CN107193996B - Similar medical record matching and retrieving system - Google Patents
Similar medical record matching and retrieving system Download PDFInfo
- Publication number
- CN107193996B CN107193996B CN201710436539.5A CN201710436539A CN107193996B CN 107193996 B CN107193996 B CN 107193996B CN 201710436539 A CN201710436539 A CN 201710436539A CN 107193996 B CN107193996 B CN 107193996B
- Authority
- CN
- China
- Prior art keywords
- medical record
- variable
- unit
- variables
- file
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
Abstract
The invention discloses a similar medical record matching retrieval system, which comprises a medical record importing unit, a variable extracting unit, a multivariate analyzing unit and a sorting and returning unit, wherein the medical record importing unit is used for importing medical records, processing the medical records through a filter, reading personal information of a patient and storing the personal information in a temporary database file; the variable extraction unit is used for extracting and storing the variables matched with the medical records in a variable extraction list; the multivariate analysis unit is used for multivariate analysis and comprehensive summarization to obtain the similarity of each medical record; the sorting and returning unit performs descending order arrangement on the obtained similarity, the sorting result is returned to the user terminal, the user judges the returned result and selects and requests a complete medical record, thus, the imported electronic medical record or the electronic medical record stored in the local database is analyzed, the variables are extracted and summarized, the file with certain similarity to the original medical record is searched in the electronic medical record archive, and the file is provided for the user as reference, so that the retrieval efficiency is improved.
Description
Technical Field
The invention relates to the technical field of medical treatment, in particular to a matching and searching system for similar medical records.
Background
When a clinician analyzes a case or writes a case report for a certain medical record, the clinician can obtain some consultable diagnosis opinions and treatment methods from the historical medical record with a certain similarity by reference. Or in clinical research, more similar medical records need to be searched for research and discussion from a certain medical record as a starting point in some cases.
Under the two situations, if the patient is manually searched in a huge electronic medical record archive, much time and energy are consumed. Therefore, there is a need to develop a medical record matching and retrieving system, which uses a certain medical record as a starting point to search the electronic medical record with similarity in the electronic medical record archive, and returns the electronic medical record with similarity to the user to provide a reference for the historical medical record.
Disclosure of Invention
Aiming at the problems in the background art, the invention aims to provide a similar medical record matching and searching system, which analyzes the imported electronic medical record or the electronic medical record stored in a local database, searches files with certain similarity with the original medical record in an electronic medical record archive library, and provides the files for a user as reference to improve the searching efficiency.
The technical scheme of the invention is realized as follows: the similar medical record matching and retrieving system comprises a medical record importing unit, a variable extracting unit, a multivariate analyzing unit and a sorting and returning unit, wherein the medical record importing unit: the system comprises a filter, a temporary database file and an original file, wherein the filter is used for importing medical records and processing the medical records through the filter, reading personal information of a patient from an electronic medical record file and storing the personal information in the temporary database file according to a mapping relation between a preset electronic medical record form control and patient data in the system, and storing the original file in the temporary database; a variable extraction unit: extracting variables for matching medical records, reading the variables from the electronic medical record documents stored in the temporary database and storing the variables in a variable extraction list according to a variable extraction relation table preset in the system and a corresponding electronic medical record document control-variable data corresponding relation; a multivariate analysis unit: the system is used for analyzing the multivariate, after the extraction of the variable is finished, the variable extraction list is transmitted to a medical record management server, the server searches in a medical record index database according to the variable extraction list and comprehensively summarizes the variable to obtain the similarity of each medical record; a sorting and returning unit: and performing descending order arrangement on the obtained similarity, returning the ordering result to the user terminal, and judging and checking the returned result by the user so as to ask for a complete medical record.
In the above technical solution, the variable types include a numeric variable, a string variable, and a logical variable.
In the above technical solution, the electronic medical record document control-variable data correspondence is that a variable name corresponds to a corresponding control and a variable type one to one.
In the above technical solution, the medical record index database stores variables extracted from each medical record in the medical record archive database according to a specified variable extraction relation table.
The similar medical record matching retrieval system comprises a medical record importing unit, a variable extracting unit, a multivariate analyzing unit and a sorting and returning unit, wherein the medical record importing unit is used for importing medical records, processing the medical records through a filter, reading personal information of a patient and storing the personal information in a temporary database file; the variable extraction unit is used for extracting and storing the variables matched with the medical records in a variable extraction list; the multivariate analysis unit is used for multivariate analysis and comprehensive summarization to obtain the similarity of each medical record; the sorting and returning unit performs descending order arrangement on the obtained similarity, the sorting result is returned to the user terminal, the user judges the returned result and selects and requests a complete medical record, thus, the imported electronic medical record or the electronic medical record stored in the local database is analyzed, the variables are extracted and summarized, the file with certain similarity to the original medical record is searched in the electronic medical record archive, and the file is provided for the user as reference, so that the retrieval efficiency is improved.
Drawings
FIG. 1 is a schematic flow chart of a matching and retrieving system for similar medical records according to the present invention;
fig. 2 is a schematic diagram of the interaction relationship between the client terminal and the database in the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The matching retrieval system for similar medical records is characterized by multivariate comprehensive analysis, particularly by using a character string type data analysis method, a numerical type data analysis method and a data comb as keys. Key variables are extracted from a certain electronic medical record imported by a user or an electronic medical record stored in a local database, multivariate comprehensive analysis is carried out, the integrated multivariate analysis result is sent to a medical record management server, multivariate searching and screening are carried out in an electronic medical record archive, and thus the medical record which is most similar to the original medical record is obtained.
Different contents of each medical record can be regarded as different variables, and a certain variable can have different values. The more variables that are equal in value between two variables, the more similar the two medical records. Some of the variables are core variables, and play the most critical role in matching medical records.
The similar medical record matching and searching system specifically comprises a medical record importing unit, a variable extracting unit, a multivariate analyzing unit and a sorting and returning unit, and the units are described in detail in combination with the flow of the similar medical record matching and searching system shown in fig. 1 as follows:
(1) a medical record importing unit:
a user introduces an electronic medical record document into the electronic medical record file management system, the introduced document enters the medical record sorting system, and a filter is adopted for processing. The filter reads the personal information of the patient from the electronic medical record document according to the mapping relation between the electronic medical record form control and the patient data preset in the system, the read data is stored in a temporary database file, and the original file is stored in a temporary database.
(2) A variable extraction unit:
the method is used for extracting the variables matched with the medical records, and mainly relates to data related to diagnosis and treatment information. A variable extraction relation table and a corresponding electronic medical record document control-variable data corresponding relation are maintained in the system in advance, and the following table shows that:
variable names | Corresponding control | Variable type |
Age | dbo. medical record front page-textbox 3.text | integer |
Gender | dbo. medical record front page-textbox 5.text | string |
C.C. | dbo. course record-textbox 3.text | string |
Imp | dbo. course record-textbox 5.text | string |
Diagnosis | dbo. course record-textbox 7.text | string |
Wherein, it is required to specify the variable type corresponding to each variable, such as numerical value, text, logic type, etc., and the method used in the multivariate analysis is different for different variable types.
According to the variable extraction relation table, the system reads the variables from the electronic medical record documents stored in the temporary database, and stores the variables in a variable extraction list, as shown in the following table:
Age | Gender | ||
C.C. | Imp | ||
Diagnosis | … |
(3) a multivariate analysis unit:
and after the user terminal of the electronic medical record file management system finishes the variable extraction, transmitting the variable extraction list to a medical record management server. The server searches in the medical record index database according to the received variable extraction list instead of directly searching in the medical record archive database.
Because the documents in the medical record database are completely stored, the medical record database contains various medical records irrelevant to diagnosis and treatment, and the content is numerous and complicated, if the medical record is directly fished in the medical record database, the great computation amount is needed. In order to reduce the amount of computation, it is necessary to extract information most closely related to medical care from medical records in a medical record archive database in advance. The specific method is that according to a specified variable extraction relation table, variables in each medical record are respectively extracted from a medical record file database and stored in a medical record index database.
The structure of the medical record index database is shown in the following table, and the main key is the hospital number:
number of hospitalization | Age | Gender | C.C. | Imp | Diagnosis | … |
And the server extracts the list according to the transmitted variable and searches in the medical record index database. The retrieval method is to carry out similarity calculation on each variable of each medical record from top to bottom, so the method is called a data comb. The data comb uses different analysis methods for different types of variables, which are respectively:
a. numerical type: the analysis of the numerical variables is calculated by the formula:
the obtained a is the similarity of the numerical variables.
b. Character string type: the analysis of string-type variables is more complex than numerical. Because the variables of the source file and the medical record variables in the medical record index database always have the same parts and also have different parts, and the logical judgment cannot be simply carried out, the analysis needs to be carried out from the perspective of character strings.
The specific method comprises the following steps: all characters in variables of a source file and all characters of corresponding variables of a medical record in a medical record index database are mixed together to form a character string pool (repeated characters do not exist in the character string pool and are combined), and then the number of characters appearing in two character strings is calculated, so that the similarity of character string type variables can be calculated by a formula as follows:
c. a logic type variable: the logic type variable analysis is simpler, if the logic judgment of the variables from the source file and the database is the same, the score is 1, otherwise, the score is 0,
after the data comb finishes the operation on the variables, the variable recording table is obtained as follows:
number of hospitalization | Age | Gender | C.C. | Imp | Diagnosis | … |
The table structure is the same as the medical record index database, but the data stored in the table structure is similarity a, b or c.
After the processing of the data comb is completed, the variables need to be summarized comprehensively. For each variable, the total similarity is calculated according to the following formula:
where weight is a weight, different weights are assigned to different variables. For example, the age is weighted 0.03, the admission diagnosis is weighted 0.1, and the weights are preset in the server. And obtaining the similarity phi of the medical records according to the formula.
(4) A sorting and returning unit:
and the server performs descending arrangement on the calculated similarity phi values and returns a plurality of items in the medical record index database before the ordering result to the user terminal. And the user judges and selects the returned result, and selects the interested medical record file, so as to ask for the complete medical record from the server. The system constructs a medical record acquisition request message and sends the medical record acquisition request message to the server, and the server calls a complete medical record in the medical record file database according to the medical record acquisition request message and returns the medical record acquisition request message to the client terminal. The interactive relationship between the client terminal and the database is shown in fig. 2.
The following is a further description taken in conjunction with a specific example:
a digestive physician needs to search similar medical records of a medical record of a digestive tract bleeding patient, and through variable extraction, the variable extraction list of the medical record is shown in the following table:
Age | 46 | Gender | for male |
C.C. | Black vomitus | Imp | Hemorrhage of digestive tract |
Diagnosis | Acute gastritis | … |
In the medical record index database, the diagnosis and treatment information of a certain medical record is shown in the following table:
Age | 44 | Gender | woman |
C.C. | Black vomitus, heartburn | Imp | Hemorrhage of digestive tract |
Diagnosis | Chronic gastritis | … |
The weight values for the different variables are shown in the following table:
Age | 0.03 | Gender | 0.03 |
C.C. | 0.15 | Imp | 0.18 |
Diagnosis | 0.32 | … |
after the data comb is performed, the similarity operation results of the medical record variables in the medical record index database are shown in the following table:
Age | 0.03 | Gender | 0.03 |
C.C. | 0.15 | Imp | 0.18 |
Diagnosis | 0.32 | … |
calculated by a multivariate analysis formula to obtain: the corresponding similarity of the medical record is as follows:
after each medical record in the medical record retrieval database completes multivariate analysis, the final descending ordering result is as follows:
number of hospitalization | φ |
136409 | 0.654 |
164923 | 0.643 |
153213 | 0.543 |
124243 | 0.476 |
And returning the result to the user terminal. After the user selects the item needing to acquire the complete medical record, the complete medical record requesting message is sent to the server, and the server calls the complete medical record in the medical record file database according to the medical record requesting message and returns the complete medical record to the client terminal.
In conclusion, compared with the prior art, the similar medical record matching and retrieving system has the following beneficial effects:
1. if all variables in each medical record in the archive are analyzed and operated, a plurality of irrelevant variables interfere the analysis result, and the accuracy is difficult to guarantee. The medical record matching retrieval system sets a plurality of variables most relevant to diagnosis and treatment, eliminates variables irrelevant to diagnosis and treatment, extracts variables from imported medical records according to a variable extraction relation table during analysis, performs multivariate analysis in a medical record index database through a data comb according to the extracted variables, analyzes different variables by dividing the variables into numerical type, character string type and logic type variables, and accordingly obtains similarity of all medical records, and is high in accuracy.
2. When a clinician or a scientific research staff needs to analyze or write a case report of a certain patient, if history similar case history needs to be found, the clinician or the scientific research staff needs to manually check and screen the case history similar case history from a database, so that the efficiency is extremely low, and a large amount of time and energy are consumed. The system searches in the medical record index database according to the data imported by the user and the extracted variables, has high operation speed, can perform multivariate analysis on medical records in the database in a very short time, and improves the working efficiency to a great extent.
3. If the electronic medical record is directly searched in the electronic medical record archive, each electronic medical record has quite a lot of data contents, and many of the data contents are irrelevant to diagnosis and treatment, and finally, the volume of data is extremely large, so that the calculation amount is large. The system extracts the variables which are most closely related to diagnosis and treatment from the electronic medical record archive in advance by newly establishing a medical record index database, so that the server does not need to carry out complicated operation in the electronic medical record archive, and directly carries out operation in the medical record index database, thereby reducing the operation amount to a certain extent.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (1)
1. A similar medical record matching retrieval system, characterized by: comprises a medical record importing unit, a variable extracting unit, a multivariate analyzing unit and a sorting and returning unit, wherein,
a medical record importing unit: the system comprises a filter, a temporary database file and an original file, wherein the filter is used for importing medical records and processing the medical records through the filter, reading personal information of a patient from an electronic medical record file and storing the personal information in the temporary database file according to a mapping relation between a preset electronic medical record form control and patient data in the system, and storing the original file in the temporary database;
a variable extraction unit: extracting variables for matching medical records, reading the variables from the electronic medical record documents stored in the temporary database and storing the variables in a variable extraction list according to a variable extraction relation table preset in the system and a corresponding electronic medical record document control-variable data corresponding relation; the variable types comprise numerical variables, character string variables and logic variables; the electronic medical record document control-variable data corresponding relation is that the variable name corresponds to the corresponding control and the variable type one by one; the medical record index database stores variables extracted from each medical record from the medical record database according to a specified variable extraction relation table;
a multivariate analysis unit: the system is used for analyzing the multivariate, after the extraction of the variable is finished, the variable extraction list is transmitted to a medical record management server, the server searches in a medical record index database according to the variable extraction list and comprehensively summarizes the variable to obtain the similarity of each medical record;
a sorting and returning unit: and performing descending order arrangement on the obtained similarity, returning the ordering result to the user terminal, and judging and checking the returned result by the user so as to ask for a complete medical record.
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 CN107193996A (en) | 2017-09-22 |
CN107193996B true 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) |
Families Citing this family (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 |
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 |
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 |
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 |
Also Published As
Publication number | Publication date |
---|---|
CN107193996A (en) | 2017-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111414393B (en) | Semantic similar case retrieval method and equipment based on medical knowledge graph | |
US9558264B2 (en) | Identifying and displaying relationships between candidate answers | |
CN107193996B (en) | Similar medical record matching and retrieving system | |
Abainia et al. | A novel robust Arabic light stemmer | |
CN104077407B (en) | A kind of intelligent data search system and method | |
CN107590128B (en) | Paper homonymy author disambiguation method based on high-confidence characteristic attribute hierarchical clustering method | |
CN101084502A (en) | System and method for digital content searching based on determined intent | |
JP6056610B2 (en) | Text information processing apparatus, text information processing method, and text information processing program | |
US8825620B1 (en) | Behavioral word segmentation for use in processing search queries | |
CN113407785B (en) | Data processing method and system based on distributed storage system | |
JP7375861B2 (en) | Related score calculation systems, methods and programs | |
CN110334343B (en) | Method and system for extracting personal privacy information in contract | |
CN109857736A (en) | The data encoding of hospital's heterogeneous system unitized method and system, equipment, medium | |
CN115497631A (en) | Clinical scientific research big data analysis system | |
CN109087688B (en) | Patient information acquisition method, apparatus, computer device and storage medium | |
CN107330111A (en) | The search method and device of domain body based on common version body | |
CN116469500A (en) | Data quality control method and system based on post-structuring of medical document | |
KR101823463B1 (en) | Apparatus for providing researcher searching service and method thereof | |
JP2013222418A (en) | Passage division method, device and program | |
JP6260678B2 (en) | Information processing apparatus, information processing method, and information processing program | |
CN110010231A (en) | A kind of data processing system and computer readable storage medium | |
Carreiras et al. | Storagebit-a metadata-aware, extensible, semantic and hierarchical database for biosignals | |
CN113505117A (en) | Data quality evaluation method, device, equipment and medium based on data indexes | |
CN113707302A (en) | Service recommendation method, device, equipment and storage medium based on associated information | |
CN111078976A (en) | Medical system crawler-based data extraction method |
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 |