CN107256222B - Electronic medical record quick retrieval system based on free word retrieval - Google Patents
Electronic medical record quick retrieval system based on free word retrieval Download PDFInfo
- Publication number
- CN107256222B CN107256222B CN201710288212.8A CN201710288212A CN107256222B CN 107256222 B CN107256222 B CN 107256222B CN 201710288212 A CN201710288212 A CN 201710288212A CN 107256222 B CN107256222 B CN 107256222B
- Authority
- CN
- China
- Prior art keywords
- retrieval
- electronic medical
- medical record
- word
- control
- 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.)
- Expired - Fee Related
Links
- 238000007781 pre-processing Methods 0.000 claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims description 15
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 5
- 238000012163 sequencing technique Methods 0.000 abstract description 3
- 238000010276 construction Methods 0.000 abstract description 2
- 208000012671 Gastrointestinal haemorrhages Diseases 0.000 description 5
- 206010020772 Hypertension Diseases 0.000 description 2
- 208000002193 Pain Diseases 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 206010019233 Headaches Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 231100000869 headache Toxicity 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
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
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2282—Tablespace storage structures; Management thereof
-
- 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
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
-
- 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 quick retrieval system of an electronic medical record based on free word retrieval, which comprises a control attribute definition unit, a retrieval structured preprocessing unit and an intelligent sequencing unit, wherein the attribute definition, the retrieval structured preprocessing method and the construction of a three-dimensional index database are carried out on the electronic medical record, and one electronic medical record passes through a retrieval structured preprocessing pipeline so as to construct a three-dimensional index database, so that a user can simultaneously limit retrieval words and retrieval fields, and the system can quickly obtain a retrieval result through the query of the three-dimensional database. Therefore, the aim of quickly searching is achieved, the target electronic medical record meeting the searching requirement is found, and the use convenience is improved.
Description
Technical Field
The invention relates to the technical field of medical treatment, in particular to an electronic medical record quick retrieval system based on free word retrieval.
Background
The free word retrieval and the subject word retrieval are a pair of opposite concepts, the free word retrieval means that a user inputs any word to be retrieved, and the system carries out retrieval in a limited field; the term searching refers to that a user searches for a specific term according to a term table, and performs the assembly between the term and an auxiliary term, thereby searching for an object to be searched in a targeted manner. The two have a length, can realize complementary functions, and have low threshold for free word retrieval, convenient and quick retrieval, less restriction conditions on retrieval words, but the recall ratio and precision ratio are inferior to those of subject word retrieval; the subject word retrieval has high recall ratio and precision ratio, but has the defects of higher retrieval threshold and lower use convenience degree than the free word retrieval.
At present, a retrieval method applied to the field of electronic medical record retrieval is not developed, retrieval is often referred by people, and the efficiency is extremely low. Therefore, it is urgently needed to develop two complementary fast free word retrieval and subject word retrieval systems so as to realize the retrieval function of the electronic medical record system.
Disclosure of Invention
Aiming at the problems in the background art, the invention aims to provide an electronic medical record quick retrieval system based on free word retrieval, which forms one of the main retrieval functions of an electronic medical record system, so that a user can quickly retrieve according to a target word to be retrieved and a selected target field, and then a target electronic medical record meeting the retrieval requirement is found.
The technical scheme of the invention is realized as follows: a quick retrieval system of electronic medical record based on free word retrieval comprises a control attribute definition unit, a retrieval structured preprocessing unit and an intelligent sequencing unit, wherein,
the attribute definition unit: the system is used for defining SaveForSearch attributes of each control in the electronic medical record form and specifying whether each control needs to be subjected to retrieval structuralization preprocessing or not; if the SaveForSearch attribute is true, performing retrieval structuring preprocessing; if the SaveForSearch attribute is false, no search structuring preprocessing is performed;
the retrieval structuring preprocessing unit: firstly, judging each control in a form, and if the SaveForSearch attribute of one control is false, entering the judgment of the next control; if the SaveForSearch attribute of one control is true, performing word segmentation processing on the data part in the control, wherein the word segmentation processing is performed by adopting a word segmentation table, if a word is distinguished according to the word segmentation table, counting the number of the word of the electronic medical record case number under the dimension corresponding to the field of the three-dimensional index database by one, and so on until all the controls are judged to be finished, and obtaining a three-dimensional index database;
the intelligent sorting unit: after receiving a user retrieval request, the system matches the dimensionality corresponding to the field defined by the user in the three-dimensional index database to find a column corresponding to the retrieval word, and then the relevance of a unit with the count of 0 in the column is marked as 0; for the cells with the counts in the columns not equal to 0, the counts are recorded as k, the maximum value of the count value is recorded as n, the conversion of the correlation is carried out, and the correlation calculation formula is as follows:
then, summing the χ values corresponding to each case number under each dimensionality, and carrying out descending order arrangement, wherein the arrangement of the case numbers from top to bottom is the search result order presented to the user; if the number of the free words is multiple, the correlation χ values of the multiple free words are summed and arranged in a descending order, and the arrangement from the top to the bottom of the case numbers is the search result order presented to the user.
In the technical scheme, in the electronic medical record form, data of different fields are stored in different controls, data of a chief complaint is stored in a Richbox control corresponding to the chief complaint, and a name is stored in a Textbox control corresponding to the name.
In the above technical solution, the three-dimensional index database has three dimensions, a first dimension is a field; the second dimension is the case number; the third dimension is words.
In the above technical solution, the first dimension field includes a chief complaint, a current medical history, a life history, and a family history; the case number of the second dimension is formed by the case numbers of all medical records; and the third dimension of the words obtains the occurrence times of certain words in different fields of different words under different electronic medical records through word segmentation statistics.
The invention relates to an electronic medical record quick retrieval system based on free word retrieval, which is characterized in that a retrieval structuralized preprocessing method and a three-dimensional index database are established for an electronic medical record, and the electronic medical record passes through a retrieval structuralized preprocessing pipeline so as to establish the three-dimensional index database, so that a user can limit retrieval words and retrieval fields at the same time, and the system can quickly obtain a retrieval result by querying the three-dimensional database. Therefore, the aim of quickly searching is achieved, the target electronic medical record meeting the searching requirement is found, and the use convenience is improved.
Drawings
FIG. 1 is a flow chart of the structured pre-processing of control attribute definition and retrieval in the retrieval system of the present invention;
FIG. 2 is a flowchart of the intelligent ranking in the search system of 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 invention relates to a quick retrieval system of an electronic medical record based on free word retrieval, which comprises a control attribute definition unit, a retrieval structured preprocessing unit and an intelligent sorting unit, wherein the following is a detailed description of the specific execution of the units.
(1) A control attribute definition unit:
in the electronic medical record form, data of different fields are stored in different controls, for example, data of a chief complaint is stored in a Richbox control corresponding to the chief complaint, and a name is stored in a Textbox control corresponding to the name. The retrieval structuralization preprocessing is carried out on the premise that fields needing to be structuralized are defined, data with retrieval value are reserved, and redundant information is abandoned. Therefore, the definition of the SaveForSearch attribute needs to be carried out on each control in the electronic medical record form, and if the SaveForSearch attribute is true, the control needs to be sorted into a three-dimensional index database in retrieval structured preprocessing; if the SaveForSearch attribute is false, no search structuring preprocessing is performed.
In this step, the developer needs to define SaveForSearch attributes for each control in an electronic medical record form template, and specify whether each control needs to perform retrieval structuring preprocessing.
(2) A retrieval structured preprocessing unit:
when a new electronic medical record of a hospital is generated, the electronic medical record system firstly stores the whole electronic medical record document into the database as a whole for service. And then the electronic medical record form is sent to a retrieval structured preprocessing pipeline for processing.
For the electronic medical record form, firstly, each control in the form needs to be judged. If the SaveForSearch attribute of one control is false, entering the judgment of the next control; if the SaveForSearch attribute of one control is true, performing word segmentation processing on the data part in the control, wherein the word segmentation processing is performed by adopting a word segmentation table, and if a word is distinguished according to the word segmentation table, counting is increased by one under the word of the electronic medical record case number in the dimension corresponding to the field of the three-dimensional index database. And so on until all the controls are judged.
The judgment significance of the SaveForSearch attribute is that there are many redundant information in an electronic medical record form, such as recording time, Label of its expression function, etc., and there is no contribution in the retrieval process. The provision of the SaveForSearch attribute greatly simplifies the search field, optimizes the storage space, and thus retains the field with the highest search value.
The result of the retrieval of the structured pre-processing is a three-dimensional index database. Compared with the index database established by most search engines, the index database has three dimensions and is suitable for electronic medical record search. The first dimension is the fields, including chief complaints, current medical history, life history, family history, etc.; the second dimension is the case number, which is formed by the case numbers of all the medical records; the third dimension is a word, and the occurrence times of certain words in different fields of different electronic medical records of different words are obtained through word segmentation statistics. An example of the construction of the three-dimensional index database is as follows:
a first layer: the main complaints are:
patient's case number | Headache pain | Pain (due to cold or dampness) | Debilitation | Generate heat | |
2017031211 | |||||
2017031212 | |||||
2017031213 |
A second layer: history of present disease
Patient's case number | Hypertension (hypertension) | Diabetes mellitus | Coronary heart disease | ||
2017031211 | |||||
2017031212 | |||||
2017031213 |
And a third layer: history of life
Patient's case number | |||||
2017031211 | |||||
2017031212 | |||||
2017031213 |
The above-described flow is shown in FIG. 1.
(3) An intelligent sequencing unit:
the user inputs one or more free words to be searched in the search box, selects one or more fields to be searched, and sends a search request to the system. Assuming that a free word is input by a user, after receiving a retrieval request, the system matches the dimension corresponding to the field defined by the user in the three-dimensional index database to find the column corresponding to the retrieval word. Subsequently, for the cell whose count is 0 in the column, the correlation is recorded as 0; for the cells with the counts in the columns not equal to 0, the counts are recorded as k, the maximum value of the count value is recorded as n, the conversion of the correlation is carried out, and the correlation calculation formula is as follows:
and then, summing the χ values corresponding to the cases under all dimensions, and performing descending arrangement, wherein the arrangement from top to bottom of the cases is the search result sequence presented to the user.
If the number of the free words is multiple, the correlation χ values of the multiple free words are summed and arranged in a descending order, and the arrangement from the top to the bottom of the case numbers is the search result order presented to the user. The flow is shown in fig. 2.
The following is further described in connection with a specific example:
the user enters the free words in the search box as follows: gastrointestinal hemorrhage, the selection search field is: the present medical history, the past medical history and the family history.
According to the search term 'gastrointestinal hemorrhage', the system calls three-dimensional data in the three-dimensional index database, namely the current medical history, the past medical history and the family history. Finding the corresponding column of the search term in three dimensions, and calculating the corresponding correlation degrees, which are respectively:
the current medical history:
patient's case number | Hemorrhage of digestive tract | Degree of correlation |
2017031211 | 1 | 1.72 |
2017031212 | 0 | 0 |
2017031213 | 1 | 1.72 |
Maximum value | 2 | 2 |
The past medical history:
family history:
patient's case number | Hemorrhage of digestive tract | |
2017031211 | 1 | 1.60 |
2017031212 | 1 | 1.60 |
2017031213 | 0 | 0 |
Maximum value | 3 | 2 |
The three fields are summed and ordered to yield:
patient's case number | Hemorrhage of digestive tract | Degree of correlation |
2017031211 | 1 | 5.19 |
2017031213 | 0 | 3.44 |
2017031212 | 1 | 3.13 |
The search result ordering is as shown in the above table.
The electronic medical record quick retrieval system based on the free word retrieval has the following beneficial effects:
1. the topic word retrieval puts higher requirements on users, and the users need to have certain understanding on a medical topic word list (MeSH) and master a grouping method among topic words, so the use threshold is higher. The free word search does not require the user to have basic knowledge about the search object, the user only needs to input the object to be searched into the search frame and select whether to limit the search field, so that a search request can be completed, and the use threshold is not high.
2. Although the recall ratio and the precision ratio of the subject term search are high, when a user searches, the user needs to match the subject terms of the searched object and then assemble the auxiliary subject terms after matching the subject terms, so that the search difficulty is high, and the use is not convenient enough. The free word retrieval only needs to input a retrieval object and select whether to limit a retrieval field, complex retrieval word assembly is not needed, and a user is concise and rapid to use. Therefore, in reality, the subject word search and the free word search are often used in combination according to the actual needs.
3. The system adopts the logarithmic function model to calculate the relevance of the retrieved medical records, and has the advantages that the relevance is at least 1 as long as the retrieval words appear in the field; and with the increase of the occurrence frequency of the search word, the relevancy also increases, but the maximum is not more than 2, so that the situation that a certain field occupies a search dominant position due to repeated occurrence of the search word is avoided. And summing the relevancy of different fields of different search terms to obtain relevancy sequences of different medical records.
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 (3)
1. A quick retrieval system of electronic medical record based on free word retrieval is characterized in that: comprises a control attribute definition unit, a retrieval structured preprocessing unit and an intelligent sorting unit, wherein,
the attribute definition unit: the system is used for defining SaveForSearch attributes of each control in the electronic medical record form and specifying whether each control needs to be subjected to retrieval structuralization preprocessing or not; if the SaveForSearch attribute is true, performing retrieval structuring preprocessing; if the SaveForSearch attribute is false, no search structuring preprocessing is performed;
the retrieval structuring preprocessing unit: firstly, judging each control in a form, and if the SaveForSearch attribute of one control is false, entering the judgment of the next control; if the SaveForSearch attribute of one control is true, performing word segmentation processing on the data part in the control, wherein the word segmentation processing is performed by adopting a word segmentation table, if a word is distinguished according to the word segmentation table, counting the number of the word of the electronic medical record case number under the dimension corresponding to the field of the three-dimensional index database by one, and so on until all the controls are judged to be finished, and obtaining a three-dimensional index database; wherein the three-dimensional index database has three dimensions, a first dimension being a field; the second dimension is the case number; the third dimension is words;
the intelligent sorting unit: after receiving a user retrieval request, the system matches the dimensionality corresponding to the field defined by the user in the three-dimensional index database to find a column corresponding to a retrieval word, and then the relevance of a unit with the count of 0 in the column is marked as 0; for the cells with the counts in the columns not equal to 0, the counts are recorded as k, the maximum value of the count value is recorded as n, the conversion of the correlation is carried out, and the correlation calculation formula is as follows:
then, corresponding each case number under each dimensionSumming the values, and arranging the values in a descending order, wherein the arrangement of the case numbers from top to bottom is the search result sequence presented to the user; if there are multiple free words, the relevance of multiple free words is determinedThe values are summed and arranged in descending order, and the arrangement from top to bottom of the case numbers is the search result ordering presented to the user.
2. The system for rapidly retrieving the electronic medical record based on the free word retrieval as claimed in claim 1, wherein: in the electronic medical record form, data of different fields are stored in different controls, data of a chief complaint is stored in a Richbox control corresponding to the chief complaint, and a name is stored in a Textbox control corresponding to the name.
3. The system for rapidly retrieving the electronic medical record based on the free word retrieval as claimed in claim 1, wherein: the fields of the first dimension comprise a chief complaint, a current medical history, a life history and a family history; the case number of the second dimension is formed by the case numbers of all medical records; and the third dimension of the words obtains the occurrence times of certain words in different fields of different words under different electronic medical records through word segmentation statistics.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710288212.8A CN107256222B (en) | 2017-04-27 | 2017-04-27 | Electronic medical record quick retrieval system based on free word retrieval |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710288212.8A CN107256222B (en) | 2017-04-27 | 2017-04-27 | Electronic medical record quick retrieval system based on free word retrieval |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107256222A CN107256222A (en) | 2017-10-17 |
CN107256222B true CN107256222B (en) | 2020-12-01 |
Family
ID=60027198
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710288212.8A Expired - Fee Related CN107256222B (en) | 2017-04-27 | 2017-04-27 | Electronic medical record quick retrieval system based on free word retrieval |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107256222B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107273405B (en) * | 2017-04-27 | 2020-12-18 | 广州慧扬健康科技有限公司 | Intelligent retrieval system of electronic medical record files based on MeSH table |
CN107818169B (en) * | 2017-11-13 | 2021-09-07 | 医渡云(北京)技术有限公司 | Electronic medical record retrieval and storage method and device, storage medium and electronic terminal |
CN109473178B (en) * | 2018-11-12 | 2022-04-01 | 北京懿医云科技有限公司 | Method, system, device and storage medium for medical data integration |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101034410A (en) * | 2007-04-16 | 2007-09-12 | 浙江大学 | Medical treatment case index cast and electronic medical record system based on same |
CN101369339A (en) * | 2007-08-17 | 2009-02-18 | 上海市闵行区卫生局 | Community sanitation service information system |
US20090112859A1 (en) * | 2007-10-25 | 2009-04-30 | Dehlinger Peter J | Citation-based information retrieval system and method |
CN101520784B (en) * | 2008-02-29 | 2011-09-28 | 富士通株式会社 | Information issuing system and information issuing method |
CN101340448A (en) * | 2008-06-10 | 2009-01-07 | 中山爱科数字科技有限公司 | Medical data integration scheme oriented to digital family |
CN102043811A (en) * | 2009-10-13 | 2011-05-04 | 北京大学 | Method and system for evaluating medical information |
CN103605665B (en) * | 2013-10-24 | 2017-01-11 | 杭州电子科技大学 | Keyword based evaluation expert intelligent search and recommendation method |
-
2017
- 2017-04-27 CN CN201710288212.8A patent/CN107256222B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
---|---|
CN107256222A (en) | 2017-10-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yuan et al. | Index-based densest clique percolation community search in networks | |
CN105893597B (en) | Similar medical record retrieval method and system | |
Chen et al. | Senbazuru: A prototype spreadsheet database management system | |
CN108897761A (en) | A kind of clustering storage method and device | |
CN107256222B (en) | Electronic medical record quick retrieval system based on free word retrieval | |
CN113407785B (en) | Data processing method and system based on distributed storage system | |
WO2021179481A1 (en) | Cold start method and apparatus for personalizing and pushing data content, device and storage medium | |
Adamu et al. | A survey on big data indexing strategies | |
CN104391908B (en) | Multiple key indexing means based on local sensitivity Hash on a kind of figure | |
CN111213212A (en) | System and method for generating query suggestions reflecting groups | |
CN107273405B (en) | Intelligent retrieval system of electronic medical record files based on MeSH table | |
KR101341816B1 (en) | System and method for extracting analogous queries | |
CN109472282A (en) | A kind of depth image hash method based on few training sample | |
CN110580255A (en) | method and system for storing and retrieving data | |
CN108733745A (en) | A kind of enquiry expanding method based on medical knowledge | |
Piech et al. | A new approach to storing dynamic data in relational databases using JSON | |
CN102915381A (en) | Multi-dimensional semantic based visualized network retrieval rendering system and rendering control method | |
Zhang et al. | Series feature aggregation for content-based image retrieval | |
WO2017065891A1 (en) | Automated join detection | |
US20150012563A1 (en) | Data mining using associative matrices | |
Yan et al. | The self-adaptive integrated algorithm on time-sensitive cluster evolution tracking | |
JP7213890B2 (en) | Accelerated large-scale similarity computation | |
JP2000112973A (en) | Space index method and medium storing space index processing program | |
Arumugam et al. | Enhancing the web image search results through direct and indirect relevance model | |
CN101984424A (en) | Mass inter-media index 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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201201 |