom medical system based on wisdom community | |
| CN112434737A (en) | Multi-agent evaluation criterion fused feature selection system for stroke analysis | |
| CN111261299A (en) | Multi-center collaborative cancer prognosis prediction system based on multi-source transfer learning | |
| CN117612679A (en) | Hospital department analysis method and system based on medical insurance DRG grouping | |
| Yu et al. | Cvcrand: A package for covariate-constrained randomization and the clustered permutation test for cluster randomized trials | |
| CN113345564B (en) | A method and device for early prediction of hospitalization length of patients based on graph neural network | |
| CN109949167A (en) | Medical insurance dynamic financing method and system based on big data modeling analysis | |
| CN113706111B (en) | Medical institution flow data processing method, device, equipment and medium | |
| CN106503889A (en) | A kind of subjective and objective integrated evaluating method of land use intensively and device | |
| CN113688854B (en) | Data processing method, device and computing equipment | |
| CN114298603A (en) | Pathological quality control index evaluation method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 610041 No. 6, floor 4, unit 2, building 1, No. 222, Tianren Road, high tech Zone, Chengdu, Sichuan Patentee after: Zhongke Houli information technology (Chengdu) Co.,Ltd. Address before: 610041 No. 6, floor 4, unit 2, building 1, No. 222, Tianren Road, high tech Zone, Chengdu, Sichuan Patentee before: CHENGDU HOULI INFORMATION TECHNOLOGY Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |