CN107193996A - Similar case history matches searching system - Google Patents
Similar case history matches searching system Download PDFInfo
- Publication number
- CN107193996A CN107193996A CN201710436539.5A CN201710436539A CN107193996A CN 107193996 A CN107193996 A CN 107193996A CN 201710436539 A CN201710436539 A CN 201710436539A CN 107193996 A CN107193996 A CN 107193996A
- Authority
- CN
- China
- Prior art keywords
- case history
- variable
- unit
- similarity
- variables
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24578—Query processing with adaptation to user needs using ranking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention discloses a kind of similar case history matching searching system, including case history import unit, variable extraction unit, multi-variables analysis unit and arrangement and returning unit, wherein, case history import unit is used to import case history, and carrying out processing by filter reads the personal information storage of patient in ephemeral data library file;The variable extraction that variable extraction unit is used for case history matching is stored in variable extraction list;The analysis and synthesis that multi-variables analysis unit is used for multivariable collects, and obtains the similarity of each part case history;Arrange and returning unit carries out descending arrangement to the similarity of gained, ranking results are returned into user terminal, user is judged returning result and chosen to ask for complete case history, so, analysis extraction variable is carried out by the electronic health record deposited in the electronic health record or local data base to importing and collected, searching the archives that have certain similarity with former case history in electronic health record file store again, there is provided, as reference, improve recall precision to user.
Description
Technical field
The present invention relates to field of medical technology, a kind of similar case history matching searching system is particularly related to.
Background technology
Clinician can have one when carrying out analysis of cases for certain part of case history or writing case report by using for reference
Determine the history case history of similarity, therefrom obtain some diagnostic comments and treatment method for referring to.Or in clinical research, certain
Needed in the case of a little from certain part of case history as starting point, find more similar case histories and carry out research discussion.
Under above two situation, if by manually being searched in huge electronic health record file store, it will consumption
Take many time energy.Accordingly, it would be desirable to a kind of case history matching searching system be developed, using certain part of case history as starting point, in electronics
The electronic health record with similarity is searched in case history archive storehouse, returns to user to provide the reference of history case history.
The content of the invention
For problem present in background technology, it is an object of the invention to provide a kind of similar case history matching searching system,
Analyzed by the electronic health record deposited in the electronic health record or local data base to importing, then in electronic health record file store
Searching the archives that have certain similarity with former case history, there is provided, as reference, improve recall precision to user.
The technical proposal of the invention is realized in this way:A kind of similar case history matching searching system, including case history import list
Member, variable extraction unit, multi-variables analysis unit and arrangement and returning unit, wherein, case history import unit:For importing case history
And handled by filter, filter according to default electronic health record form controls-patient data mapping relations in system,
The personal information of patient is read into electronic medical record document and is stored in ephemeral data library file, and original document is then deposited in
In volatile data base;Variable extraction unit:Extracted for the variable that case history is matched, relation is extracted according to default variable in system
Table, and corresponding electronic medical record document control-variable data corresponding relation, to the electronic health record being stored in volatile data base
Document reads variable storage and extracted in variable in list;Multi-variables analysis unit:For the analysis of multivariable, complete variable and extract
Afterwards, variable extraction list is transmitted to medical record management server, server extracts list in case history index data base according to variable
It is middle to be retrieved, and synthesis is carried out to variable, obtain the similarity of each part case history;Arrange and returning unit:To gained
Similarity carries out descending arrangement, ranking results is returned into user terminal, user is judged and chosen to the result of return, from
And ask for complete case history.
In the above-mentioned technical solutions, the types of variables includes numeric type variable, character string type variable and logical type variable.
In the above-mentioned technical solutions, the electronic medical record document control-variable data corresponding relation be name variable with it is right
Control and types of variables is answered to correspond.
In the above-mentioned technical solutions, deposited in the case history index data base and relation table is extracted according to the variable of regulation, from
The variable extracted respectively to every part of case history in case history archive database.
The similar case history matching searching system of the present invention, including case history import unit, variable extraction unit, multi-variables analysis list
Member and arrangement and returning unit, wherein, case history import unit is used to import case history, carries out processing reading patient's by filter
Personal information storage is in ephemeral data library file;Variable extraction unit is stored in variable for the variable extraction that case history is matched and carried
Take in list;The analysis and synthesis that multi-variables analysis unit is used for multivariable collects, and obtains the similarity of each part case history;Arrange simultaneously
Returning unit carries out descending arrangement to the similarity of gained, ranking results is returned into user terminal, user enters to returning result
Row judges and chosen to ask for complete case history, in this way, passing through the electronics disease deposited in the electronic health record or local data base to importing
Go through progress analysis to extract variable and collect, then the shelves that there is certain similarity with former case history are searched in electronic health record file store
There is provided, as reference, improve recall precision to user for case.
Brief description of the drawings
Fig. 1 is similar case history matching searching system schematic flow sheet of the invention;
Fig. 2 is the interactive relation schematic diagram of client terminal and database in the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on this
Embodiment in invention, the every other reality that those of ordinary skill in the art are obtained under the premise of creative work is not made
Example is applied, the scope of protection of the invention is belonged to.
A kind of similar case history matching searching system of the present invention, is that multiple variable synthetical is analyzed, especially with character string type
The use of data analysing method and numeric type data analysing method and data comb is key.Pass through certain part of electricity imported to user
The electronic health record deposited in sub- case history or local data base extracts key variables, carries out multiple variable synthetical analysis, then will integrate
The multi-variables analysis result crossed is sent to medical record management server, and multivariable is carried out in electronic health record file store and is searched and sieve
Choosing, so as to obtain and original case history the most similar case history.
The different content of every part of case history can be considered as different variables, can there is different take for some variable
Value.If value identical variable is more between two variables, then two parts of case histories are more similar.Wherein, some variables are core
Heart variable, the most key effect is played between case history matching.
The similar case history matching searching system of the present invention specifically includes case history import unit, variable extraction unit, multivariable
Analytic unit and arrangement and returning unit, the similar case history matching searching system flow of the invention with reference to shown in Fig. 1 is to above-mentioned each
Unit describes in detail as follows:
(1) case history import unit:
User imports a electronic medical record document into electronic health record archive management system, and the document of importing is whole into case history
Reason system, is handled using filter.Filter maps according to default electronic health record form controls in system-patient data
Relation, the personal information of patient is read into electronic medical record document, by the data storage obtained by reading in ephemeral data library file
In, and original document is then deposited in volatile data base.
(2) variable extraction unit:
Extracted for the variable that case history is matched, mainly the data related to medical information.Safeguarding in advance in systems has
A variable extracts relation table, and corresponding electronic medical record document control-variable data corresponding relation, as shown in the table:
Name variable | Correspondence control | Types of variables |
Age | Dbo. first page of illness case-Textbox3.Text | integer |
Gender | Dbo. first page of illness case-Textbox5.Text | string |
C.C. | Dbo. progress note-Textbox3.Text | string |
Imp | Dbo. progress note-Textbox5.Text | string |
Diagnosis | Dbo. progress note-Textbox7.Text | string |
Wherein, be required for providing its types of variables corresponding to each variable, such as numerical value, text, logical type etc., no
The method that same types of variables is used in multi-variables analysis is different.
Relation table is extracted according to variable, system reads above-mentioned change to the electronic medical record document being stored in volatile data base
Amount, is stored in variable and extracts in list, as shown in the table:
Age | Gender | ||
C.C. | Imp | ||
Diagnosis | … |
(3) multi-variables analysis unit:
After electronic health record archive management system user terminal completes variable extraction, variable extraction list is transmitted to medical record
Management server.Server extracts list according to the variable that receives, retrieved in case history index data base and and it is non-straight
It is connected in case history archive database and is retrieved.
Because the document in case history archive database is full storage, wherein containing a variety of medical treatment notes unrelated with diagnosis and treatment
Record, the contents are multifarious and disorderly, if directly case history is fished in case history archive database will need great operand.In order to reduce computing
Amount from the case history in case history archive database, it is necessary to extract and diagnosis and treatment the most closely related information in advance.Specific method is
Relation table is extracted according to the variable of regulation, central variable is extracted respectively to every part of case history from case history archive database, deposits
It is put in case history index data base.
The structure of case history index data base is as shown in the table, and major key is admission number:
Admission number | Age | Gender | C.C. | Imp | Diagnosis | … |
Server extracts inventory according to the variable sent, is retrieved in case history index data base.The method of retrieval
It is each variable progress similarity computing for every part of case history, from top to down, therefore referred to as data are combed.Data are combed to different type
The different analysis method of variable uses, be respectively:
A. numeric type:The analysis of numeric type variable is calculated using formula:
The a tried to achieve is the similarity of numeric type variable.
B. character string type:The analysis of character string type variable is complexity compared with numeric type.Due to the variable and case history rope of source file
The case history variable drawn in database always has points of resemblance, and also has difference, it is impossible to carry out logic judgment merely, it is therefore desirable to
Analyzed from the angle of character string.
Specific method is:By pair of certain part of case history in the alphabet in the variable of source file and case history index data base
The alphabet of dependent variable is mixed, and as a character string pond, (central without repeat character (RPT), repeat character (RPT) is closed
And), then calculating the number of characters all occurred in two character strings, then the similarity of character string type variable can be calculated such as with formula
Under:
C. logical type variable:Logical type variable analysis is relatively simple, if the variable in source file and database
Logic judgment is identical, then score is 1, is otherwise 0,
Data have been combed after the computing of paired variates, obtain variable record sheet as follows:
Admission number | Age | Gender | C.C. | Imp | Diagnosis | … |
The table structure is identical with case history index data base, but the data of central storage are similarity a, b or c.
Complete after the processing of data comb, it is necessary to carry out synthesis to variable.For every a variable, according to following formula
Calculate total similarity:
Wherein, weight is weight, and different weights are imparted for different variables.For example, the weight at age is
0.03, the weight of admission diagnosis is 0.1, and weighted value is preset in the server.According to above-mentioned formula, the similar of each part case history is obtained
Spend φ.
(4) arrange and returning unit:
Server carries out descending arrangement to the similarity φ values obtained by calculating, and some positions before ranking results are indexed in case history
Entry in database returns to user terminal.User is judged and chosen to the result of return, interested among selection
Case history archive, so as to ask for complete case history to server.System constructing case history asks for request message, is sent to server, service
Device asks for request message further according to case history, transfers the complete case history in case history archive database, returns to client terminal.Wherein,
The interactive relation of client terminal and database is as shown in Figure 2.
The following is further illustrating that one instantiation of combination is carried out:
Certain GI Medicine doctor needs to carry out the case history of a Diagnosing Gastrointestinal Bleeding the retrieval of similar case history, by becoming
Amount is extracted, and the variable of the case history, which is extracted, to be listed as follows shown in table:
Age | 46 | Gender | Man |
C.C. | Vomitus niger | Imp | Hemorrhage of digestive tract |
Diagnosis | Acute gastritis | … |
In case history index data base, the medical information of certain part of case history is as shown in the table:
Age | 44 | Gender | Female |
C.C. | Vomitus niger, it is heartburn | Imp | Hemorrhage of digestive tract |
Diagnosis | Chronic gastritis | … |
The weighted value of different variables is as shown in the table:
Age | 0.03 | Gender | 0.03 |
C.C. | 0.15 | Imp | 0.18 |
Diagnosis | 0.32 | … |
The similarity operation result of each variable of the case history such as following table institute in the combing combed by data, case history index data base
Show:
Age | 0.03 | Gender | 0.03 |
C.C. | 0.15 | Imp | 0.18 |
Diagnosis | 0.32 | … |
Gained is calculated by multi-variables analysis formula:The corresponding similarity of the case history:
Each part case history in case history searching database is completed after multi-variables analysis, and last descending sort result is:
Admission number | φ |
136409 | 0.654 |
164923 | 0.643 |
153213 | 0.543 |
124243 | 0.476 |
The result is returned into user terminal.User's selection needs to obtain after the entry of complete case history, sends complete disease
Go through and ask for request message, be sent to server, server asks for request message further according to case history, transfer in case history archive database
Complete case history, return to client terminal.
To sum up, similar case history matching searching system of the invention, has the advantages that compared with prior art:
1. in a electronic medical record document, include medical information and other medical documents record, if to archives
All variables in each part case history in storehouse all carry out analytic operation, it will have many unrelated variables to cause to do to analysis result
Disturb, the degree of accuracy is difficult to ensure that.And this kind of case history matching searching system sets some variables mostly concerned to diagnosis and treatment, eliminate
The variable unrelated with diagnosis and treatment, extracts relation table according to variable during analysis and carries out variable extraction from the case history of importing, and according to carrying
The variable of taking-up in case history index data base by data comb carry out multi-variables analysis, be divided into numeric type, character string type and
Logical type variable is analyzed different variables, so as to obtain the similarity of each part case history, accuracy is higher.
2. conventional clinician or scientific research personnel need to analyze the case history of some patient or write case report,
The similar case history of history is looked for if desired, it is necessary to manually carry out checking screening from database, it is extremely inefficient, and when expending substantial amounts of
Between energy.And the data that the system is then imported according to user, server is according to the variable extracted in case history index data base
Middle to search, arithmetic speed is high, can carry out multi-variables analysis to the case history in database within the extremely short time, largely
Improve operating efficiency.
If 3. directly searched in electronic health record file store, because every part of electronic health record has considerable data
Content, and many of which is unrelated with diagnosis and treatment, the final data scale of construction is extremely huge, causes operand larger.And the system pass through it is new
Build a case history index data base, extracted in advance in electronic health record file store with diagnosis and treatment relation the closest variable, this
Sample server avoids the need for carrying out complicated computing in electronic health record file store, is directly transported in case history index data base
Calculate, operand is reduced to a certain extent.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God is with principle, and any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.
Claims (4)
1. a kind of similar case history matching searching system, it is characterised in that:Including case history import unit, variable extraction unit, changeable
Measure analytic unit and arrange and returning unit, wherein,
Case history import unit:For importing case history and being handled by filter, filter is according to default electronics in system
Case history form controls-patient data mapping relations, the personal information of patient is read into electronic medical record document and is stored in interim
In database file, and original document is then deposited in volatile data base;
Variable extraction unit:Extracted for the variable that case history is matched, relation table is extracted according to default variable in system, and it is right
Electronic medical record document control-variable data the corresponding relation answered, reads to the electronic medical record document being stored in volatile data base
Variable storage is extracted in list in variable;
Multi-variables analysis unit:For the analysis of multivariable, complete after variable extraction, variable extraction list is transmitted to medical record pipe
Server is managed, server extracts list according to variable and retrieved in case history index data base, and variable is carried out to integrate remittance
Always, the similarity of each part case history is obtained;
Arrange and returning unit:Similarity to gained carries out descending arrangement, and ranking results are returned into user terminal, user couple
The result of return is judged and chosen, so as to ask for complete case history.
2. similar case history matching searching system according to claim 1, it is characterised in that:The types of variables includes numerical value
Type variable, character string type variable and logical type variable.
3. similar case history matching searching system according to claim 1, it is characterised in that:The electronic medical record document control
Part-variable data corresponding relation is that name variable is corresponded with corresponding control and types of variables.
4. similar case history matching searching system according to claim 1, it is characterised in that:In the case history index data base
Storage extracts relation table, the variable extracted respectively to every part of case history from case history archive database according to the variable of regulation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710436539.5A CN107193996B (en) | 2017-06-09 | 2017-06-09 | Similar medical record matching and retrieving system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710436539.5A CN107193996B (en) | 2017-06-09 | 2017-06-09 | Similar medical record matching and retrieving system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107193996A true CN107193996A (en) | 2017-09-22 |
CN107193996B CN107193996B (en) | 2021-02-12 |
Family
ID=59876261
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710436539.5A Active CN107193996B (en) | 2017-06-09 | 2017-06-09 | Similar medical record matching and retrieving system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107193996B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108932981A (en) * | 2018-07-18 | 2018-12-04 | 深圳市有钱科技有限公司 | Medical data exchange method and device |
CN110197214A (en) * | 2019-05-22 | 2019-09-03 | 浙江大学 | A kind of patient identity matching process based on multi-field similarity calculation |
CN110517789A (en) * | 2019-08-30 | 2019-11-29 | 深圳市汇健医疗工程有限公司 | The digital composite operating room of a variety of image documentation equipments |
CN111724873A (en) * | 2020-06-18 | 2020-09-29 | 北京嘉和海森健康科技有限公司 | Data processing method and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140036941A (en) * | 2012-09-17 | 2014-03-26 | 삼성전자주식회사 | Method for query formulation for pattern searches in health records |
CN104572675A (en) * | 2013-10-16 | 2015-04-29 | 中国人民解放军南京军区南京总医院 | Similar medical history searching system and method |
CN104881463A (en) * | 2015-05-22 | 2015-09-02 | 清华大学深圳研究生院 | Reference medical record search method and device based on structural medical record database |
CN105893597A (en) * | 2016-04-20 | 2016-08-24 | 上海家好科技有限公司 | Similar medical record retrieval method and system |
-
2017
- 2017-06-09 CN CN201710436539.5A patent/CN107193996B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140036941A (en) * | 2012-09-17 | 2014-03-26 | 삼성전자주식회사 | Method for query formulation for pattern searches in health records |
CN104572675A (en) * | 2013-10-16 | 2015-04-29 | 中国人民解放军南京军区南京总医院 | Similar medical history searching system and method |
CN104881463A (en) * | 2015-05-22 | 2015-09-02 | 清华大学深圳研究生院 | Reference medical record search method and device based on structural medical record database |
CN105893597A (en) * | 2016-04-20 | 2016-08-24 | 上海家好科技有限公司 | Similar medical record retrieval method and system |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108932981A (en) * | 2018-07-18 | 2018-12-04 | 深圳市有钱科技有限公司 | Medical data exchange method and device |
CN110197214A (en) * | 2019-05-22 | 2019-09-03 | 浙江大学 | A kind of patient identity matching process based on multi-field similarity calculation |
CN110517789A (en) * | 2019-08-30 | 2019-11-29 | 深圳市汇健医疗工程有限公司 | The digital composite operating room of a variety of image documentation equipments |
CN110517789B (en) * | 2019-08-30 | 2023-06-16 | 深圳市汇健医疗工程有限公司 | Digital composite operating room with multiple image devices |
CN111724873A (en) * | 2020-06-18 | 2020-09-29 | 北京嘉和海森健康科技有限公司 | Data processing method and device |
CN111724873B (en) * | 2020-06-18 | 2024-01-09 | 北京嘉和海森健康科技有限公司 | Data processing method and device |
Also Published As
Publication number | Publication date |
---|---|
CN107193996B (en) | 2021-02-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhao et al. | Analysis and visualization of citation networks | |
Lu et al. | Collaborative graph learning with auxiliary text for temporal event prediction in healthcare | |
CN107193996A (en) | Similar case history matches searching system | |
CN109753516A (en) | A kind of sort method and relevant apparatus of case history search result | |
CN108446260A (en) | The method and system of automation disease code conversion are carried out based on semantic approximate match algorithm | |
Li et al. | Lightweight end-to-end neural network model for automatic heart sound classification | |
CN108986907A (en) | A kind of tele-medicine based on KNN algorithm divides the method for examining automatically | |
Basco et al. | Real-time analysis of healthcare using big data analytics | |
Park et al. | Knowledge discovery with machine learning for hospital-acquired catheter-associated urinary tract infections | |
Deng et al. | An ensemble CNN method for biomedical entity normalization | |
Hong et al. | Event2vec: Learning representations of events on temporal sequences | |
Biseda et al. | Prediction of ICD codes with clinical BERT embeddings and text augmentation with label balancing using MIMIC-III | |
Ding et al. | Diagnosing crop diseases based on domain-adaptive pre-training BERT of electronic medical records | |
CN107273405A (en) | The intelligent retrieval system of electronic health record archives based on MeSH tables | |
JP6963535B2 (en) | Analytical methods, analyzers and programs | |
Memarzadeh et al. | A study into patient similarity through representation learning from medical records | |
Kim et al. | User–Topic Modeling for Online Community Analysis | |
CN113821641B (en) | Method, device, equipment and storage medium for classifying medicines based on weight distribution | |
Mathew et al. | Distributed privacy-preserving decision support system for highly imbalanced clinical data | |
CN113590845A (en) | Knowledge graph-based document retrieval method and device, electronic equipment and medium | |
CN110010231A (en) | A kind of data processing system and computer readable storage medium | |
JP7384705B2 (en) | Analytical equipment, analytical methods, and analytical programs | |
CN110033862B (en) | Traditional Chinese medicine quantitative diagnosis system based on weighted directed graph and storage medium | |
Trigo et al. | Retrieval, visualization and validation of affinities between documents | |
Theodorou et al. | TREEMENT: Interpretable Patient-Trial Matching via Personalized Dynamic Tree-Based Memory Network |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |