CN117438079A - Method and medium for evidence-based knowledge extraction and clinical decision assistance - Google Patents
Method and medium for evidence-based knowledge extraction and clinical decision assistance Download PDFInfo
- Publication number
- CN117438079A CN117438079A CN202311743250.XA CN202311743250A CN117438079A CN 117438079 A CN117438079 A CN 117438079A CN 202311743250 A CN202311743250 A CN 202311743250A CN 117438079 A CN117438079 A CN 117438079A
- Authority
- CN
- China
- Prior art keywords
- evidence
- clinical
- knowledge
- database
- knowledge graph
- 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
- 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
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Epidemiology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application discloses a method and medium for evidence-based knowledge extraction and clinical decision assistance. The method may include: establishing and updating a evidence-based evidence database according to a first time interval; matching and integrating according to evidence-based documents of an evidence-based evidence database to construct a clinical knowledge graph; updating the clinical knowledge graph according to the updated evidence-based evidence database; and matching in a clinical knowledge graph according to the real-time clinical acquisition data to realize clinical decision reminding. The invention correlates the evidence-based document with the medical knowledge, solves the correlation between the evidence and the final application, realizes automatic update, and can meet the application requirements for decision support in different scenes.
Description
Technical Field
The invention relates to the field of clinical decision, in particular to a method and a medium for evidence-based knowledge extraction and clinical decision assistance.
Background
Evidence Based Medicine (EBM) means "Evidence-compliant medicine", which is a medical diagnosis and treatment method, emphasizing that applying studies (Evidence) of perfect design and execution optimizes decisions. Traditional medicine is based on personal experience, and doctors treat patients based on their own practical experience, the guidance of advanced physicians, textbooks, and scattered research reports in medical journals. Practice of evidence-based medicine both pays attention to individual clinical experience and emphasizes the adoption of existing, best research bases.
The clinical decision support system (Clinical Decision Support System, hereinafter referred to as CDSS) is system software which combines evidence-based evidence with computer science, and can combine clinical knowledge and basic patient information and give reasonable clinical diagnosis and treatment decision advice in a man-machine interaction mode. There are a number of types of clinical decision support today:
a decision-making system based on evidence-based knowledge base is mainly based on knowledge literature content, and because medicine is a complex discipline with fast updating iteration, the medical knowledge is updated quickly, and the system has the advantages of providing accurate and authoritative clinical knowledge, authoritative source and timely updating, and is a primary way for a clinician to acquire information. However, the system is generally based on static knowledge base reference, and cannot be deeply combined with a clinical service system to realize real-time reminding aiming at different requirements.
A decision making system for special departments or special diseases, such as a reasonable medication system, is mainly used for a specific scene, is early in system starting and relatively mature in business process, but cannot meet the core requirements of clinical decisions, particularly the support of diagnosis and treatment decision advice around the diagnosis and treatment angle, so that the decision making value is limited.
The system mainly depends on the existing medical record data of the hospital, and the data are not gold standard from the point of view of evidence, so that the accuracy of decision support cannot be ensured on data sources, the risk of clinical decision is high, and the risk of misdiagnosis is difficult to control.
A clinical medicine knowledge graph is built based on guideline isogold standards, matching deduction is carried out by combining user information and medical record information, the method can ensure the accuracy of evidence and rules, but because a medical knowledge system is numerous and complicated, the knowledge graph is built by experienced clinicians, the updating and maintenance of the knowledge graph is a difficult problem, and some clinical problems cannot be expressed by the knowledge graph.
Therefore, there is a need to develop a method and medium for evidence-based knowledge extraction and assistance in clinical decisions.
The information disclosed in the background section of the invention is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.
Disclosure of Invention
The invention provides a method and a medium for extracting evidence-based knowledge and assisting clinical decision.
In a first aspect, embodiments of the present disclosure provide a method for evidence-based knowledge extraction and clinical decision assistance, comprising:
establishing and updating a evidence-based evidence database according to a first time interval;
matching and integrating according to evidence-based documents of the evidence-based evidence database to construct a clinical knowledge graph;
updating the clinical knowledge graph according to the updated evidence-based evidence database;
and matching in the clinical knowledge graph according to the real-time clinical acquisition data to realize clinical decision reminding.
Preferably, establishing and updating the evidence-based evidence database at the first time interval includes:
formulating a rule expression, dividing evidence-based evidence into a plurality of categories, wherein each category corresponds to a group of retrieval feature words;
dividing words aiming at all document data titles, keywords and abstracts, and matching with the rule expression to obtain a matching result of each document;
indexing the documents according to the matching result, and calculating weights by adopting a TF-IDF mode;
storing the indexed documents and the weights corresponding to the indexed documents into the evidence-based evidence database;
repeating the steps with a new document after the first time interval, and updating the evidence-based evidence database.
Preferably, the matching and integrating are performed according to the evidence-based document of the evidence-based evidence database, and the constructing a clinical knowledge graph includes:
collecting medical knowledge and terms to form a medical knowledge system, and forming the knowledge system into a glossary comprising subject words, synonyms, english names, other names and classification system dimensions;
performing term indexing on evidence-based documents of the evidence-based evidence database according to medical terms;
based on the rule index and the term index of each type of medical knowledge classification system and evidence-based literature, the knowledge graph venation of all medical terms is obtained by calculating the association contribution relation of each term system.
Preferably, updating the clinical knowledge graph according to the updated evidence-based database of evidence-based evidence comprises:
aiming at the terms with the existing knowledge association relationship, if the updated terms with the association relationship reappear in the evidence-based evidence database, the association in the clinical knowledge graph is enhanced.
Preferably, updating the clinical knowledge graph according to the updated evidence-based database of evidence-based evidence comprises:
if no association is established between the existing knowledge terms, and an association relationship appears in the updated evidence-based evidence database, the knowledge association is added in the clinical knowledge graph.
Preferably, updating the clinical knowledge graph according to the updated evidence-based database of evidence-based evidence comprises:
if a new term type is added, re-indexing is carried out on all the existing documents, so that the new term and the existing knowledge graph are associated and expanded.
Preferably, updating the clinical knowledge graph according to the updated evidence-based database of evidence-based evidence comprises:
if the medical term is newly added with a different name, re-indexing the evidence-based evidence database according to the different name, and updating the co-occurrence times of the different name and other terms.
Preferably, the real-time clinical acquisition data includes basic information, complaint information, advice information, and case information of the patient.
Preferably, matching is performed in the clinical knowledge graph according to real-time clinical acquisition data, and implementing clinical decision reminding includes:
based on the real-time clinical acquisition data, matching is carried out in the clinical knowledge graph through an inference engine;
if the matching result is obtained, sequencing and reminding are carried out according to the weight of the matching result from big to small;
and if the matching result is not obtained, carrying out correlation retrieval in the evidence-based evidence database according to the knowledge term to obtain a clinical decision.
In a second aspect, embodiments of the present disclosure also provide a computer readable storage medium storing a computer program which, when executed by a processor, implements the method of evidence-based knowledge extraction and clinical decision assistance.
The beneficial effects are that:
the invention solves the combination of evidence-based evidence and knowledge, combines a clinical decision model, constructs a set of evidence-based knowledge discovery, indexing, organization, expression and application methods, not only considers the problem of data updating of evidence-based knowledge documents and the like, but also combines the evidence-based knowledge library to construct a knowledge map with evidence support, and realizes the support of clinical decision under multiple scenes through a multi-terminal construction technology, a knowledge reasoning technology and an interface technology.
The method and apparatus of the present invention have other features and advantages which will be apparent from or are set forth in detail in the accompanying drawings and the following detailed description, which are incorporated herein, and which together serve to explain certain principles of the present invention.
Drawings
The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular descriptions of exemplary embodiments of the invention as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the invention.
FIG. 1 shows a flow chart of the steps of a method of evidence-based knowledge extraction and assistance in clinical decision making, according to an embodiment of the invention.
FIG. 2 shows a schematic diagram of a knowledge organization architecture, according to an embodiment of the invention.
Fig. 3 shows a schematic diagram of a clinical decision support function architecture according to one embodiment of the invention.
Detailed Description
Preferred embodiments of the present invention will be described in more detail below. While the preferred embodiments of the present invention are described below, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.
In order to facilitate understanding of the solution and the effects of the embodiments of the present invention, two specific application examples are given below. It will be understood by those of ordinary skill in the art that the examples are for ease of understanding only and that any particular details thereof are not intended to limit the present invention in any way.
Example 1
FIG. 1 shows a flow chart of the steps of a method of evidence-based knowledge extraction and assistance in clinical decision making, according to an embodiment of the invention.
As shown in FIG. 1, the evidence-based knowledge extraction and clinical decision assistance method comprises the following steps: step 101, establishing and updating a evidence-based evidence database according to a first time interval; step 102, matching and integrating according to evidence-based documents of an evidence-based evidence database to construct a clinical knowledge graph; step 103, updating the clinical knowledge graph according to the updated evidence-based evidence database; and 104, matching in a clinical knowledge graph according to the real-time clinical acquisition data to realize clinical decision reminding.
In one example, establishing and updating the evidence-based evidence database at first time intervals includes:
formulating a rule expression, dividing evidence-based evidence into a plurality of categories, wherein each category corresponds to a group of retrieval feature words;
word segmentation is carried out on all the document data titles, keywords and abstracts, and matching is carried out on the document data titles, keywords and abstracts and regular expressions, so that a matching result of each document is obtained;
indexing the documents according to the matching result, and calculating weights by adopting a TF-IDF mode;
storing the indexed documents and the weights corresponding to the indexed documents into a evidence-based evidence database;
repeating the steps with the new document after the first time interval, and updating the evidence-based evidence database.
In one example, matching and integrating from evidence-based documents of the evidence-based evidence database, constructing the clinical knowledge graph includes:
collecting medical knowledge and terms to form a medical knowledge system, and forming the knowledge system into a glossary comprising subject words, synonyms, english names, other names and classification system dimensions;
performing term indexing on evidence-based documents of the evidence-based evidence database according to medical terms;
based on the rule index and the term index of each type of medical knowledge classification system and evidence-based literature, the knowledge graph venation of all medical terms is obtained by calculating the association contribution relation of each term system.
In one example, updating the clinical knowledge graph from the updated evidence-based database of evidence-based evidence comprises:
aiming at the terms with the existing knowledge association, if the terms with the association reappear in the updated evidence-based evidence database, the association in the clinical knowledge graph is enhanced.
In one example, updating the clinical knowledge graph from the updated evidence-based database of evidence-based evidence comprises:
if the association between the existing knowledge terms is not established, the association is increased in the clinical knowledge graph when the association relationship appears in the updated evidence-based evidence database.
In one example, updating the clinical knowledge graph from the updated evidence-based database of evidence-based evidence comprises:
if a new term type is added, re-indexing is carried out on all the existing documents, so that the new term and the existing knowledge graph are associated and expanded.
In one example, updating the clinical knowledge graph from the updated evidence-based database of evidence-based evidence comprises:
if the medical term is newly added with a different name, re-indexing the evidence-based evidence database according to the different name, and updating the co-occurrence times of the different name and other terms.
In one example, the real-time clinical acquisition data includes basic information, complaint information, order information, and case information of the patient.
In one example, matching in a clinical knowledge graph according to real-time clinical acquisition data, implementing a clinical decision reminder includes:
based on real-time clinical acquisition data, matching is carried out in a clinical knowledge graph through an inference engine;
if the matching result is obtained, sorting and reminding are carried out according to the weight of the matching result from big to small;
and if the matching result is not obtained, carrying out correlation retrieval in a evidence-based evidence database according to the knowledge term to obtain a clinical decision.
Specifically, the number of traditional medical documents is more than ten millions, hundreds of thousands of documents are updated each year, the response time cannot be guaranteed, the accuracy cannot be guaranteed by adopting a traditional database instant retrieval mode, and the documents are indexed and classified in advance by the preset indexing rules, so that data support is provided for later retrieval and recommended utilization, and meanwhile, the continuous update of evidence is guaranteed. The technical content related to the invention mainly comprises: formulating rule expression, expression conversion, matching expression, document indexing, weight calculation, result storage and the like. The process operations are as follows:
1. formulating a rule expression: according to evidence-based evidence standard, dividing evidence-based evidence into a plurality of categories, wherein each category corresponds to a group of retrieval feature words and is used for matching documents.
TABLE 1 evidence-based category names
| First class of | Second class of |
| Guide specification library | Guide |
| Guide specification library | Expert consensus |
| Guide specification library | Guide interpretation |
| Guide specification library | Medical standard |
| Guide specification library | Cause of disease of the pathogenic factors of the circulation pattern |
| Evidence-based medical library | Diagnosis of evidence-based disease |
| Evidence-based medical library | Treatment of evidence-based disease |
| Evidence-based medical library | Prognosis of evidence-based disease |
| Evidence-based medical library | System evaluation and Meta analysis |
| Evidence-based medical library | Random control study |
| Evidence-based medical library | Case control study |
| Evidence-based medical library | Queue study |
| Evidence-based medical library | Multiple case reporting |
| Library of difficult cases | Personal case reporting |
| Library of difficult cases | Case analysis |
| Library of difficult cases | Document review |
| Library of difficult cases | Misdiagnosis and mistreatment |
| Library of difficult cases | Case of evidence-based disease |
2. Expression conversion: the program splits the expression, then complements the expression understood by the machine, supplements the format if the rule has incomplete format, and prompts the rule error if the rule cannot be supplemented. And in the conversion process, a sample data mode is adopted for testing, and the test passing is marked as an effective rule.
3. Matching the expression: the method comprises the steps of word segmentation on all the titles, keywords and abstracts of the literature data, wherein word segmentation is to segment a text, a plurality of words are segmented through rules or algorithms, and each word serves as a single word or a word with the finest granularity of search. Only if the word is separated, the search can be performed, and the accuracy of the word separation is very important. In the word segmentation process, the problems of word segmentation correctness, granularity, stop words, term normalization, synonyms and the like are considered, and finally, the matching result of each paper is output.
4. Literature indexing and weight calculation: based on the matching result of each article, the weight is calculated by using a TF (word frequency) -IDF (inverse document frequency) method, which is a method for evaluating the importance degree of a word to one of the documents in a document set or a corpus. Wherein TF represents the frequency of occurrence of a phrase in the document, the higher the description the more important; DF indicates how many documents in the set of documents the word appears, the larger DF indicates that the word is common, not important, the smaller it indicates that he is more important, IDF is the inverse of DF (log), and the larger IDF indicates that the word is more important. And (3) carrying out rule expression by using TF-IDF, calculating the rule relevance of each paper, and classifying and archiving the documents in the document set.
5. And (3) storing results: and storing basic data information of the literature, index classification and rule tag data by adopting a solr technology to form a evidence-based evidence database.
Repeating the steps with the new document after the first time interval, and updating the evidence-based evidence database.
FIG. 2 shows a schematic diagram of a knowledge organization architecture, according to an embodiment of the invention.
The evidence-based document data can be related to medical knowledge to form a knowledge organization, and the traditional knowledge base or the evidence-based base is focused on a certain aspect, or evidence-based evidence is provided, or knowledge base content is provided, so that the two can not be connected, and the purpose of continuous automatic updating is achieved. The invention is based on evidence-based documents, and associates with the evidence-based documents by collecting various medical terms and knowledge, and further builds an evidence-based knowledge map based on the evidence-based documents, so that the data content can be continuously updated. The technical scheme related by the invention mainly comprises the following steps: the process of collecting medical knowledge and terms, organizing the association of the knowledge, constructing a knowledge graph, developing a terminal and the like is shown in fig. 2. The specific operation of each process is as follows:
1. collecting medical knowledge and terminology
Medical knowledge includes knowledge of diseases, medicines, examination, inspection, operation and the like, a medical knowledge system is formed by collecting knowledge data and knowledge terms in various fields, and the knowledge system is made into a glossary which comprises dimensions of subject matters, synonyms, english names, other names, classification systems and the like.
2. Knowledge organization association
Referring to the indexing mode of evidence-based evidence, medical terms are used for carrying out secondary indexing on documents, and a bidirectional maximum matching method is adopted for word segmentation. In the word segmentation process, the problems of word segmentation accuracy, granularity, stop words, term normalization, synonyms and the like are considered, such as type 2 diabetes, english name diabetes mellitus type and abbreviation name T2DM, the words are matched and uniformly classified into type 2 diabetes, after word segmentation, the words are sequenced in a reverse mode according to TF-IDF indexes of each phrase, and the first 5 terms are indexed.
3. Constructing a knowledge graph
Based on the rule index and the term index of each type of medical knowledge classification system and evidence-based document, by calculating the association contribution relation of each term system, the two terms are commonly recorded as 1 in 1 times, recorded as 2 in 2 times and the like in the same document. Finally, the knowledge graph venation of all medical terms is formed. The update condition of the knowledge graph has the following conditions:
a. for the terms of the existing knowledge association relationship, the association is enhanced based on each updating of the knowledge graph, for example, the terms (diseases) A and the terms (medicines) B are co-present for 10 times, when 1 time appears after matching in a new batch of documents, the co-occurrence frequency is added with 1, and the co-occurrence frequency is updated for 11 times.
b. If no association is established among the existing knowledge terms, a contribution relationship appears in a new batch of documents, then new knowledge association is added, if the terms (diseases) A and the terms (medicines) C do not appear in the documents before, when the new documents co-appear 1 time, then a knowledge graph association relationship between the terms A and the terms C is newly added, and the co-occurrence number is recorded as 1 time.
c. The medical terms relate to aspects, medical knowledge maps are also continuously increased, on the basis of original terms and maps, if a type of term is newly added, such as nursing operation, the whole library of all existing documents is required to be re-indexed, so that the nursing operation terms and the existing knowledge maps are associated and expanded, if the terms (diseases) A and the terms (nursing) D co-occur for N times, the association relation of the terms A and the terms C knowledge maps is newly added, and the co-occurrence number is recorded as N times.
d. Since the medical term has various different names (alias, english name, abbreviation name, trade name, etc.), if the new different name is added in the update of the medical term, the cases of a, b, etc. appear at the same time, and the whole library needs to be re-indexed, so that the co-occurrence times of the term and other terms are updated.
4. Terminal development
Based on evidence-based documents, medical knowledge and knowledge patterns, developing corresponding carrier pages by adopting a multi-terminal response technology, correspondingly displaying different page styles after a user accesses a knowledge base through different inlets, displaying a complete webpage through a browser opening webpage, displaying a mobile terminal page through a mobile phone opening webpage, opening a docking page through a docking interface, and the like.
Fig. 3 shows a schematic diagram of a clinical decision support function architecture according to one embodiment of the invention.
As shown in fig. 3, the final scenario of clinical decision is to remind in the middle of business, construct a knowledge reasoning engine based on evidence-based evidence, medical knowledge and knowledge graph, analyze the acquired clinical data and then give advice:
1. collecting service data: and interfacing with a plurality of service systems to acquire data such as basic information, main complaint information, doctor's advice information, medical history information, inspection and inspection results, and carrying out normalization processing to remove noise, wherein a hospital term system is in contrast association with the term system of the device, so that a unified term coding system is formed, for example, the coding of type 2 diabetes in the hospital term system is YY001, and corresponds to the coding WF001 of type 2 diabetes in the device.
2. Interface call: and calling different service interfaces according to different scenes, wherein the service interfaces comprise a disease diagnosis interface, a medicine order interface, an examination report interface, a medical record writing interface and the like.
3. Internal reasoning: based on the collected information data, the information data is compared with a knowledge base, a knowledge graph and evidence-based documents through an inference engine, so that a corresponding result is calculated. And directly feeding back the content with the clear reasoning result, sorting the feedback content according to the reasoning result weight (the least correlation is 0-5 minutes, the most correlation), and if the corresponding result cannot be found, searching and recommending the related suggestion from the knowledge base and evidence-based evidence by adopting a term name correlation searching mode.
4. Decision advice: and finally displaying the page of the reasoning result, wherein the user can select reminding modes according to preference, including popup window reminding, bubble reminding, closing reminding and the like.
The invention relates to a data indexing mining technology for medical evidence-based knowledge documents, and provides a method for extracting core relations from clinical documents, constructing knowledge paths and applying clinical decisions. According to evidence-based evidence grade, a literature indexing strategy is formulated, and medical literature titles are indexed regularly, so that the literature is divided into evidence types such as guidelines, expert consensus, guideline interpretation, diagnosis and treatment standards, evidence-based etiology, evidence-based diagnosis, evidence-based treatment, evidence-based prognosis, system evaluation and Meta analysis, random control research, case control research, queue research, multiple cases of report, individual case report, case analysis, literature review, misdiagnosis and misdiagnosis, evidence-based case and the like. The evidence-based literature is combined with clinical knowledge points to construct a clinical knowledge graph, a knowledge system of evidence and knowledge is formed, and a user can find corresponding clinical evidence through different scenes and entrances.
The invention can realize the butt joint with different clinical service systems, map and match the two parties of entities through the interface form, realize the unification of the unique identification, realize the reminding function under different scenes of the clinical service systems, and achieve the purpose of decision support.
According to the invention, knowledge content is extracted from massive document data, and real-time reminding is performed in a clinical service scene, so that on one hand, time and energy for workers to find data are saved, and on the other hand, diagnosis and treatment quality is improved.
Example 2
Embodiments of the present disclosure provide a computer readable storage medium storing a computer program which, when executed by a processor, implements the method of evidence-based knowledge extraction and assisting clinical decisions.
A computer-readable storage medium according to an embodiment of the present disclosure has stored thereon non-transitory computer-readable instructions. When executed by a processor, perform all or part of the steps of the methods of embodiments of the present disclosure described above.
The computer-readable storage medium described above includes, but is not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or removable hard disk), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).
It will be appreciated by persons skilled in the art that the above description of embodiments of the invention has been given for the purpose of illustrating the benefits of embodiments of the invention only and is not intended to limit embodiments of the invention to any examples given.
The foregoing description of embodiments of the invention has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described.
Claims (10)
1. A method for evidence-based knowledge extraction and clinical decision assistance, comprising:
establishing and updating a evidence-based evidence database according to a first time interval;
matching and integrating according to evidence-based documents of the evidence-based evidence database to construct a clinical knowledge graph;
updating the clinical knowledge graph according to the updated evidence-based evidence database;
and matching in the clinical knowledge graph according to the real-time clinical acquisition data to realize clinical decision reminding.
2. The method of evidence-based knowledge extraction and clinical decision assistance as claimed in claim 1, wherein establishing and updating the evidence-based evidence database at first time intervals includes:
formulating a rule expression, dividing evidence-based evidence into a plurality of categories, wherein each category corresponds to a group of retrieval feature words;
dividing words aiming at all document data titles, keywords and abstracts, and matching with the rule expression to obtain a matching result of each document;
indexing the documents according to the matching result, and calculating weights by adopting a TF-IDF mode;
storing the indexed documents and the weights corresponding to the indexed documents into the evidence-based evidence database;
repeating the steps with a new document after the first time interval, and updating the evidence-based evidence database.
3. The method for evidence-based knowledge extraction and clinical decision assistance according to claim 2, wherein the matching and integration according to the evidence-based documents of the evidence-based evidence database, the construction of the clinical knowledge graph comprises:
collecting medical knowledge and terms to form a medical knowledge system, and forming the knowledge system into a glossary comprising subject words, synonyms, english names, other names and classification system dimensions;
performing term indexing on evidence-based documents of the evidence-based evidence database according to medical terms;
based on the rule index and the term index of each type of medical knowledge classification system and evidence-based literature, the knowledge graph venation of all medical terms is obtained by calculating the association contribution relation of each term system.
4. The method of evidence-based knowledge extraction and clinical decision assistance as claimed in claim 3, wherein updating the clinical knowledge graph based on the updated evidence-based database of evidence-based evidence comprises:
aiming at the terms with the existing knowledge association relationship, if the updated terms with the association relationship reappear in the evidence-based evidence database, the association in the clinical knowledge graph is enhanced.
5. The method of evidence-based knowledge extraction and clinical decision assistance as claimed in claim 4, wherein updating the clinical knowledge graph based on the updated evidence-based database of evidence-based evidence comprises:
if no association is established between the existing knowledge terms, and an association relationship appears in the updated evidence-based evidence database, the knowledge association is added in the clinical knowledge graph.
6. The method of evidence-based knowledge extraction and clinical decision assistance as claimed in claim 5, wherein updating the clinical knowledge graph based on the updated evidence-based database of evidence-based evidence comprises:
if a new term type is added, re-indexing is carried out on all the existing documents, so that the new term and the existing knowledge graph are associated and expanded.
7. The method of evidence-based knowledge extraction and clinical decision assistance as claimed in claim 6, wherein updating the clinical knowledge graph based on the updated evidence-based database of evidence-based evidence comprises:
if the medical term is newly added with a different name, re-indexing the evidence-based evidence database according to the different name, and updating the co-occurrence times of the different name and other terms.
8. The method for evidence-based knowledge extraction and clinical decision assistance according to claim 7, wherein the real-time clinical acquisition data includes basic information, complaint information, order information and case information of the patient.
9. The method for evidence-based knowledge extraction and clinical decision assistance according to claim 8, wherein matching in the clinical knowledge graph according to real-time clinical acquisition data, implementing clinical decision reminding comprises:
based on the real-time clinical acquisition data, matching is carried out in the clinical knowledge graph through an inference engine;
if the matching result is obtained, sequencing and reminding are carried out according to the weight of the matching result from big to small;
and if the matching result is not obtained, carrying out correlation retrieval in the evidence-based evidence database according to the knowledge term to obtain a clinical decision.
10. A computer readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements the method of evidence-based knowledge extraction and clinical decision assistance as claimed in any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311743250.XA CN117438079B (en) | 2023-12-19 | 2023-12-19 | Method and medium for evidence-based knowledge extraction and clinical decision assistance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311743250.XA CN117438079B (en) | 2023-12-19 | 2023-12-19 | Method and medium for evidence-based knowledge extraction and clinical decision assistance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117438079A true CN117438079A (en) | 2024-01-23 |
| CN117438079B CN117438079B (en) | 2024-03-12 |
Family
ID=89553687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311743250.XA Active CN117438079B (en) | 2023-12-19 | 2023-12-19 | Method and medium for evidence-based knowledge extraction and clinical decision assistance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117438079B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119207829A (en) * | 2024-11-28 | 2024-12-27 | 北京万方医学信息科技有限公司 | Medical terminology mapping method, electronic device and medium based on knowledge graph |
| CN119324070A (en) * | 2024-09-29 | 2025-01-17 | 中国中医科学院中医药信息研究所 | Traditional Chinese medicine evidence-based knowledge extraction method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902871A (en) * | 2012-07-03 | 2013-01-30 | 成都中医药大学 | Clinical acupuncture evidence-based decision support system and application method thereof |
| CN106971071A (en) * | 2017-03-27 | 2017-07-21 | 为朔医学数据科技（北京）有限公司 | A kind of Clinical Decision Support Systems and method |
| CN114003734A (en) * | 2021-11-22 | 2022-02-01 | 四川大学华西医院 | Breast cancer risk factor knowledge system model, knowledge map system and construction method |
| CN114255884A (en) * | 2021-12-13 | 2022-03-29 | 首都医科大学附属北京安贞医院 | Method and device for constructing knowledge graph of hypertension drug treatment |
| WO2023061377A1 (en) * | 2021-10-13 | 2023-04-20 | 浙江大学 | Multi-center knowledge graph joint decision support method and system |
| CN116013534A (en) * | 2022-10-10 | 2023-04-25 | 睿愈(南京)数字医疗科技有限公司 | A clinical auxiliary decision-making method and system based on medical guidelines and data |
| CN117038062A (en) * | 2023-08-24 | 2023-11-10 | 北方健康医疗大数据科技有限公司 | Clinical decision support system, method and storage medium |
-
2023
- 2023-12-19 CN CN202311743250.XA patent/CN117438079B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902871A (en) * | 2012-07-03 | 2013-01-30 | 成都中医药大学 | Clinical acupuncture evidence-based decision support system and application method thereof |
| CN106971071A (en) * | 2017-03-27 | 2017-07-21 | 为朔医学数据科技（北京）有限公司 | A kind of Clinical Decision Support Systems and method |
| WO2023061377A1 (en) * | 2021-10-13 | 2023-04-20 | 浙江大学 | Multi-center knowledge graph joint decision support method and system |
| CN114003734A (en) * | 2021-11-22 | 2022-02-01 | 四川大学华西医院 | Breast cancer risk factor knowledge system model, knowledge map system and construction method |
| CN114255884A (en) * | 2021-12-13 | 2022-03-29 | 首都医科大学附属北京安贞医院 | Method and device for constructing knowledge graph of hypertension drug treatment |
| CN116013534A (en) * | 2022-10-10 | 2023-04-25 | 睿愈(南京)数字医疗科技有限公司 | A clinical auxiliary decision-making method and system based on medical guidelines and data |
| CN117038062A (en) * | 2023-08-24 | 2023-11-10 | 北方健康医疗大数据科技有限公司 | Clinical decision support system, method and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| FEI CHEN 等: "Multi-contrastive Learning Recommendation Combined with Knowledge Graph", 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2 August 2023 (2023-08-02) * |
| 夏冬云;史婷奇;陆巍;陆瑶;程建平;傅巧美;吴玲;: "压力性损伤临床决策支持系统的研发与应用", 中华护理杂志, no. 01, 15 January 2020 (2020-01-15) * |
| 李范;赵玉虹;: "从循证病例报告看我国循证临床决策过程", 中国循证医学杂志, no. 11, 25 November 2008 (2008-11-25) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119324070A (en) * | 2024-09-29 | 2025-01-17 | 中国中医科学院中医药信息研究所 | Traditional Chinese medicine evidence-based knowledge extraction method and system |
| CN119207829A (en) * | 2024-11-28 | 2024-12-27 | 北京万方医学信息科技有限公司 | Medical terminology mapping method, electronic device and medium based on knowledge graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117438079B (en) | 2024-03-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3948701B1 (en) | Artificial intelligence advisory systems and methods for vibrant constitutional guidance | |
| CN109299239B (en) | ES-based electronic medical record retrieval method | |
| US20200381087A1 (en) | Systems and methods of clinical trial evaluation | |
| Roberts et al. | State-of-the-art in biomedical literature retrieval for clinical cases: a survey of the TREC 2014 CDS track | |
| US11468363B2 (en) | Methods and systems for classification to prognostic labels using expert inputs | |
| US10593431B1 (en) | Methods and systems for causative chaining of prognostic label classifications | |
| US20200234801A1 (en) | Methods and systems for healthcare clinical trials | |
| US11937939B2 (en) | Methods and systems for utilizing diagnostics for informed vibrant constituional guidance | |
| CN117438079B (en) | Method and medium for evidence-based knowledge extraction and clinical decision assistance | |
| US20210391075A1 (en) | Medical Literature Recommender Based on Patient Health Information and User Feedback | |
| Wulff et al. | Designing an openEHR-based pipeline for extracting and standardizing unstructured clinical data using natural language processing | |
| Edinger et al. | Evaluation of clinical text segmentation to facilitate cohort retrieval | |
| US20130060793A1 (en) | Extracting information from medical documents | |
| Feng et al. | Usability of the clinical care classification system for representing nursing practice according to specialty | |
| US20240136070A1 (en) | Application-Specific Processing of a Disease-Specific Semantic Model Instance | |
| Norman | Systematic review automation methods | |
| Pakhomov et al. | Automatic quality of life prediction using electronic medical records | |
| US11810669B2 (en) | Methods and systems for generating a descriptor trail using artificial intelligence | |
| Shah-Mohammadi et al. | Contextualized Large Language Model-Based Architecture for Outcome Measure Alignment in Clinical Trials | |
| Berlanga et al. | Medical data integration and the semantic annotation of medical protocols | |
| US20250307296A1 (en) | Method and system for generating semantic response to query | |
| Suda-King et al. | Representation of Social Determinants of Health terminology in medical subject headings: impact of added terms | |
| Eisman et al. | Clinical Note Section Detection Using a Hidden Markov Model of Unified Medical Language System Semantic Types | |
| Xu et al. | Mining biomedical literature for terms related to epidemiologic exposures | |
| Kusa | Automated Eligibility Screening and its Evaluation in the Medical Domain |
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 |