CN110060776A - Assessment performance data - Google Patents
Assessment performance data Download PDFInfo
- Publication number
- CN110060776A CN110060776A CN201811517790.5A CN201811517790A CN110060776A CN 110060776 A CN110060776 A CN 110060776A CN 201811517790 A CN201811517790 A CN 201811517790A CN 110060776 A CN110060776 A CN 110060776A
- Authority
- CN
- China
- Prior art keywords
- processor
- observation result
- performance
- variable
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
It is configured for assessing the system (100) for showing data the invention discloses a kind of, the system comprises: memory (106) comprising indicate the director data of the set of instruction；And processor (102), it is configured as that the set of described instruction is communicated and executed with the memory.The set of described instruction makes the processor when being executed by the processor: acquisition multiple performance data records associated with performance indicators；Classified according to the performance indicators to each performance data record；Identify multiple variables that the classification is contributed in the performance data record；Multiple observations relevant at least one variable in the performance indicators and the multiple variable are determined based on the multiple variable；And the multiple observation is delivered for being presented to the user.Also disclose a kind of method.
Description
Technical field
The present invention relates to assessment data, and are more particularly to assessment data relevant to the movement having been carried out.
Background technique
General background of the invention is data analysis and parsing.In many fields, it can be used to compare one or more
A measurement is to measure the performance specifically acted (performance), to assess the performance and to recognize the need for any
Modification.In some cases, data relevant to the movement having been carried out can be according to one or more crucial performance indicators
(KPI) it is assessed.Crucial performance indicators are a type of performance measurements, and make it possible to about specific activities, appoint
The success of the performance of business or movement is assessed.
KPI for example can be arranged or be chosen by tissue or individual, and can indicate the specific objective for being intended to realize or
Purpose.Data about having been carried out for task and its mode being performed (e.g., including time scale) can be can
?.Pass through data as observation result, it may be possible to determine whether task is performed to comply with standard or desired
Within time scale.In other words, it can determine whether performed task meets specific KPI.
Such KPI statistical result and data can provide the details in the field that can be further investigated for user, with
Just the possible cause for meeting or being unsatisfactory for specific KPI is determined.However, the data cannot be provided for user to can or not be able to satisfy
The understanding that is useful, being easy to interpretation of the reason of KPI.
Therefore, it is intended that can be used for assessing or analyze performance data with a kind of so as to according to about specifically executing
The such data why movement meets or do not meet specific performance measurement (for example, KPI) are valuable to determine
The system of information.
Summary of the invention
According in a first aspect, a kind of system for being configured for assessment performance data, comprising: memory comprising indicate
The director data of the set of instruction；And processor, it is configured as communicating with the memory and executes described instruction
Set, wherein the set of described instruction makes the processor when being executed by the processor: acquisition is associated with performance indicators
Multiple performance data records；Classified according to the performance indicators to each performance data record；Identify the performance number
According to the multiple variables for contributing to the classification in record；It is determined based on the multiple variable and performance indicators and described more
The relevant multiple observation results of at least one variable in a variable；And the multiple observation result is delivered for being presented to use
Family.Suffered observation result is considered ' it was found that ' and can referred to as ' it was found that '.
By the way that the set of observation result is determined or generated based on the variable, the system can provide logarithm for user
According to the summary for being easy to interpret or about the opinion of data, otherwise from initial data may not be obvious.Cause
This, the amount of user time and system time needed for the system can reduce analysis initial data.By being provided pair for user
The convenient summary of data, more times can be spent in using the system for other purposes, and user can
It takes more time to execute other tasks, the relevant task of such as medicine.
The set of described instruction can also make the processor deliver the multiple sight when being executed by the processor
Result is examined for classifying before presentation to the multiple observation result.Therefore, more weights, which can be given, is considered as
Than other observation results there is those of higher priority to observe result.Therefore, can make with about performance indicators can be by
It easily adjusts and can have those of the factor significantly affected or variable correlation observation result pair to the performance is improved
User is clearly visible.
In some embodiments, making the processor may include making the processing to the graduation of the multiple observation result
Device calculates the statistical significance of each observation result.The set of described instruction can make described when being executed by the processor
Reason device classifies to the multiple observation result according to statistical significance calculated.
It is the multiple that the set of described instruction can also make the processor generate summary when being executed by the processor
Observe the summary of at least one of result observation result.Such summary summarizes the meaning for allowing user to understand data, and
Without analyzing initial data or carrying out a large amount of observation results.In fact, the system provides for user based on initial data
Conclusion allows to rapidly take any necessary operation to improve situation.
In some embodiments, so that the processor is generated summary may include carrying out the processor in following extremely
One item missing: predefined template is applied to the data at least one described observation result；And by natural language processing algorithm
Applied to the data at least one described observation result.
In some embodiments, the processor is made to identify that the multiple variable may include that come from the processor will
The data of the multiple performance data record are supplied to one or more prediction models.
In some embodiments, the set of described instruction can also make the processor when being executed by the processor:
Generate the classification standard for being directed to the performance indicators.Make the processor to each performance data record carry out classification may include
The processor is set to be classified according to the classification standard to each performance data record.
In some embodiments, the classification standard may include about whether the instruction for meeting the performance indicators.
The set of described instruction can also make the processor according at least one measurement when being executed by the processor
The multiple variable is filtered.At least one described measurement may include user-defined measurement.
In some embodiments, the set of described instruction can also make the processor when being executed by the processor:
Generate the graphical representation of at least one of the multiple observation result observation result；And the graphical representation is delivered for being in
Now give the user.It is described observation result graphical representation can be used family be easier to solution read raw data, and obtain about
The quick conclusion of any necessary operation of performance relevant to performance indicators should be improved by carrying out.
According to second aspect, a method of data are showed for assessing, comprising: acquisition is associated with performance indicators more
A performance data record；Classified according to the performance indicators to each performance data record；Identify the performance data note
Multiple variables of the classification are contributed in record；It is determined based on the multiple variable and the performance indicators and the multiple change
The relevant multiple observation results of at least one variable in amount；And the multiple observation result is delivered for being presented to the user.
In some embodiments, the method may include generate to summarize the sight of at least one of the multiple observation result
Examine the summary of result.
The method may include the summary and the observation corresponding with the summary presented is presented to user
As a result graphical representation.
According to the third aspect, a kind of computer program product including non-transient computer-readable media, the computer
Readable medium is embedded with computer-readable code wherein, and the computer-readable code is configured such that by suitable meter
Calculation machine or processor make the method for the computer or processor execution herein disclosed when executing.
With reference to embodiments described just below, these and other aspects of the present invention will be apparent and be explained
It is bright.
Detailed description of the invention
It, now will be only with model for a better understanding of the present invention and in order to more clearly illustrate how it can be carried out
The mode of example is referring to attached drawing, in the accompanying drawings:
Fig. 1 is the schematic illustration for the example for assessing the system of performance data；
Fig. 2 be include for multiple records performance data table；
Fig. 3 is the diagram of the example of graphical representation and observation result；
Fig. 4 is the flow chart of the example of the method for assessment performance data；
Fig. 5 is the flow chart of the other example of the method for assessment performance data；
Fig. 6 is the flow chart of the other example of the method for assessment performance data；And
Fig. 7 is the schematic illustration of machine readable media and processor.
Specific embodiment
Present disclose provides can inquire and analyze mass data record to obtain from initial data not to be by it
The mechanism of apparent valuable information immediately.Example herein is described under the background of medical data.However, will
, it is realized that disclosed system and method can apply to the field of wide scope, and any type can be applied to and come
Data from many different sources.
The specific area that system and method disclosed herein can be applied is health care industry.It is cardiovascular
Information system (CVIS) can be used to store and assess in some health care facilities the record of patient.For example, CVIS can
To be integrated in other one or more electronic systems (such as electric health record (EHR) or laboratory information system (LIS))
Together.The user of CVIS can observe the health records that result its data record is stored in the patient in CVIS, and the number
According to the purpose that can be used to report, arrange and manage.User can for example check the data for multidigit patient, and according to
One or more measurement (such as crucial performance indicators (KPI)) assesses the data.
The example for the such data element that can be specifically evaluated under the background of CVIS is that patient has to wait for
The time quantum of specific process (such as performing the operation).This is referred to as patient's waiting time or is called for short " waiting time ".Specific group
It knits (such as healthcare organization) and can have patient and be disposed within the period of definition (for example, executing specific stream to patient
Journey) specific objective or requirement.The defined period can be arbitrarily arranged by individual or it can for example be included in
In guide.Such example in UK is that patient waits the waiting time of medical should be lower than 6 weeks.In other words, it anticipates
Figure is should to wait for the specific process patient to be performed no longer than 6 weeks.
Data relevant to the waiting time of patient can be with other data (data such as relevant with patient, process
Type and/or the medical facilities or mechanism for executing process) it is included in together, and the data can be stored in form CVIS
Part database on.In general, the part that data can be used as database is stored in calculating as described below
It is on memory in system or associated with computing system as described below.
Fig. 1 shows the schematic block diagram of the system 100 according to the embodiment that can be used for assessment performance data.With reference to
Fig. 1, system 100 includes processor 102, the operation of 102 control system 100 of processor, and the processor 102 can
Implement method described herein.
System 100 further includes memory 106, and the memory 106 includes the director data for indicating the set of instruction.Storage
Device 106 can be configured as storage and can be executed by processor 102 to execute the program code of method described herein
The director data of form.In some embodiments, described instruction data can include being both configured to execute either to be used for
Execute the individual step of method described herein or multiple software and/or hardware modules of multiple steps.In some realities
Apply in example, memory 106 can be further include system 100 one or more other components (for example, the processor of system 100
102 and/or one or more other component) equipment part.In an alternative embodiment, memory 106, which can be to be directed to, is
The part of the specific installation of the other component of system 100.
In some embodiments, memory 106 may include multiple sub memories, first each sub memory can store
Director data.In some embodiments that memory 106 includes multiple sub memories, the instruction number of the set of described instruction is indicated
According to can be stored at single sub memory.In the other embodiments that memory 106 includes multiple sub memories, institute is indicated
The director data for stating the set of instruction can be stored at multiple sub memories.For example, at least one sub memory can deposit
Storage indicates the director data of at least one instruction in the set of described instruction, and at least one other sub memory can store
Indicate the director data of at least one other instructions in the set of described instruction.Therefore, according to some embodiments, indicate different
The described instruction data of instruction can be stored at one or more different locations in system 100.In some embodiments,
Memory 106 can be used to store by system 100 processor 102 acquisition or make or from any of system 100
Information, data (for example, image), signal and the measurement result of other component.
The processor 102 of system 100 can be configured as being communicated with memory 106 to execute the set of described instruction.Institute
The set for stating instruction can make processor 102 execute method described herein when being executed by processor 102.Processor
102 can include be configured as or be programmed to by retouch herein it is described in a manner of come one or more of control system 100
A processor, processing unit, multi core processor and/or module.In some embodiments, for example, processor 102 may include
It is configured for multiple (for example, interoperability) processor, processing unit, multi core processor and/or the modules of distributed treatment.
It will be appreciated by persons skilled in the art that such processor, processing unit, multi core processor and/or module can be positioned in
In different positions, and the different step of method described herein and/or the different portions of single step can be executed
Point.
Fig. 1 is again returned to, in some embodiments, system 100 may include at least one user interface 104.One
In a little embodiments, user interface 104 can be further include system 100 one or more other components (for example, system 100
Processor 102, memory 106 and/or one or more other component) equipment part.In an alternative embodiment, Yong Hujie
Mouth 104 can be the part of the specific installation of the other component for system 100.
User interface 104 can be used for the user for system 100 (for example, researcher (such as medical research person), medicine are special
Industry personnel or any other user of neural network model) it provides by the letter that is obtained according to the method for embodiment herein
Breath.The set of described instruction can make processor 102 control one or more user interfaces 104 when being executed by processor 102
To provide the information by obtaining according to the method for embodiment herein.Alternatively or additionally, user interface 104 can be matched
It is set to and receives user's input.In other words, the user that user interface 104 can permit system 100 manually input instruction, data or
Information.When being executed by processor 102 the acquisition of processor 102 can connect the set of described instruction from one or more users
User's input of mouth 104.
User interface 104 may be such that can draw (or output or display) information, data for the user of system 100
Or any user interface of signal.Alternatively or additionally, user interface 104 may be such that the user of system 100 can mention
For user's input, interact with system 100 and/or any user interface of control system 100.For example, user interface 104 can wrap
Include one or more switches, one or more buttons, keypad, keyboard, mouse, mouse roller, (for example, tablet computer or intelligence
Can be on mobile phone) touch screen or application, display screen, graphical user interface (GUI) or other visions draw component, one or more
Loudspeaker, one or more microphones or any other acoustic component, one or more lamps, for provide touch feedback (for example,
Vibrating function) component or any other user interface or user interface combination.
In some embodiments, illustrated such as in Fig. 1, system 100 can also include for enabling system 100
The communication interface (or circuit) 108 communicated with the interface of the part as system 100, memory and/or equipment.Communication interface
108 can communicate wirelessly or via wired connection with interface, memory and equipment.
It will be realized that Fig. 1 illustrate only this aspect of the diagram disclosure needed for component, and in actual embodiment
In, system 100 may include the additional components other than those of shown component.For example, system 100 may include being used for
The battery or other power supplys powered for system 100 or the device for system 100 to be connected to main power source.
In more detail, as mentioned above, system 100 is configured for assessment performance data.Memory 106 includes table
Show the director data of the set of instruction.Processor 102 is configured as that the set of described instruction is communicated and executed with memory.Letter
The set of Yan Zhi, described instruction make processor 102 when the execution of processor 102 by system 100: acquisition is related to performance indicators
Multiple performance data records of connection；Classified according to the performance indicators to each performance data record；Identify the performance
Multiple variables of the classification are contributed in data record；It is determined based on the multiple variable and performance indicators and described
The relevant multiple observation results of at least one variable in multiple variables；And the multiple observation result is delivered for being presented to
User.The observation result be considered " it was found that " and can referred to as " it was found that ".For example, observation result (or discovery)
It is considered the result based on the analysis to performance data.
The performance data record can be related to any kind of activity or performance of task previously executed.With this side
Formula, the performance data record may be considered that including history or passing data.Each performance data record can for example be related to
Single-unit activity or task and/or it is related to special entity.For example, each performance data record can relate in health care environment
And movable or task performance relevant to particular patient.Fig. 2 shows the multiple performance data records being disposed in table 200
Example.In this example, every row in table 200 is related to the process for the particular patient for being numbered as 1 to N.In table 200
Each column includes and is related to performed process and/or be related to and executes the associated number of the particular variables of patient of process about it
According to.Column 202 indicate the gender of patient (' M ' is used for male, and ' F ' is used for women)；Whether column 204 indicate patient by certain medical
Situation, in this case, the specific medical conditions be diabetes (if patient really by the medical condition, for
' Y ', if it is not, then being ' N ')；The age of the instruction patient of column 206；Process relevant to patient is responsible for or is executed in the instruction of column 208
Consulting staff cognizance code or number；Column 210 indicate the waiting time of patient (that is, being placed on from patient with process
Waiting list on duration on date for being performed to the process on date for waiting).In other examples, the waiting
Time can be variously defined.
It will be realized that may will include than more data records shown in table 200.In some examples, data
Library may include hundreds of or thousands of performance data records.For example, database may include being related to and entire hospital or healthy group
The record of the relevant performed all medicals of the patient knitted.Furthermore, it will be appreciated that the amount of available data is depended on,
Each data record may include more or fewer data (that is, table 200 may include more or fewer column).For example, number
According to library may include the instruction of the type of performed process, the instruction of the priority of the process (for example, the process whether
Optionally executed or patient be that the process is awaited orders), the instruction whether disposed privately of patient is (for example, at as payment
The part set), the instruction of medical facilities that is disposed of patient etc..
Each performance data record is associated with one or more performance indicators.In example shown in Fig. 2, the table
Existing data record is associated with the performance indicators for involving waiting for the time.Specifically, the performance indicators involve ensuring that patient etc.
Wait be longer than the target or expectation of the time (for example, six weeks) for the definition being directed in the waiting list of process.If process is defining
Period in be performed, then the specific condition can be considered having met or having met the performance indicators.However, if stream
Cheng Wei is interior at the defined time period to be performed, then the specific condition is not considered having met or having met the performance indicators.Such as
Mentioned above, performance indicators can be referred to as crucial performance indicators (KPI), and these indexs can for example be used to comment
Estimating special entity, company or tissue, to meet one group of guide how well.
In some examples, performance data record can be associated with multiple performance indicators, so that the data record can
Performance is assessed be used to compare multiple performance indicators.For example, data record can be showed with involve waiting for the time first
Index and be related to process has how successful second performance indicators associated (such as in example above in the case where disposing specific medical conditions
In).In this example, the data record can be used to assess the success of waiting time and/or process.
Once the performance data record has been collected, described instruction just makes the processor 102 of system 100 according to described
Performance indicators classify to each performance data record.The classification of each performance data record can be and classified based on two-value
Mechanism, thus if meeting specific criteria (or one group of standard), data record is classified as meet the performance
Index, and if being unsatisfactory for specific criteria (or one group of standard), it is classified as not meet the performance indicators.It is above
In the example discussed with reference to Fig. 2, whether each data record (1 to N) can meet the requirement about the waiting time according to it
To classify.For example, if the waiting time of greatest expected is six weeks (that is, 42 days), in data shown in Fig. 2
In, data record 1,2, N-1 and N may be considered that and met the waiting time standard, because its waiting time is both less than
42 days, and data record 3 and 4 may be considered that and not meet waiting time standard, because its waiting time is longer than 42 days.
In some embodiments, performance indicators associated with each performance data record can be transformed into classification type
Or rule.Therefore, in such embodiments, the set of the described instruction in memory 106 is stored in when by processor 102
The processor can also be made to generate the classification standard for being directed to the performance indicators when execution.As above, the classification standard can
To be two-value classification, such as according to whether meet the standard to classify.For example, the classification standard may include about
Whether the instruction of the performance indicators is met.In some embodiments, the performance indicators can be transformed into classification standard, make
Obtaining data record can be according to whether there are problems that classifying about the classification standard (that is, problem or worry).Example
Such as, it if data record meets classification standard (for example, the waiting time is less than the duration defined), can be classified
The problem of having ' N ', being not present with instruction about the specific criteria；However, if the data record is unsatisfactory for classification standard (example
Such as, the waiting time is longer than the duration of definition), then it can be classified ' Y ', exist with instruction about the specific criteria
The problem of.
For example by medical professional each record can will be distributed to the classification of data record or by responsible
Consulting staff is manually done after process, such as rule-based automated procedure can be used and be automatically completed.?
In some embodiments, classifying rules can automatically be speculated according to data.For example, such as information gain cut-off and/or cluster
Technology may be used to determine whether each data record meets classification standard.In one example, passing performance data
Record is stored with particular order (for example, with sequence of the waiting time of rising), and mean shift clustering method is applied
To find the turning point in data.Recording those of on turning point will be classified as be unsatisfactory for the contingency table for performance indicators
Standard, and record will be classified as meet the classification standard those of under turning point.In some embodiments, contingency table
Standard, such as maximum latency can be determined based on the requirement illustrated in one group of guide.In some embodiments, machine
Device learning art can be used to automatically determine classification standard or rule.
For performance indicators come determine classification standard embodiment in, make the processor to each performance data record
Carrying out classification may include that the processor is made to be classified according to the classification standard to each performance data record.
In some embodiments, update can be provided for user or change performance indicators or differently described point of definition
The option of class standard.For example, user can update institute in the case where the classification standard requires process to be performed in six weeks
Standard is stated to require process to be performed within the different periods (such as surrounding).
The instruction being stored in memory 106 also makes the processor identification performance number when being executed by processor 102
According to the multiple variables for contributing to classification in record.In the performance data record, there may be the multiple changes for contributing to classification
Amount (such as, if meet the performance indicators).In the example that reference Fig. 2 is discussed, whether specific process is being wanted
Any variable being included in data record can be depended on by being performed in the waiting time asked.For example, being directed to specific condition
Be not able to satisfy the target latency time can with the gender of patient, with patient whether by specific medical conditions, with the age of patient,
To the consulting staff for being responsible for the process or related with any other factor or variable that can be included in data record
Connection.Some variables, the gender of such as patient, it is impossible to have strong tribute to whether the process is completed within the target latency time
Offer influence.However, its dependent variable, is such as responsible for the consulting staff of the process, relatively strong influence can have.Such as institute above
It mentions, hundreds of or thousands of (or even more) a data records can be collected and be evaluated, and each data are remembered
Record may include tens of or hundreds of data fields, wherein each data field in the data field may be constructed about
Whether the variable or contribution factor of classification standard are met.In addition, if there are other contribution factors, then some contribution factors can be with
More strongly contribute.For example, specific consulting staff can contribute to the longer waiting time, but only when the certain types of stream of execution
Cheng Shi, or only when executing process in certain medical facility.It is thus impossible to which enough manually interpret all data and establishment
Which variable most contributes to specific classification by force.
In some embodiments, processor 102 is made to identify that the multiple variable may include that come from the processor will
The data of multiple performance data records are supplied to one or more prediction models.Various prediction models can be used, to identify
Show the contribution variable in data record.In some embodiments, can be used machine learning model, such as neural network and/
Or deep learning network.In some embodiments, identify that the multiple variable may include using one or more mode discoveries
Model, decision tree, rule generate model, and/or maximum information gain model.The performance data are recorded in it and have been classified
It can be fed in one or more prediction models later to be trained to the model.In this way, the mould
Which variable is type can learn or which kind of combination of variable most contributes to specific classification by force.For example, the model can learn
Which factor most contributes to satisfaction (or being unsatisfactory for) waiting time target by force.
Various data processings and data analysis technique can be used to analyze the output of the data from various prediction models, with
Just determine which prediction model identifies the variable most contributed by force.Including ' area below recipient's characteristic working curve '
(AUC) prediction performance measurement can be used to assess the prediction model.AUC measurement instruction covers more true targets and introduces
Tradeoff between more vacation candidates.Cross-Validation technique can be used for the measurement of more various prediction models.Specifically, exist
In cross validation, partial data is taken out, and the rest part of data be used to construct prediction model.Constructed model is then
It is used to compare taken out data and carried out prediction.The prediction is compared with the actual value of the performance indicators, in terms of
Calculate the prediction performance measurement.
According to the analysis of prediction model and the output of prediction model, which prediction model optimum prediction can be provided about
Performance measurement is determined (according to the variable for contributing to classification).It is then possible to select to be confirmed as providing the pre- of optimal representation
It surveys model, and may be used to determine whether as discussed below associated with the data one by the variable that the prediction model determines
A or multiple observation results.
It may include some changes for contributing to carried out classification by force by multiple variables that optimum prediction model identifies
Amount and some variables for not contributing to carried out classification by force.Therefore, in some embodiments, can execute to variable
Further analysis so that variable is included in short-list, and identify which variable is most related.It is described more in some examples
Each variable in a variable can be compared with each of the multiple variable its dependent variable (for example, changing on it
Generation).
By the analysis or filtering according to one or more measurement, it can quantitatively reduce the multiple variable.Due to
Comparing each variable and its each dependent variable may be large-scale and data-intensive task, therefore in some embodiments
In, the system can reduce comparison task by limiting the quantity of compared variable.Can about when and its dependent variable
Which variable may provide most useful information and be determined when comparing.Such determination for example can be by analyzing previous user
Behavior, passing analysis record are (that is, result of the similar data analysis technique previously executed) and/or relevant to performance indicators
Document (for example, guide) Lai Jinhang.Such as when analyzing the waiting time, prediction model can indicate the consultation of doctors for being responsible for the process
Doctor is about whether the strong contribution factor for meeting performance indicators.Divide according to passing KPI data relevant to the waiting time
Analysis, the system can speculate that the period of season and the moon are especially relevant when considering the waiting time of consulting staff.Institute
The system of stating can also speculate that the type for the process to be performed is the notable contribution factor to the waiting time.When it is to other
It is continually investigated then, such as this can be speculated in the analysis of KPI.Therefore, in this example, the system can be with
For every consulting staff in terms of being broken down into its serving-time distribution of season and the moon and performed process class
Data are analyzed in terms of type.
In one embodiment, all text elements of variable associated with performance indicators and/or guide can root
Recorded according to passing performance data record analysis of history, and maximally related variable can be identified as relative to
The variable that its dependent variable is further analyzed.In other embodiments, all unique values of variable can be transformed into item I1,
I2, I3 ... IN }, wherein performance indicators continuous item (such as classification standard) is subset { I ' 1, I ' 2, I ' 3 ... the I ' of the item
K }, wherein k < N.Frequent item set mining can be used to identify the item frequently occurred with performance indicators continuous item.In other realities
It applies in example, " word2vec " method can be used for the user for identifying that expression is registered as performance indicators continuous item and background item
The term vector of the distributional semantic model of behavior and the term vector of other performance indicators parsing.It shows close with performance indicators term vector
The activity and/or behavior of similitude can be identified as the variable for further analyzing.
By analyzing passing performance data record, the performance trend with the time can be identified, allow user's energy
Whether enough activities of rapidly saying and the performance of task are totally improving or are deteriorating.Such understanding can permit about in table
The particular variables identified in existing data record are made decision, such as so that the performance of difference is enhanced.
In example above, the determination for the variable that be compared with one another and be analyzed is by being made by the system
Decision automatically carry out.However, in some embodiments, user can choose the change that be compared with one another and be analyzed
Amount.For example, in some embodiments, user be considered that the age of patient may contribute to by force for execute process etc.
To the time, and therefore, before its dependent variable is put into short-list and is analyzed, user can choose patient age work
For strobe utility.
It is intended to reduce the size for the data set to be analyzed to the analysis that the multiple variable executes, so that the analysis can
To be performed faster, and to require less process resource.This data analysis can be referred to as " alternate analysis ", by
This big data set " being sliced " and " dicing ", to assess particular variables relative to its specific dependent variable, but are not necessarily to
Consider all variables in data set.
In some embodiments, certain variables may seem to contribute to by force specific classification (for example, particular variables can be seemingly
Lead to the longer waiting time), but there may be why these variables are contributed by force and these variables are inevitable
The reason of.For example, specific process can the process be need to performed before one month period patient view's result.?
Under such circumstances, process can be requested, particular patient is made to be added to waiting list, but since patient must be observed knot
Fruit one month, necessarily it can be extended one month for the waiting time of the process.In this case, wherein tribute
The factor of offering be inevitable and be it is known, can be ignored or for the contribution factor of variable by from analysis
It removes.
In some embodiments, multiple performance indicators can be interested, or can be relevant.In such model
Example in, the system can provide multiresolution analysis, wherein variable associated with a performance indicators can relative to
Another associated variable of performance indicators is analyzed.
Therefore, it is however generally that, the instruction being stored in memory can make the processor when being executed by a processor
The multiple variable is filtered according at least one measurement.In some embodiments, at least one measurement may include using
The measurement that family defines.As mentioned above, the big of the data set to be analyzed will be reduced by being filtered to the multiple variable
Small, thus reduction analyzes the data and determines time and the processing capacity of the observation result.The multiple variable can be with
The one or more measurements being filtered according to it automatically or by user are manually selected by the system.As solved above
It releases, the measurement can be based on its dependent variable, and such as, such as whether the process is optional process, about its execution
The age of the patient of the process, the hospital for executing the process or facility etc..Variable is filtered can make user from
Removed in data set it is incoherent, useless in assessment, and/or be considered it is trifling (for example, in classification be not it is important because
Element) data.
Filtering can be used to reduce the size for the data set to be analyzed, to include and be considered most contributing to point by force
The relevant data of the variable of class.In some embodiments, be filtered to variable may include using baseline or threshold value；It drops to
Variable can be ignored from analysis or omits those of under baseline or threshold value, and variable can those of on baseline or threshold value
To be included in analysis and further be considered.In some examples, the baseline or threshold value can by user setting, and
And for example may include statistically significant p- value or the variable to be considered number of thresholds.
Once performing analysis to the multiple variable, the variable just for example can be in the contribution made to classification
Sequence in terms of classify.For example, be confirmed as most by force contribute to classification those of variable can be ranked as height
In be confirmed as on classification almost without contribution influence those of variable.The variable of highest level is (for example, the first 10 most contribute
Variable) it is used as from the subset of the variable of result from its determination.In some embodiments, the subset of variable can be with one
A little other modes (for example, using baseline as discussed above or threshold value) determine.
The instruction being stored in memory make when being executed by a processor the processor based on the multiple variable come
Determine multiple observations relevant at least one variable in the performance indicators and the multiple variable as mentioned above
As a result.In some examples, the observation result can the subset based on the variable for using method as described above to establish come
It determines.Observation result discussed herein is considered following opinion, because the observation result provides pair for user
The slave initial data for the data being analyzed is to user's not obvious opinion immediately.
Observation result is considered the summary to the data analyzed, and allows users to infer about being considered
Data the most important and related fact.For example, in the case where discussing herein, wherein analysis is directed to various meetings
Examine the waiting time of doctor, wherein the waiting time is broken down into week and season, and observation result is confirmed as and specific consulting staff
C03 is related.According to an example, the system can determine the observation of " C03: 6.1,6.4,7.2 weeks average to Q2, Q3, Q4 "
As a result.In other words, it is determined during season Q2 based on the data analyzed for the specific consulting staff with identifier C03
Waiting time be average 6.1 weeks, the waiting time during season Q3 be averaged 6.4 weeks, and during season Q4 etc.
It is 7.2 weeks average to the time.In another example, wherein for every consulting staff waiting time about process whether be
Option program (that is, patient chooses with the process, rather than the process is ordered by medical professional) Lai Jinhang is commented
Estimate, system can determine the sight of " C15 average 6.2 weeks in optional priority, other are in the range from 3.2 weeks to 5.8 week "
Examine result.In other words, for the specific consulting staff with identifier C15, determine that the waiting time for optional process is flat
Equal 6.2 weeks, and for other (non-optional) programs, the waiting time is averagely between 3.2 weeks and 5.8 weeks.Although only give herein
Gone out the observation of two examples as a result, it will be appreciated that, the system can be determined based on many variables it is many observe as a result,
And it is decomposed with being permitted various ways.
In some embodiments, the instruction in the memory can make the processor when being executed by the processor
It classifies before delivering the multiple observation result for presentation to the multiple observation result.The observation result can be by
Graduation, or the subset of observation result can be identified or be selected by considering the statistical significance of the observation result.Cause
This, making the processor may include that the processor is made to calculate each observation result to the graduation of the multiple observation result
Statistical significance.In some embodiments, the set of described instruction can make the processor when being executed by the processor
It is classified according to statistical significance calculated to the multiple observation result.For example, being considered to have relatively higher statistics
The observation result of conspicuousness can be ranked as higher than the relatively lower observation result of its statistical significance.In some examples
In, p- value (that is, probability value) can be used to measure in the statistical significance for observing result, and thus lower p- value instruction is bigger
Statistical significance.In some embodiments, the statistical significance (for example, p- value) can together with the observation result by
It is presented to the user.
It in some embodiments, may include the scoring system according to definition to the multiple graduate process of result of observing
System scores to the embodiment.Maximally related observation knot is considered with those of highest score observation result
Fruit, and therefore, it can be ranked in the list of the multiple observation result higher.The observation result of these greater degrees
It is considered " primarily observing result " or " primary opinion ".In some embodiments, it is ranked in baseline or threshold value water
Observation result or scoring are less than those of threshold quantity observation result and can be saved in any other analysis those of under flat
Slightly or it is ignored.
In this way, the system can be considered dividing priority to observation result, and in position more outstanding
In presented (that is, higher in graduate list) for user there is those of higher priority to observe result.Higher priority
Observation result be given more weights because these observation results can be used by a user in carry out it is relatively small change so as to
Improve activity relevant to performance indicators or the aggregate performance of task.For example, if result is apparent from high priority
Be waiting time of specific consulting staff gradually to become longer within recent months, then it can indicate consulting staff
Workload it is too big.This may not be from initial data immediately it will be apparent that particularly in view of that can have to the waiting time
The every other variable influenced.By providing this analysis for user, user can take any necessary operation to change
Kind waiting time performance, such as the workload by reducing consulting staff.Family is not used only and more readily identifies any performance for this
Related problem, and enable other potential problems processed (for example, the workload energy of consulting staff in early stage
It is enough to be contracted by before it becomes too big).In addition, by the quantity of reduction variable (for example, by filtering, dividing priority
And graduation), it to be contracted by by the data volume that processor is analyzed, thus reduce processing load and calculate the time.
In other embodiments, other technologies can be used to classify in the observation result.For example, the system can be with
For the error of the classification of each variable analysis data record in variable relevant to performance indicators.Based on minimum quantity
Those of variable observation result of missing classification can be used as preferred (greater degree) observation result and be put into candidate name
It is single.
In other embodiments, it can analyze and have been classified as being unsatisfactory for specific mark associated with the performance indicators
The subset (for example, having been classified as that there is those of " problem " to show data record) of quasi- performance data record, and can
With counting accuracy and sensitivity.F- score (that is, harmomic mean of accuracy and sensitivity) or g- mean value (that is, sensitivity and
The geometric mean of specificity) it can be calculated for each performance data record in subset, and there is highest f- score
The record (for example, those of f- score or g- mean value on threshold value with definition data record) of (or g- mean value) can be by
It is included in short-list.It can be used as preferred observation result quilt with those the associated observation results that record for being included in short-list
It is included in short-list.
Although the observation result that is generated by the system facilitates summary for the relevant portion that user provides data,
The data still can be with numeric form or be some formats to be difficult to interpret.Therefore, according to some embodiments,
The system can be converted into the observation result that may be easier to the alternate formats of user's digestion.In some embodiments,
Instruction in the memory can be such that the processor generates in the multiple observation result of summary when being executed by a processor
At least one observation result.As used in this article, term " summary " is intended to mean the summary or remittance of information or data
Collection.For example, the summary may include explaining sentence, phrase or the wording of the observation result.It is observed for example above
As a result, example summary can be generated, display is " when C03 has waiting deteriorate and higher within nearest three season
Between ".Another example summary can show " C15 has the significant longer waiting time in optional process ".Therefore, substitution is simple
Ground generate require user's unscrambling data observation as a result, the system can be provided for user understanding get up it is simple and easy
It summarizes.
In some embodiments, so that the processor is generated summary may include making the processor: by predefined template
In at least one predefined template be applied at least one observation result in data；And natural language processing algorithm is answered
For the data at least one observation result.Therefore, in some examples, the information being included in the observation result can
To be incorporated into one or more predefined summary templates.For example, memory can store multiple template, and the system
It can choose the one or more correlate templates to use during generating summary.In other examples, at natural language
Reason can be used for: analysis is included in the information in the observation result, and generation can be used as the suitable of summary
Wording.
So that the processor is delivered the multiple observation result for being presented to the user may include by the multiple observation
As the result is shown on display associated with the system.For example, the use of system 100 can be used in the multiple observation result
Family interface 104 is displayed to user.In some embodiments, the one or more summaries generated by the system can also or
It is alternatively shown over the display.
In some embodiments, described instruction can also be such that the processor generates the multiple when being executed by a processor
Observe the graphical representation of at least one of result observation result.Described instruction can also make the processor deliver the figure
It indicates for presenting.For example, the expression can be together with observation result and/or summary or substitution observation result and/or general
It is to be displayed over the display.Any graphical representation can be generated, such as chart, curve graph, attached drawing, schematic diagram, image
Deng.Fig. 3 is the example that the display 300 that system 100 herein disclosed is presented to user can be used.300 packet of display
Include the information title 302 relevant to what explained in display.Below title 302, display 300 include with according to performance data
Record determining relevant two summaries 304,306 of observation result and can be via the link 308 of the other information of acquisition.
Display 300 further includes two graphical representations 310,312 relevant to one or more of summary 304,306.In this example,
Graphical representation 310,312 be show it is related to the percentage of performance data record of the problem that contributes to (meeting specific classification standard)
Data chart.
According to some embodiments, user can be able to be for example in by selecting or clicking on item or hovering over pointer
It is interacted above existing item with the item (for example, observation result, summary and/or graphical representation) presented.Such selection can be with
Additional information is set to be provided to user.For example, the observation subsequent reasoning of result can be provided to user.In some embodiments
In, analysis mechanisms and/or filtering and the various steps for being included in short-list can be presented to user.In this way, described
User can be able to access that the necessary data that be used to generate the observation result for being presented to them.This can enable users can
With the deeper understanding why being generated to the observation result.
A kind of method for assessing performance data is disclosed according to another aspect,.Fig. 4 is to show data for assessing
The flow chart of the example of method 400.At step 402, method 400 includes acquiring multiple performance numbers associated with performance indicators
According to record.At step 404, method 400 includes being classified according to the performance indicators to each performance data record.Method
400 include, and at step 406, multiple variables of the classification are contributed in identification performance data record.At step 408, side
Method 400 includes being determined based on the multiple variable and at least one variable phase in the performance indicators and the multiple variable
The multiple observation results closed.At step 410, method 400 includes delivering the multiple observation result for being presented to the user.?
In some examples, (step 402) may include that acquisition has at least one variable associated with performance indicators the step of acquisition
Multiple performance data records.
It will be realized that method 400 can be executed by the processor 102 of system described above 100.Therefore, method
400 are considered the method implemented by computer.
Fig. 5 is the flow chart for the other example for assessing the method 500 of performance data.On method 500 may include
The one or more steps for the method 400 that text is discussed.Method 500 can also include, and at step 502, generate described in summarizing
The summary of at least one of multiple observation results observation result.At step 504, method 500 can also include being in user
The existing summary and graphical representation corresponding with the observation result of summary presented.As mentioned above, the sight
The graphical representation for examining result for example can be in the form of chart or curve graph.
Fig. 6 is the flow chart for the other example for assessing the method 600 of performance data.Method 600 describe including
Such as it can be executed by processor 102 and the specific embodiment of step corresponding with the instruction discussed above with reference to Fig. 1
Example.Method 600 includes the steps that method 400 402 is to step 410 as discussed above.Acquiring multiple performance data notes
After the step of record (step 402), method 600 may include, and at step 602, generate the contingency table for being directed to the performance indicators
It is quasi-.Method 600 then continues to step 404, wherein each performance data record is classified according to the performance indicators.
In some embodiments, each performance data record can classify according to the classification standard generated in step 602.
According to method 600, multiple variable (steps that the classification is contributed in the performance data record are then identified
406).Method 600 is then with step 604 continuation, wherein the multiple variable is filtered according at least one measurement.Institute
Stating at least one measurement may include user-defined measurement.After filtering (step 604), method 600 is continued with step 408,
Wherein it is determined that multiple observation results relevant at least one variable in the performance indicators and the multiple variable.In step
At rapid 606, method 600 classifies to the multiple observation result.At step 608, method 600, which generates, summarizes the multiple sight
Examine the summary of at least one of result observation result.Method 600 is then with step 610 continuation, wherein generates the multiple sight
Examine the graphical representation of at least one of result.Method 600 is then with step 410 continuation, wherein the multiple observation knot of delivering
Fruit is for being presented to the user.At step 612, method 600 further includes delivering the graphical representation for being presented to the user.
According to other aspect, a kind of computer program product is disclosed.The computer program product includes non-transient
Computer-readable medium 702, the computer-readable medium is embedded with computer-readable code wherein, described computer-readable
Code, which is configured such that when being executed by suitable computer or processor 704, executes the computer or processor
Any means in method disclosed herein.
Processor 102,704 can include being configured as or being programmed to control in mode described herein being
One or more processors, processing unit, multi core processor or the module of system 100.In certain embodiments, processor 102,
704 can include being both configured to execute either for executing the individual step or multiple steps of method described herein
Rapid multiple software and/or hardware modules.
Term " module " as used in this article is intended to include that hardware component (is such as configured as executing specific function
Processor or processor component) or software component (such as when being executed by a processor with specific function one group of instruction
Data).
It will be realized that the embodiment of the present invention is also applied to computer program, especially on carrier or in the carrier
Computer program, the computer program are suitable for putting the invention into practice.Described program can be with source code, target generation
In the form of code, code intermediate source and object code such as in the form of partial compilation, or to be suitable for carrying out according to this
Any other form of the method for the embodiment of invention.It will additionally appreciate, such program can have many different frameworks
Design.For example, implementing according to the method for the present invention or the program code of the function of system can be subdivided into one or more sons
Routine.Many different modes of distributed function will be apparent to those skilled in the art between these subroutines
's.The subroutine can be collectively stored in an executable file, to form self-contained program.Such executable text
Part may include computer executable instructions, for example, processor instruction and/or plug-in reader instruction are (for example, Java plug-in reader refers to
It enables).Alternatively, one or more subroutines or all subroutines can be stored at least one external library file, and
Either statically or dynamically (for example, at runtime) linked with main program.Main program includes at least the one of at least one subroutine
A calling.Subroutine can also include to mutual function call.The embodiment for being related to computer program product includes computer
Executable instruction corresponds to each processing stage of at least one method proposed herein.These instructions can be by
It is subdivided into subroutine and/or is stored in the one or more files that can be either statically or dynamically linked.It is related to computer
Another embodiment of program product includes computer executable instructions, is corresponded in system and/or product presented herein
The each unit of at least one.These instruction can be subdivided into subroutine and/or be stored in can by statically or dynamic
In one or more files of ground link.
The carrier of computer program can be any entity or equipment that can carry program.For example, carrier may include
Storage medium, such as ROM (for example, CDROM or semiconductor ROM) or magnetic recording media (for example, hard disk).In addition, carrier can
To be such as electrically or optically transmittable carrier of signal, via cable or optical cable or radio or other works can be passed through
Tool is to convey.When program is embodied in this signal, carrier can be made of such cable or other equipment or unit.It is standby
Selection of land, carrier can be program and be embodied in integrated circuit therein, and integrated circuit is adapted for carrying out correlation technique, or is used for phase
The execution of pass method.
Those skilled in the art are practicing claimed invention by research attached drawing, disclosure and claim
When can understand and realize other variants of the disclosed embodiments.In the claims, one word of " comprising " is not excluded for other
Element or step, and word "a" or "an" be not excluded for it is multiple.Single processor or other units may be implemented in right
Several functions of being recorded in it is required that.Although certain measures are described in mutually different dependent claims, this
The combination that these measures cannot be used to advantage is not indicated that.Computer program can be stored/distributed on suitable medium,
Such as the optical storage medium or solid state medium of the part supply together with other hardware or as other hardware, but can also be with
It is distributed otherwise, such as via internet or other wired or wireless telecommunication systems.It is any attached in claim
Icon note shall not be interpreted the limitation to range.
Claims (15)
1. a kind of system (100) for being configured for assessment performance data, the system comprises:
Memory (106) comprising indicate the director data of the set of instruction；And
Processor (102) is configured as that the set of described instruction is communicated and executed with the memory, wherein the finger
The set of order makes the processor when being executed by the processor:
Acquire multiple performance data records associated with performance indicators；
Classified according to the performance indicators to each performance data record；
Identify multiple variables that the classification is contributed in the performance data record；
It is determined based on the multiple variable relevant at least one variable in the performance indicators and the multiple variable
Multiple observation results；And
The multiple observation result is delivered for being presented to the user.
2. system according to claim 1 (100), wherein the set of described instruction is executed when by the processor (102)
When also make the processor:
It classifies before delivering the multiple observation result for presentation to the multiple observation result.
3. system (100) according to claim 2, wherein make the processor (102) to the multiple observation result point
Grade includes the statistical significance for making the processor calculate each observation result；And
Wherein, the set of described instruction makes the processor when being executed by the processor:
It is classified according to statistical significance calculated to the multiple observation result.
4. system (100) according to any one of the preceding claims, wherein the set of described instruction is when by described
Reason device (102) also makes the processor when executing:
Generate the summary for summarizing at least one of the multiple observation result observation result.
5. system (100) according to claim 4, wherein so that the processor (102) is generated summary includes making the place
Reason device carries out at least one of the following: predefined template is applied to the data at least one described observation result；And
Natural language processing algorithm is applied to the data at least one described observation result.
6. system (100) according to any one of the preceding claims, wherein the processor (102) is made to identify institute
Stating multiple variables includes making the processor that will be supplied to one or more in advance from the data of the multiple performance data record
Survey model.
7. system (100) according to any one of the preceding claims, wherein the set of described instruction is when by described
Reason device (102) also makes the processor when executing:
Generate the classification standard for being directed to the performance indicators；
Wherein, so that the processor is carried out classification to each performance data record includes making the processor according to the contingency table
Standard classifies to each performance data record.
8. system (100) according to claim 7, wherein the classification standard includes about whether meeting the performance
The instruction of index.
9. system (100) according to any one of the preceding claims, wherein the set of described instruction is when by described
Reason device (102) also makes the processor when executing:
The multiple variable is filtered according at least one measurement.
10. system (100) according to claim 9, wherein at least one described measurement includes user-defined measurement.
11. system (100) according to any one of the preceding claims, wherein the set of described instruction is when by described
Processor (102) also makes the processor when executing:
Generate the graphical representation of at least one of the multiple observation result observation result；And
The graphical representation is delivered for being presented to the user.
12. a kind of for assessing the method (400) of performance data, which comprises
Acquire (402) multiple performance data records associated with performance indicators；
Classified (404) to each performance data record according to the performance indicators；
Multiple variables of the classification are contributed in identification (406) the performance data record；
(408) and at least one variable phase in the performance indicators and the multiple variable are determined based on the multiple variable
The multiple observation results closed；And
(410) the multiple observation result is delivered for being presented to the user.
13. according to the method for claim 12 (400,500), further includes:
Generate the summary that (502) summarize at least one of the multiple observation result observation result.
14. according to the method for claim 13 (400,500), further includes:
The graphical representation of (504) described summary and the observation result corresponding with the summary presented is presented to user.
15. one kind includes the computer program product of non-transient computer-readable media (702), the computer-readable medium exists
It is wherein embedded with computer-readable code, the computer-readable code is configured such that by suitable computer or processing
Device (704) makes the computer or processor execute method described in any one of 3 and 14 according to claim 1 when executing.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2017116322 | 2017-12-15 | ||
| CNPCT/CN2017/116322 | 2017-12-15 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110060776A true CN110060776A (en) | 2019-07-26 |
Family
ID=67315862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811517790.5A Pending CN110060776A (en) | 2017-12-15 | 2018-12-12 | Assessment performance data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110060776A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1684082A (en) * | 2005-02-18 | 2005-10-19 | 北京大学深圳医院 | Evaluation management system and its method for modern hospital performance |
| US20100174557A1 (en) * | 2009-01-05 | 2010-07-08 | Markus Bundschus | System and Method for Ranking Quality Improvement Factors in Patient Care |
| US20120065987A1 (en) * | 2010-09-09 | 2012-03-15 | Siemens Medical Solutions Usa, Inc. | Computer-Based Patient Management for Healthcare |
| CN103136629A (en) * | 2011-11-23 | 2013-06-05 | 通用电气公司 | Real-time contextual KPI-based autonomous alerting agent |
| CN106462664A (en) * | 2014-05-19 | 2017-02-22 | 皇家飞利浦有限公司 | Method and system for guiding patient self-care behaviors |
| WO2017072651A1 (en) * | 2015-10-30 | 2017-05-04 | Koninklijke Philips N.V. | Integrated healthcare performance assessment tool focused on an episode of care |
| WO2017191227A1 (en) * | 2016-05-04 | 2017-11-09 | Koninklijke Philips N.V. | Estimation and use of clinician assessment of patient acuity |
-
2018
- 2018-12-12 CN CN201811517790.5A patent/CN110060776A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1684082A (en) * | 2005-02-18 | 2005-10-19 | 北京大学深圳医院 | Evaluation management system and its method for modern hospital performance |
| US20100174557A1 (en) * | 2009-01-05 | 2010-07-08 | Markus Bundschus | System and Method for Ranking Quality Improvement Factors in Patient Care |
| US20120065987A1 (en) * | 2010-09-09 | 2012-03-15 | Siemens Medical Solutions Usa, Inc. | Computer-Based Patient Management for Healthcare |
| CN103136629A (en) * | 2011-11-23 | 2013-06-05 | 通用电气公司 | Real-time contextual KPI-based autonomous alerting agent |
| CN106462664A (en) * | 2014-05-19 | 2017-02-22 | 皇家飞利浦有限公司 | Method and system for guiding patient self-care behaviors |
| WO2017072651A1 (en) * | 2015-10-30 | 2017-05-04 | Koninklijke Philips N.V. | Integrated healthcare performance assessment tool focused on an episode of care |
| WO2017191227A1 (en) * | 2016-05-04 | 2017-11-09 | Koninklijke Philips N.V. | Estimation and use of clinician assessment of patient acuity |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Galetsi et al. | Big data analytics in health sector: Theoretical framework, techniques and prospects | |
| Hoyt et al. | IBM Watson analytics: automating visualization, descriptive, and predictive statistics | |
| US11257574B1 (en) | Information system providing explanation of models | |
| AU2012245343A1 (en) | Predictive modeling | |
| Mazorchuck et al. | Web-application development for tasks of prediction in medical domain | |
| US11450434B2 (en) | Implementation of machine-learning based query construction and pattern identification through visualization in user interfaces | |
| US20130282404A1 (en) | Integrated access to and interation with multiplicity of clinica data analytic modules | |
| JP2020529058A (en) | Human-participatory interactive model training | |
| King et al. | Using machine learning to selectively highlight patient information | |
| EP3669376A1 (en) | System and method for generating query suggestions reflective of groups | |
| JP2023544335A (en) | Adaptive training system and method for machine learning models | |
| Hayn et al. | Predictive analytics for data driven decision support in health and care | |
| US20220058749A1 (en) | Medical fraud, waste, and abuse analytics systems and methods | |
| Chen et al. | Process mining and data mining applications in the domain of chronic diseases: A systematic review | |
| de Moura et al. | Recognizing context-aware human sociability patterns using pervasive monitoring for supporting mental health professionals | |
| Zhao et al. | Visual analytics for health monitoring and risk management in CARRE | |
| US20230055277A1 (en) | Medical fraud, waste, and abuse analytics systems and methods using sensitivity analysis | |
| WO2019185853A1 (en) | Systems and methods for dynamic generation of structured quality indicators and management thereof | |
| Amirian et al. | Data science and analytics | |
| CN110060776A (en) | Assessment performance data | |
| Manuaba et al. | The evaluation of supervised classifier models to develop a machine learning API for predicting cardiovascular disease risk | |
| Bautista et al. | Health Disparities Through Generative AI Models: A Comparison Study Using a Domain Specific Large Language Model | |
| US11347753B2 (en) | Assessing performance data | |
| Gaur et al. | Early Prediction and Prevention of Lifestyle Diseases | |
| Jahangiri et al. | An Improved Naïve Bayes Approach to Diagnose Cardiovascular Disease: A Case Study |
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 |