CN109390058A - A kind of method for building up of case history Computer Aided Analysis System and the system - Google Patents
A kind of method for building up of case history Computer Aided Analysis System and the system Download PDFInfo
- Publication number
- CN109390058A CN109390058A CN201811144073.2A CN201811144073A CN109390058A CN 109390058 A CN109390058 A CN 109390058A CN 201811144073 A CN201811144073 A CN 201811144073A CN 109390058 A CN109390058 A CN 109390058A
- Authority
- CN
- China
- Prior art keywords
- case history
- module
- record information
- medical record
- medical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- 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
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present invention relates to a kind of case history Computer Aided Analysis Systems, comprising: medical record information module, for storing the medical record information of simultaneously managing patient；Base module comprising the mapping ruler of several medical knowledges and interrelated medical knowledge；Assistant analysis module, can knowledge based library module the medical record information in medical record information module is analyzed and is drawn a conclusion.Paramedical personnel's routine work and managerial role are played by above system, compared to existing assistant diagnosis system, it passes through the standardization to index, accurately analyze the standardized index for meeting natural language expressing, so that analysis conclusion is more acurrate, there is the case where mistaken diagnosis, fail to pinpoint a disease in diagnosis in reduction, and greatly reduces the time that medical worker reads case history and analysis, improves work efficiency.
Description
Technical field
The present invention designs medical assistant analysis field, the foundation side of especially a kind of case history Computer Aided Analysis System and the system
Method.
Background technique
Construction of country in terms of carrying forward vigorously medical treatment & health+big data+internet.The clinic of current main-stream is determined
Plan supports system to be defined as follows: being a kind of sufficiently with available, suitable computer technology, for semi-structured or non-knot
Structure medical problem, the system for improving clinical decision efficiency by man-machine interaction mode, clinical assistant diagnosis are clinical
One pith of DSS, major technique mode are that the technical side of auxiliary diagnosis is carried out based on document repository
Formula.Although Clinical Decision Support Systems brings many conveniences to medical institutions and staff, there is also many disadvantages.
The quantitative calculating of auxiliary diagnosis possibility is not supported, which is mainly foundation history information in document repository
Keyword search matching is the thoughtcast of kind of association type to export the technical approach of possible diagnosis, cannot quantitatively according to
A possibility that suffering from diagnosis according to history information calculating.
Since keyword search matching degree exists, difference semantically is easy to form to be failed to pinpoint a disease in diagnosis, and which is searched using keyword
The auxiliary diagnosis mode of rope matching association, if that from the matching in the collected keyword of history information and document repository
Word is easy to appear there may be difference and fails to pinpoint a disease in diagnosis phenomenon.For example the symptom of microscopic hematuria is described in medical history, come from diagnosis and treatment feature
Say that gross hematuria is also to meet its feature, but Keywords matching mode may match this knowledge base less than gross hematuria
Information, so as to cause failing to pinpoint a disease in diagnosis.
Matching can not be scanned for so as to cause failing to pinpoint a disease in diagnosis, in practical diagnosis and treatment process for stealthy symptom or possibility symptom
In often occur some cases be certain signs it is possible that the case where be not in, such as in " clinic diagnosis guide
(nephrology fascicle) " clinical manifestation in chronic renal failure is described in this way: " patient can without any symptom, or it is only out of strength,
The soreness of waist, enuresis nocturna increase etc. slight uncomfortable ", then this may result in failing to pinpoint a disease in diagnosis in support decision process for asymptomatic patient.
Keyword is the mode manually entered, may cause and fails to pinpoint a disease in diagnosis in the case where keyword inputs incomplete situation: existing
It is typically all to be manually entered when clinic diagnosis auxiliary system inputs keyword, these artificial refinements are inherently held with input process
It is easy the case where imperfect input or omission occur, it inputs imperfect just appear in document repository and searches for less than appropriate
Diagnosis, this will certainly will also occur failing to pinpoint a disease in diagnosis situation.
The prior art is manually entered keyword patterns, cannot change timely auxiliary diagnosis and treatment according to conditions of patients
Target: patient its medical record information and sign situation in treatment process be it is continually changing, this mode is that one kind is manually entered
Offline operating mode, it is very necessary in fact how timely auxiliary diagnosis to be changed according to conditions of patients and indication, exactly this
It is also a major defect.
Summary of the invention
In view of the above existing problems in the prior art, the present invention provides a kind of case history Computer Aided Analysis System and the systems
Method for building up.
A kind of case history Computer Aided Analysis System, comprising:
Medical record information module, for storing the medical record information of simultaneously managing patient；
Base module comprising the mapping ruler of several medical knowledges and interrelated medical knowledge；
Assistant analysis module, can knowledge based library module the medical record information in medical record information module is analyzed and is obtained
Conclusion out.
Further, base module includes: the document repository established based on medical literature data and based on clinical disease
Go through the clinical case history knowledge base of big data foundation；Medical literature knowledge base and clinical case history knowledge base respectively include the medicine
Knowledge and the mapping ruler.
Further, base module further include: the diagnosis established based on document repository and/or clinical case history knowledge base
Put knowledge base comprising the mapping ruler of illness information, symptom information, the illness information and illness information.
Further, clinical case history knowledge base included be medical knowledge Jing Guo standardization.
Further, medical record information module includes: medical record information storage unit and medical record information Standardisation Cell, wherein disease
Information standardization unit is gone through for analyzing the medical record information in medical record information storage unit, obtains the standard for meeting natural language
Index value.
Further, it is specifically included for the medical record information Standardisation Cell of standardization:
Case history project definition module, for newly established according to case history it is several being capable of identified case history project；
Extracting rule definition module, for setting the extraction conditions for extracting case history project from case history document；
Case history item recognition module, for extracting specific case history item from case history document according to the extraction conditions
Mesh content；
Standard index definition module: for setting the standard index to be standardized；
Normalisation rule definition module, for establishing the rule being standardized to the case history contents of a project；
Standardization unit, the normalisation rule for being established according to normalisation rule definition module is to extraction case history item
Mesh content is standardized, and obtains the standard index value for meeting natural language.
Further, assistant analysis module includes the analysis model for analyzing medical record information, same disaggregated model base
It is established in more than one classifiers.
Further, assistant analysis module further include: for training the model training unit of optimizing and analyzing model, training is single
Member can establish the mapping ruler between medical knowledge.
The invention also discloses a kind of methods for establishing case history Computer Aided Analysis System, and the specific method is as follows:
The medical record information module for storing patient is established,
Base module is established, base module includes the mapping rule of several medical knowledges and interrelated medical knowledge
Then；
Establish assistant analysis module, assistant analysis module can knowledge based library module to the case history in medical record information module
Information is analyzed and is drawn a conclusion
The invention has the following advantages:
The present invention is based on medical literatures and clinical case history big data to establish reliable knowledge base, since clinical setting is complicated,
The content that document is recorded cannot include whole the case where being likely to occur, thus be based on clinical case history knowledge base and greatly improve be
The accuracy of system analysis.The case history of patient is analyzed automatically by each analysis model, obtains related conclusions, for examining for doctor
Disconnected, therapeutic scheme planning etc. provides auxiliary.The present invention, can be greatly less by the standardization to medical record information
The calculation amount and fault rate of system, and this system does not need the keyword that user is manually entered in case history, simplifies operation, simultaneously
Prevent from omitting the input of key message.The medical record information of patient is more detailed, and the analysis conclusion obtained is also more accurate, therefore assists
Analysis module can obtain whole medical record informations of patient in time, and the accuracy of analysis conclusion is continuously improved.By to practical phase
The case where some medical knowledges closed are associated, and prevent omission, erroneous judgement.For there is some different descriptions in case history,
But the meaning of expression has very big relevance, even expresses meaning of the same race, in the prior art, will be because of different expression
Mode and omit or judged by accident these key messages, assistant analysis module of the invention can establish correlation and reflect for above-mentioned relation
It penetrates, so that the case where different expression ways is also considered in analysis, so that the accuracy of analysis result greatly improves.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is system block diagram of the invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, technical solution of the present invention will be carried out below
Detailed description.Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art are obtained all without making creative work
Other embodiment belongs to the range that the present invention is protected.
As shown in Figure 1, a kind of case history Computer Aided Analysis System, mainly includes following three parts on the whole:
(1) medical record information module, for storing the medical record information of patient.The module both can individually be built with analysis system
It is vertical, it can also be imported by electronic medical record system/management system of existing hospital.A large amount of disease produced by each patient assessment
Information capacity is gone through, comprising medical history record, doctor's advice, examines Examination report sheet, pathological replacement list, temperature chart, nursing record form etc., this
A little medical record informations embody development and the therapeutic process of conditions of patients symptom and sign, and each information controls auxiliary diagnosis and auxiliary
Treatment suffers from more or less meaning.Due to not perfect, the current overwhelming majority medical outcomes of existing intelligent analysis of medical record system
Or diagnosis and treatment are carried out by the way of manual read's analysis, this undoubtedly excessively relies on profile and the knowledge reserves of doctor, no
It only needs to take more time and energy, and accuracy rate is also difficult to ensure.Therefore of the invention by base module and auxiliary
It helps analysis module to realize through computer or other Intelligent hardwares to medical record information intellectual analysis and show that corresponding auxiliary is examined
Disconnected conclusion.
(2) base module comprising the mapping ruler of several medical knowledges and interrelated medical knowledge.Described knows
Know the R. concomitans that library relates to artificial intelligence field and database field, not all program with intelligence, which is owned by, to be known
Know library, only KBS Knowledge Based System just possesses knowledge base.Many application programs all utilize knowledge, wherein some has also achieved very
High level, still, the system that these application programs may be not based on knowledge, they also do not possess knowledge base.It is general
Difference between application program and KBS Knowledge Based System is, general application program be the knowledge of problem solving impliedly
It encodes in a program, and KBS Knowledge Based System then will solve knowledge and explicitly express the problem of application field, and individually group
At a relatively independent program entity.Knowledge in knowledge base has consistency of axioms and nonredundancy, deposits needed for knowledge base
It is huge to store capacity.Not individually storage also contains the mapping ruler with corresponding relationship, example to medical knowledge in knowledge base
It such as, can be by the corresponding symptom of a certain item disease, the cause of disease, body parameter, therapeutic scheme, therapeutic agent by mapping ruler
Etc. information establish mapping relations, these mapping rulers are also one of the key component that assistant analysis module is analyzed.
(3) assistant analysis module, can knowledge based library module the medical record information in medical record information module is analyzed
And it draws a conclusion.In some embodiments of the invention, assistant analysis module is established according to the base module, can
Enough medical record information is analyzed using knowledge base.
In some embodiments of the invention, base module includes: the Document Knowledge established based on medical literature data
Library and the clinical case history knowledge base established based on clinical case history big data；Document repository and clinical case history knowledge base respectively include
The medical knowledge and the mapping ruler.Above two knowledge base ensure that the usable correctness of medical knowledge and authority
Property, wherein the information of medicine rule-based knowledge base derives from authoritative documents and materials, " the medical diagnosis on disease mark issued such as the Ministry of Public Health
It is quasi- ", " national essential drugs formulary ", Chinese Medical Association " the clinic diagnosis guide " write.These documents and materials are with guide
Or teaching material form exists, in order to reach the target of computer aided decision making, it is necessary first to which these documents and materials are organized into meter
The knowledge information that machine information system can read, search for, analyzing is calculated, and then forms information-based document repository.Clinical case history
Knowledge base is based on big data thought, information technology, with the knowledge architecture that existing literature is recorded, such as symptom and sign, inspection
Test inspection result etc. be foundation, by information system arrange, refine, cleaning integration and etc. after establish.
In some embodiments of the invention, base module further include: known based on document repository and/or clinical case history
Know the diagnostic point knowledge base that library is established comprising illness information, symptom information, the illness information are reflected with illness information
Penetrate rule.It is undoubtedly most heavy accurately to obtain patient institute illness disease by analysis as far as possible for the system that this system is assisted as case history
One of work wanted, therefore preferably to reach above-mentioned purpose, the present invention also advanced optimizes knowledge base, establishes diagnosis and wants
Point knowledge base enables assistant analysis module targetedly by modes such as matching/lookup/retrievals, based on symptom information etc.
The medical information that other diagnostic point knowledge bases are included, and corresponding mapping ruler are analyzed to obtain the illness information of patient,
For example, such as being wanted in " clinic diagnosis guide (the nephrology fascicle) " of People's Health Publisher to the diagnosis of nephrotic syndrome
Point is described as follows: (one) High-grade Proteinuria (quantity of proteinuria > 3.5g/d)；(2) hypoalbuminemia (plasma albumin < 30g/
L)；(3) treating serious edema caused；(4) hyperlipidemia (plasma cholesterol, triglyceride obviously increase).Upper in information system
The symptom information that four diagnostic points are organized into nephrotic syndrome (illness information) is stated, i.e. diagnosis differentiates knowledge information, using phase
As mode to all kinds of diseases diagnosis require arrange, formed diagnostic point knowledge base.
In some embodiments of the invention, it can not only be established based on document repository and/or clinical case history knowledge base
Diagnostic point knowledge base, can also according to analysis conclusion actual purpose demand, establish other with specific aim type
Knowledge base, such as drug knowledge base, therapeutic scheme knowledge base can use foundation identical with above-mentioned diagnostic point knowledge base
Principle and method, therefore be no longer described in detail herein.
In some embodiments of the invention, clinical case history knowledge base included be that medicine Jing Guo standardization is known
Know.As described above, clinical case history knowledge base be can by information system arrange, refine, cleaning integration and etc. after establish.
In fact, clinical case history content is often actively write by doctor, the expression form of the inside is rich and varied and meets natural language
Expression way can undoubtedly take a significant amount of time understanding for the clinical case history of magnanimity when establishing knowledge base, extract relevant doctor
It gains knowledge, if whole process all can even more increase burden to the foundation of whole system by being accomplished manually, therefore, can pass through
A large amount of case history is handled and extracted by computer program, its effective information is obtained, natural language can be used to this
Processing technique is an important directions in computer science and artificial intelligence field, it, which is studied, is able to achieve people and meter
The various theory and methods for carrying out efficient communication between calculation machine with natural language, electronic health record data are just for clinical treatment
It is a part of natural language.With mainly used in natural language processing technique participle technique, case history text marking,
Name the contents such as Entity recognition, semantic association extraction technique.Have in the prior art and can be realized above-mentioned purpose program, herein not
An another explanation.
In some embodiments of the invention, medical record information module includes: medical record information storage unit and medical record information mark
Standardization unit, wherein medical record information Standardisation Cell is used to analyze the medical record information in medical record information storage unit, is accorded with
Close the standard index value of natural language.It is similar with purpose described above, by the standardization to medical record information, convenient for height
Effect establishes system and improves the analysis efficiency to case history.
In some embodiments of the invention, assistant analysis module includes the analysis model for analyzing medical record information, together
A kind of disaggregated model is established based on more than one classifier algorithms.Analysis model is based on knowledge base information, using artificial intelligence
Energy algorithm is established, and analysis model includes algorithm, program and configuration parameter composition.Wherein, it is identified and is calculated based on artificial intelligence diagnosis
Method includes Bayesian Classification Arithmetic, random forest Decision Tree Algorithm, support vector machine classifier, is based on K nearest neighbor algorithm point
Class device etc..It for the disaggregated model of same purpose type, is established, can be obtained most by practical comparison by algorithms of different
Excellent one or more of analysis models guarantee that this system has optimal analysis conclusion, it may also be said to, it is established by algorithms of different
Analysis model simultaneously compares screening, is a kind of tuning mode to this system.Analysis model can be established based on machine learning mode,
Which is commonly applied to AI related fields, and concrete principle and structure are no longer described in detail.
In some embodiments of the invention, assistant analysis module further include: for training the model of optimizing and analyzing model
Training unit, training unit can establish the mapping ruler between medical knowledge.Model training unit can be with clinical big data
Sample constantly trains optimizing and analyzing model, and training and the content optimized include medical record information recognition capability, medical record information identification
The correlation between discriminant parameter value section, diagnostic point in granule size, diagnostic point etc. content.For example, in training
It finds that blood urine situation is possible to be described below in case history in the process: " hematuria ", " it was found that microscopic hematuria ", " macroscopic blood
Urine ", gross hematuria centainly includes microscopic hematuria, in the presence of very big correlation between these descriptors, if doctor is in disease
Into being described as gross hematuria in going through, and blood under mirror is described as the diagnostic characteristic of the patient institute illness disease in the knowledge base
Urine is failed to pinpoint a disease in diagnosis or the case where mistaken diagnosis then just will appear, therefore assistant analysis module just needs to establish by model training unit
And support the correlation between these diagnostic characteristics, that is, mapping relations are established, to preferably improve auxiliary diagnosis confidence level.It is right
It is widely used means in machine learning field in being trained for optimization, the application has been disclosed creative
Specific training content, and for how to train above-mentioned content, no longer information explanation.
The present invention is based on medical literature knowledge base, and analysis refines clinical big data and generates clinical case history knowledge base,
Then artificial intelligence approach establishes assisted diagnosis algorithms model again, comprehensive in time by system when carrying out auxiliary diagnosis to patient
Acquisition according to patient medical record information, then substitute into assisted diagnosis algorithms model the working mechanism for carrying out auxiliary diagnosis.
Embodiment two
Standardization of the present embodiment based on each technical solution disclosed in embodiment one, to case history is able to carry out
Structures/methods are described further.
A kind of medical record information Standardisation Cell for standardization is present embodiments provided, is specifically included:
Case history project definition module, for according to case history document establish it is several being capable of identified case history project.Case history text
There is each in shelves the sentence of independent expression meaning can be understood as a case history project, substantially case history document be exactly by
Several case history item designs, this system greatly reduce natural language analysis based on the identification to case history project and extraction
It is required based on database size.
Extracting rule definition module, for setting the extraction conditions for extracting case history project.
Case history item recognition module, extraction conditions for being set according to extracting rule definition module extract case history item
Mesh.
Standard index definition module: for setting the standard index to be standardized.
Normalisation rule definition module, for establishing the normalisation rule to the quasi- index of the case history successful project bidding.
Standardization unit, the normalisation rule for being established according to normalisation rule definition module is to extraction case history item
Standard index in mesh is standardized, and obtains the standard index value for meeting natural language.Standardization unit needle
Target (standard index) in the case history project of one case history project is standardized, due to reducing to non-targeted
Processing so that treatment effeciency it is higher and obtain information it is intuitively effective.
Preferably, case history project includes project information unit, and project information unit includes case history project code, case history project
At least one of title, the affiliated document of case history project, higher level's case history project.In the prior art, to the form of presentation in case history
Carrying out natural languageization analysis is a current difficult point, since natural-sounding representation mode is abundant and diversification, to be reached
Automatic identification needs to establish a huge sound bank as support.The present invention first by the content of a large amount of continuous case history archives into
Row is split, each section of one case history project of case history project correspondence establishment with useful information, that is to say, that as long as finding this
Case history project can obtain the corresponding case history project of the case history project, and case history project is additionally provided with project information, passes through project
Information searching is easily more more than case history document full text is handled by natural languageization to corresponding case history project.In the present invention
Some embodiments in, case history project can be established respectively to whole case history projects of case history document, so as to case history text
Shelves all can be with being identified and handled.
Preferably, case history item recognition module its be used to identify the case history project having had built up, according to extraction
The extraction conditions of rule-definition module obtain corresponding case history project, by obtaining more targeted case history project, so that
Normalized analysis greatly improves the efficiency of identification without setting up huge analytical database and logical relation library.User's energy
It is enough to set extraction conditions according to their own needs, and then the case history project in required case history document can be accurately found,
The extracted case history project of case history item recognition module be also it is subsequent it is standardized play the role of it is key.
Preferably, the standardization identifying system further include: for the case history file catalogue to case history document classification, disease
Several case history projects can be established to each case history document in case history file catalogue by going through project definition module.For some trouble
The case history archive of person may add up huge content, even if carrying out establishing case history project to case history project, these case history projects
It is still very huge to enumerate quantity, can be classified to case history project by case history file catalogue, for mark
The standard index of standardization processing, the effective information that can be analyzed is contained only in some case history documents, therefore, to case history document
It carries out catalog classification and substantially increases the speed and accuracy of extraction so that having more specific aim for the extraction of case history project.
Preferably, case history item recognition module identifies case history project using regular expression.Regular expression is to character
String, including general character (for example, letter between a to z) and spcial character (referred to as " metacharacter "), a kind of logic of operation is public
Formula is exactly the combination with predefined some specific characters and these specific characters, forms one " regular character string ",
This " regular character string " is used to express a kind of filter logic to character string.Regular expression is a kind of Text Mode, mode
Matched one or more character strings are wanted in description when searching for text.
Preferably, the extraction conditions of extracting rule definition module include characteristic key words, case history statement prefix, case history sentence
At least one of suffix, exclusion keyword.
Since different case histories is to the same case history project, project information may have multiple representations, only with one
Group extraction conditions are possible to occur extracting the situation of no result or mistake in advance, then just needing to define the mould of multiple identifications
Formula to guarantee precision, and is needed to adjust each identification and is ranked up, and it is forward that the high identification of accuracy adjusts sequence.Cause
This, extracting rule definition module of the present invention includes several identification groups being made of extraction conditions, and any two identification group
Extraction conditions are not exactly the same, and case history item recognition module can extract case history project according at least one identification group.
In order to ensure the result of extraction is accurate, the system that S2 according to the present invention is established further includes authentication module, is used for
Whether verifying case history item recognition module has extracted correct case history project, is verified, and case history project has been extracted in expression
Pair extracting mode of the information based on a variety of identification groups, specifically, the present embodiment can be using more identification group verification methods be based on, i.e.,
It extracts to obtain several case history projects using multiple groups identification group in same case history document, the case history project and basis that original is extracted
The case history comparison of item that multiple groups identification group is extracted is such as all identical it may be considered that being verified.
Preferably, standard index definition module is for setting the standard index to be standardized.Standardize case history project
According to the standardized index information of medicine teaching material and documents management, comprising sign, symptom, inspection project, inspection item etc., such as
Oedema, cough, Urine proteins, renal cyst etc., these standardized informations be all will appear in case history and case history in it is highly useful
Information.The definition of these standardization projects and specific case history document all come without direct relationship, but for every part of case history
From the effective information for being included in case history project, standardizing case history project is to further refine to case history document items and markization
As a result.
Preferably, Standardization Order definition structure includes case history item information unit, specifically: mapping relations are retouched
It states, data type, data requirement, standardized calculation method, standardization ginseng value.Conclusion unit is standardized, is specifically included: case history item
The corresponding standardized index of mesh project information and standardized index value standard index.Several reference values in table are considered as analyzing
Condition can be regular, in case history project, by that just can obtain to case history item information unit whether is corresponded in case history project
To the corresponding standard index value of standard index.It is noted that the table discloses the case history project of literal type, based on upper
Normalisation rule definition module and corresponding method are stated, can be applied equally to the case history project of value type.
Preferably, the extracted standard index of standard index extraction module includes sign, symptom, inspection project, check item
At least one of mesh.Above-mentioned standard index is that doctor obtains the common index of sufferer physical condition, by These parameters
Standardization analysis, be conducive to it is intuitive, quickly grasp the state of an illness.
When standardization, it is first determined the case history project to be standardized, the case history project are by case history project
Identification module identifies and extracts, standardization conclusion unit established standards index and mark of the user in normalisation rule definition module
Standardization rule, obtains standardized index information.
Embodiment three
A method of case history Computer Aided Analysis System being established, the specific method is as follows:
Establish the medical record information module for storing patient.
Base module is established, base module includes the mapping rule of several medical knowledges and interrelated medical knowledge
Then.
Establish assistant analysis module, assistant analysis module can knowledge based library module to the case history in medical record information module
Information is analyzed and is drawn a conclusion.
In some embodiments of the invention, it may not be necessary to individually establish medical record information module, but call healthcare structure
Existing medical record information module, and then complete the foundation of the medical record information module.
System one of which application method described in embodiment one and two is as follows: when carrying out diagnostic analysis to new patient,
Patient's complete medical record information at that time is obtained by system and case history is analyzed comprehensively, all kinds of diagnosis are generated by analysis of medical record
Diagnostic point parameter value analyzed then according to diagnostic model and relevant knowledge library information by analysis model, and then generate
The disease that patient may suffer from, or even the probability of illness is analyzed, further, provide the foundation and needs of system auxiliary diagnosis
Content further identified etc..
For in the prior art, the simple search simply based on documents and materials matches, it cannot obtain at all and meet the mankind
The conclusion of thinking and natural language, and the present invention is based on document repository and clinical case history knowledge base, and can break up/thin
Change more targeted diagnostic point knowledge base, cooperates the various analysis models in assistant analysis module, finally obtaining can
The analysis authentic and valid to medical staff is as a result, not only reduce the difficulty of diagnosis, treatment, while greatly improving accurately
Property, this is also the present invention and the maximum difference of existing assistant diagnosis system and improvements.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.
Claims (9)
1. a kind of case history Computer Aided Analysis System characterized by comprising
Medical record information module, for storing the medical record information of simultaneously managing patient；
Base module comprising the mapping ruler of several medical knowledges and interrelated medical knowledge；
Assistant analysis module, can knowledge based library module the medical record information in medical record information module is analyzed and obtains knot
By.
2. a kind of case history Computer Aided Analysis System according to claim 1, which is characterized in that base module includes: to be based on
The document repository and the clinical case history knowledge base established based on clinical case history big data that documents and materials are established；Document repository and
Clinical case history knowledge base respectively includes the medical knowledge and the mapping ruler.
3. a kind of case history Computer Aided Analysis System according to claim 2, which is characterized in that base module further include: base
In the diagnostic point knowledge base that document repository and/or clinical case history knowledge base are established comprising illness information, symptom information,
The mapping ruler of the illness information and illness information.
4. a kind of case history Computer Aided Analysis System according to claim 2, which is characterized in that clinical case history knowledge base included
It is the medical knowledge Jing Guo standardization.
5. a kind of case history Computer Aided Analysis System according to claim 1, which is characterized in that medical record information module includes: disease
Information memory cell and medical record information Standardisation Cell are gone through, wherein medical record information Standardisation Cell is used to store medical record information single
Medical record information analysis in member, obtains the standard index value for meeting natural language.
6. a kind of case history Computer Aided Analysis System according to claim 4 or 5, which is characterized in that for standardization
Medical record information Standardisation Cell specifically includes:
Case history project definition module, for newly established according to case history it is several being capable of identified case history project；
Extracting rule definition module, for setting the extraction conditions for extracting case history project from case history document；
Case history item recognition module, for according to the extraction conditions come from being extracted in case history document in specific case history project
Hold；
Standard index definition module: for setting the standard index to be standardized；
Normalisation rule definition module, for establishing the rule being standardized to the case history contents of a project；
Standardization unit, the normalisation rule for being established according to normalisation rule definition module is in extraction case history project
Appearance is standardized, and obtains the standard index value for meeting natural language.
7. a kind of case history Computer Aided Analysis System according to claim 1, which is characterized in that assistant analysis module includes being used for
The analysis model of medical record information is analyzed, same disaggregated model is established based on more than one classifiers.
8. a kind of case history Computer Aided Analysis System according to claim 7, which is characterized in that assistant analysis module further include:
For training the model training unit of optimizing and analyzing model, training unit can establish the mapping ruler between medical knowledge.
9. a kind of method for establishing case history Computer Aided Analysis System, which is characterized in that the specific method is as follows:
The medical record information module for storing patient is established,
Base module is established, base module includes the mapping ruler of several medical knowledges and interrelated medical knowledge；
Establish assistant analysis module, assistant analysis module can knowledge based library module to the medical record information in medical record information module
It is analyzed and is drawn a conclusion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811144073.2A CN109390058A (en) | 2018-09-28 | 2018-09-28 | A kind of method for building up of case history Computer Aided Analysis System and the system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811144073.2A CN109390058A (en) | 2018-09-28 | 2018-09-28 | A kind of method for building up of case history Computer Aided Analysis System and the system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109390058A true CN109390058A (en) | 2019-02-26 |
Family
ID=65418237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811144073.2A Pending CN109390058A (en) | 2018-09-28 | 2018-09-28 | A kind of method for building up of case history Computer Aided Analysis System and the system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109390058A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109918436A (en) * | 2019-03-08 | 2019-06-21 | 上海一健事信息科技有限公司 | A kind of Medical Knowledge management and inquiry system |
| CN110379507A (en) * | 2019-06-27 | 2019-10-25 | 南京市卫生信息中心 | A kind of aided diagnosis method based on patient's vector-valued image |
| CN110471941A (en) * | 2019-08-12 | 2019-11-19 | 贵州医渡云技术有限公司 | It is automatically positioned the method, apparatus and electronic equipment of judgment basis |
| CN110660459A (en) * | 2019-08-30 | 2020-01-07 | 腾讯科技（深圳）有限公司 | Method, device, server and storage medium for controlling medical record quality |
| CN111177356A (en) * | 2019-12-31 | 2020-05-19 | 佛山科学技术学院 | Acid-base index medical big data analysis method and system |
| CN111354459A (en) * | 2020-02-25 | 2020-06-30 | 成都联客信息技术有限公司 | Auxiliary diagnosis expert system for traditional Chinese medicine massage |
| CN111599483A (en) * | 2020-05-29 | 2020-08-28 | 安徽科大讯飞医疗信息技术有限公司 | Medical record set optimization method, device, equipment and storage medium |
| CN112133385A (en) * | 2020-09-14 | 2020-12-25 | 黄军彩 | Inquiry information acquisition and analysis system and acquisition and analysis method |
| WO2021114620A1 (en) * | 2020-06-16 | 2021-06-17 | 平安科技（深圳）有限公司 | Medical-record quality control method, apparatus, computer device, and storage medium |
| CN113051373A (en) * | 2021-04-19 | 2021-06-29 | 安徽科大讯飞医疗信息技术有限公司 | Text analysis method and device, electronic equipment and storage medium |
| CN113707307A (en) * | 2021-08-30 | 2021-11-26 | 康键信息技术（深圳）有限公司 | Disease analysis method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149751A (en) * | 2007-10-29 | 2008-03-26 | 浙江大学 | Generalized relating rule digging method for analyzing traditional Chinese medicine recipe drug matching rule |
| CN105608091A (en) * | 2014-11-21 | 2016-05-25 | 中国移动通信集团公司 | Construction method and device of dynamic medical knowledge base |
| AU2017210493A1 (en) * | 2014-08-14 | 2017-08-17 | Accenture Global Services Limited | System for automated analysis of clinical text for pharmacovigilance |
| CN107863147A (en) * | 2017-10-24 | 2018-03-30 | 清华大学 | The method of medical diagnosis based on depth convolutional neural networks |
| CN107887036A (en) * | 2017-11-09 | 2018-04-06 | 北京纽伦智能科技有限公司 | Construction method, device and the clinical decision accessory system of clinical decision accessory system |
-
2018
- 2018-09-28 CN CN201811144073.2A patent/CN109390058A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101149751A (en) * | 2007-10-29 | 2008-03-26 | 浙江大学 | Generalized relating rule digging method for analyzing traditional Chinese medicine recipe drug matching rule |
| AU2017210493A1 (en) * | 2014-08-14 | 2017-08-17 | Accenture Global Services Limited | System for automated analysis of clinical text for pharmacovigilance |
| CN105608091A (en) * | 2014-11-21 | 2016-05-25 | 中国移动通信集团公司 | Construction method and device of dynamic medical knowledge base |
| CN107863147A (en) * | 2017-10-24 | 2018-03-30 | 清华大学 | The method of medical diagnosis based on depth convolutional neural networks |
| CN107887036A (en) * | 2017-11-09 | 2018-04-06 | 北京纽伦智能科技有限公司 | Construction method, device and the clinical decision accessory system of clinical decision accessory system |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109918436A (en) * | 2019-03-08 | 2019-06-21 | 上海一健事信息科技有限公司 | A kind of Medical Knowledge management and inquiry system |
| CN109918436B (en) * | 2019-03-08 | 2022-12-20 | 麦博(上海)健康科技有限公司 | Medical knowledge management and query system |
| CN110379507A (en) * | 2019-06-27 | 2019-10-25 | 南京市卫生信息中心 | A kind of aided diagnosis method based on patient's vector-valued image |
| CN110379507B (en) * | 2019-06-27 | 2023-03-14 | 南京市卫生信息中心 | Auxiliary diagnosis method based on patient vector image |
| CN110471941B (en) * | 2019-08-12 | 2020-10-02 | 贵州医渡云技术有限公司 | Method and device for automatically positioning judgment basis and electronic equipment |
| CN110471941A (en) * | 2019-08-12 | 2019-11-19 | 贵州医渡云技术有限公司 | It is automatically positioned the method, apparatus and electronic equipment of judgment basis |
| CN110660459A (en) * | 2019-08-30 | 2020-01-07 | 腾讯科技（深圳）有限公司 | Method, device, server and storage medium for controlling medical record quality |
| CN111177356A (en) * | 2019-12-31 | 2020-05-19 | 佛山科学技术学院 | Acid-base index medical big data analysis method and system |
| CN111177356B (en) * | 2019-12-31 | 2023-03-31 | 佛山科学技术学院 | Acid-base index medical big data analysis method and system |
| CN111354459A (en) * | 2020-02-25 | 2020-06-30 | 成都联客信息技术有限公司 | Auxiliary diagnosis expert system for traditional Chinese medicine massage |
| CN111599483A (en) * | 2020-05-29 | 2020-08-28 | 安徽科大讯飞医疗信息技术有限公司 | Medical record set optimization method, device, equipment and storage medium |
| CN111599483B (en) * | 2020-05-29 | 2023-12-22 | 讯飞医疗科技股份有限公司 | Medical record set optimization method, device, equipment and storage medium |
| WO2021114620A1 (en) * | 2020-06-16 | 2021-06-17 | 平安科技（深圳）有限公司 | Medical-record quality control method, apparatus, computer device, and storage medium |
| CN112133385A (en) * | 2020-09-14 | 2020-12-25 | 黄军彩 | Inquiry information acquisition and analysis system and acquisition and analysis method |
| CN113051373A (en) * | 2021-04-19 | 2021-06-29 | 安徽科大讯飞医疗信息技术有限公司 | Text analysis method and device, electronic equipment and storage medium |
| CN113051373B (en) * | 2021-04-19 | 2024-02-13 | 讯飞医疗科技股份有限公司 | Text analysis method, text analysis device, electronic equipment and storage medium |
| CN113707307A (en) * | 2021-08-30 | 2021-11-26 | 康键信息技术（深圳）有限公司 | Disease analysis method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109390058A (en) | A kind of method for building up of case history Computer Aided Analysis System and the system | |
| US11176150B2 (en) | Method and system for text understanding in an ontology driven platform | |
| CN106919671B (en) | Intelligent system for traditional Chinese medicine text case mining and assistant decision making | |
| Nebot et al. | Multidimensional integrated ontologies: A framework for designing semantic data warehouses | |
| CN113505243A (en) | Intelligent question-answering method and device based on medical knowledge graph | |
| Xiang et al. | Knowledge graph-based clinical decision support system reasoning: a survey | |
| Wu et al. | A novel community answer matching approach based on phrase fusion heterogeneous information network | |
| CN115293161A (en) | Reasonable medicine taking system and method based on natural language processing and medicine knowledge graph | |
| Sun | Methods for automated concept mapping between medical databases | |
| Wang et al. | Research on named entity recognition of doctor-patient question answering community based on bilstm-crf model | |
| Zubke | Classification based extraction of numeric values from clinical narratives. | |
| Saint-Dizier et al. | Knowledge and reasoning for question answering: Research perspectives | |
| Zhitomirsky-Geffet et al. | Maximizing agreement on diverse ontologies with “wisdom of crowds” relation classification | |
| CN116719840A (en) | Medical information pushing method based on post-medical-record structured processing | |
| Zhang et al. | An effective standardization method for the lab indicators in regional medical health platform using n-grams and stacking | |
| CN116227594A (en) | Construction method of high-credibility knowledge graph of medical industry facing multi-source data | |
| CN114860852A (en) | Knowledge graph construction method for military field | |
| CN113314236A (en) | Intelligent question-answering system for hypertension | |
| Li et al. | RCMR 280k: refined corpus for move recognition based on PubMed abstracts | |
| Dubey et al. | Smart Patient Records using NLP and Blockchain | |
| Fan et al. | Knowledge base construction based on knowledge fusion process model | |
| Chen et al. | Research on the key technology of course ontology construction | |
| Feng et al. | Extracting meaningful correlations among heterogeneous datasets for medical question answering with domain knowledge | |
| Rassinoux | Knowledge Representation and Management: Benefits and Challenges of the Semantic Web for the Fields of KRM and NLP | |
| Siklósi | Clustering relevant terms and identifying types of statements in clinical records |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190226 |
|
| RJ01 | Rejection of invention patent application after publication |