CN116344041B - Traditional Chinese medicine auxiliary decision-making system based on knowledge graph - Google Patents

Traditional Chinese medicine auxiliary decision-making system based on knowledge graph Download PDF

Info

Publication number
CN116344041B
CN116344041B CN202310593821.XA CN202310593821A CN116344041B CN 116344041 B CN116344041 B CN 116344041B CN 202310593821 A CN202310593821 A CN 202310593821A CN 116344041 B CN116344041 B CN 116344041B
Authority
CN
China
Prior art keywords
keyword
knowledge
information
image
unit
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
Application number
CN202310593821.XA
Other languages
Chinese (zh)
Other versions
CN116344041A (en
Inventor
顾扬
贾冬梅
金清
刘鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Daan Health Technology Beijing Co ltd
Original Assignee
Daan Health Technology Beijing Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Daan Health Technology Beijing Co ltd filed Critical Daan Health Technology Beijing Co ltd
Priority to CN202310593821.XA priority Critical patent/CN116344041B/en
Publication of CN116344041A publication Critical patent/CN116344041A/en
Application granted granted Critical
Publication of CN116344041B publication Critical patent/CN116344041B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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; CALCULATING OR 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
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • 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)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Primary Health Care (AREA)
  • Pathology (AREA)
  • Epidemiology (AREA)
  • Artificial Intelligence (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention relates to the technical field of intelligent medical treatment, in particular to a traditional Chinese medicine auxiliary decision-making system based on a knowledge graph, which comprises the following components: a database; the identification unit comprises a text information extraction module and an image information extraction module which are connected with the database; the effective information judging unit is respectively connected with the text information extracting module and the image information extracting module; the information integration unit is connected with the identification unit and used for integrating each keyword set; the knowledge retrieval unit is respectively connected with the database, the identification unit and the information integration unit and is used for retrieving medical knowledge in the database according to the keyword set; and the interaction unit is connected with the identification unit and is used for acquiring the information of the external information source and transmitting the search result to the external information source. The invention can realize the guidance of information correction of the external information source so as to increase the information credibility of the external information source.

Description

Traditional Chinese medicine auxiliary decision-making system based on knowledge graph
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to a traditional Chinese medicine auxiliary decision-making system based on a knowledge graph.
Background
The traditional Chinese medicine auxiliary decision-making system is a system for carrying out auxiliary decision making on traditional Chinese medicine diagnosis and treatment by combining a calculation method and a technology and combining a traditional Chinese medicine medical thought and method, and is characterized in that based on a traditional Chinese medicine medical thought and method and a modern computer technology and an artificial intelligence technology, medical knowledge and decision-making models based on methods such as rules, cases and models are established to realize auxiliary decision making on doctors, and the artificial intelligence technology is an important support of the traditional Chinese medicine auxiliary decision-making system, such as a rule-based inference engine, a case-based inference engine, a neural network-based inference engine, a machine learning-based inference engine and the like, and can extract medical knowledge according to information input by users.
Chinese patent CN112420191a discloses a traditional chinese medical auxiliary decision making system comprising: the patient information acquisition module is used for receiving patient information of a patient in the traditional Chinese medical records; the symptom scoring module is connected with the patient information acquisition module and is used for scoring the symptom body according to a preset rule by adopting a dynamic weighting method, obtaining and outputting the weight of each index and carrying out weighted summation on the indexes to calculate the weight of each symptom corresponding to the symptom body; the auxiliary decision module is connected with the symptom scoring module, trains a symptom-to-medicine model based on the SeqCAN algorithm, outputs an optimal traditional Chinese medicine group for each symptom corresponding to a symptom body as an auxiliary decision result according to a preset rule according to the weight of each symptom corresponding to the symptom body, and can improve the traditional Chinese medicine decision efficiency. However, there is a problem that the confidence of the patient information is insufficient due to the influence of the main objective condition of the user, so that the reference of the traditional Chinese medicine decision result is low.
Disclosure of Invention
Therefore, the invention provides a traditional Chinese medicine auxiliary decision-making system based on a knowledge graph, which can solve the problem of lower referential property of medical knowledge acquired according to an external information source due to insufficient confidence of the external information source influenced by main objective conditions.
In order to achieve the above object, the present invention provides a knowledge graph-based auxiliary decision-making system for traditional Chinese medicine, comprising:
a database for storing knowledge-graph based medical knowledge, an effective image element training set, and an effective keyword training set;
the identification unit comprises a text information extraction module connected with the database for extracting text keywords in text information from an external information source according to the effective keyword training set and an image information extraction module connected with the database for identifying and extracting effective information elements in image feedback information according to the effective image element training set and sequentially converting each effective information element into corresponding image keywords;
the information integration unit is connected with the identification unit and is used for integrating the keywords to generate an integrated keyword set meeting the corresponding conditions;
The knowledge retrieval unit is respectively connected with the database, the identification unit and the information integration unit and is used for retrieving medical knowledge in the database according to the keywords or the integrated keyword set to obtain a plurality of retrieval results;
an effective information judging unit connected with each module in the identifying unit and used for judging whether the text keyword is effective or not according to the number of the search results obtained by the knowledge searching unit based on the text keyword and the category number of the search results, and obtaining corresponding keywords based on each search result obtained by the text keyword and secondary search of the effective keyword training set when judging that the text keyword is invalid so as to establish a reference set;
the interaction unit is connected with the identification unit and is used for acquiring text information of an external information source, transmitting the text information to the text information extraction module, transmitting keywords belonging to the reference set to the external information source to acquire feedback information of the external information source on the keywords, transmitting an image information acquisition request to the external information source to acquire image feedback information of the external information source, and transmitting the image feedback information to the image information extraction module; the interaction unit is also used for transmitting each search result to an external information source and labeling the recommendation priority of each transmitted search result.
Further, the knowledge retrieval unit marks the set of text keywords as a text keyword set a, and the knowledge retrieval unit retrieves medical knowledge in the database according to the text keyword set a, wherein:
the effective information judging unit judges that the keywords in the text keyword set A are invalid under a first judging condition;
the effective information judging unit judges that the keywords in the text keyword set A are effective under a second judging condition, the knowledge searching unit sets recommendation priorities of the search results acquired based on the text keyword set A according to the matching degree of the search results acquired based on the text keyword set A and the text keyword set A, and the interaction unit transmits the search results acquired based on the text keyword set A to an external information source and marks the recommendation priorities of the transmitted search results;
the first judging condition is that the number of the search results obtained based on the text keyword set A is larger than a maximum threshold value of the number of the search results, or the number of the search results is smaller than a minimum threshold value of the number of the search results, or the number of the search result categories is larger than a maximum threshold value of the number of the search result categories; the second judging condition is that the number of the search results obtained based on the text keyword set A is larger than or equal to a minimum threshold value of the number of the search results, smaller than or equal to a maximum threshold value of the number of the search results, and smaller than or equal to a maximum threshold value of the number of the search results categories.
Further, the interaction unit sends an image information acquisition request to an external information source under the condition of a first search result;
the first search result condition is that the knowledge search unit is based on the fact that the number of search results obtained by the text keyword set A is larger than a maximum threshold value of the number of search results and the number of keywords in the text keyword set A is larger than or equal to a minimum threshold value of the number of keywords, or based on the fact that the number of search result categories obtained by the text keyword set A is larger than the maximum threshold value of the number of search result categories and the number of keywords in the text keyword set A is larger than or equal to the minimum threshold value of the number of keywords.
Further, the knowledge retrieval unit calculates the matching degree of the text keyword set A and each retrieval result obtained based on the text keyword set A under the condition of a second retrieval result, and takes a plurality of retrieval results with the matching degree larger than or equal to a preset matching degree as a reference information group, the effective information judgment unit extracts keywords which belong to the effective keyword training set and do not belong to the text keyword set A in the reference information group as a reference set, and sets the priority for each keyword in the reference set according to the descending order of the number of each keyword in the reference set in the reference information group,
The second search result condition is that the knowledge search unit obtains the number of search results based on the text keyword set A, wherein the number of the search results is larger than a maximum threshold value of the number of the search results, and the number of keywords in the text keyword set A is smaller than a minimum threshold value of the number of the keywords, or the number of the search result categories is larger than a maximum threshold value of the number of the search result categories, and the number of the keywords in the text keyword set A is smaller than a minimum threshold value of the number of the keywords;
the effective information judging unit sets each keyword in the reference set as a plurality of priority reference keywords according to the priority, and when a plurality of keywords exist in the reference set and the number of the keywords in the reference information set is the same, the effective information judging unit sets the priority of each keyword according to an initial ordering rule.
Further, the interaction unit sequentially transmits the keywords in the reference set to an external information source according to the sequence of a plurality of priority reference keywords to acquire feedback information of the external information source on the keywords in the reference set, the information integration unit selects corresponding keywords from the reference set according to the feedback information of the external information source and adds the keywords to the text keyword set A to form a supplementary keyword set B, the knowledge retrieval unit retrieves medical knowledge in the database according to the supplementary keyword set B, the interaction unit judges whether to send an image information acquisition request to the external information source according to the number of retrieval results acquired by the knowledge retrieval unit based on the supplementary keyword set B,
And if the number of the search results acquired by the knowledge search unit based on the supplementary keyword set B is larger than the maximum threshold value of the number of the search results, the interaction unit judges that an image information acquisition request is sent to an external information source.
Further, the image information extraction module respectively converts the effective information elements identified and extracted according to the image feedback information into corresponding keywords under the first interaction condition, marks each keyword as an image keyword set C, the knowledge retrieval unit obtains the confidence level of the image keyword set C based on the coincidence ratio of the image retrieval result obtained by the image keyword set C and the retrieval result obtained by the supplementary keyword set B, wherein,
the effective information identification module acquires the confidence coefficient of the image keyword set C as a first confidence coefficient under the first coincidence rate condition; the effective information identification module acquires the confidence coefficient of the image keyword set C as a second confidence coefficient under the second coincidence rate condition;
the first interaction condition is that the interaction unit sends an image information acquisition request to an external information source and acquires image feedback information of the external information source, the first coincidence rate condition is that the coincidence rate of a search result acquired based on the image keyword set C and a search result acquired based on the supplementary keyword set B is smaller than a preset search result coincidence rate, and the second coincidence rate condition is that the coincidence rate of the search result acquired based on the image keyword set C and the search result acquired based on the supplementary keyword set B is larger than or equal to the preset search result coincidence rate.
Further, the information integrating unit integrates each keyword in the supplemental keyword set B with each keyword in the image keyword set C according to the confidence of the image keyword set C, wherein,
the information integration unit sets the union of the supplemental keyword set B and the image keyword set C as a first keyword union D1 under the first confidence level condition;
the information integration unit sets the union of the supplementary keyword set B and the image keyword set C as a second keyword union D2 under the condition of second confidence, sets the intersection of the supplementary keyword set B and the image keyword set C as a keyword intersection E, and sets the relative complement of the keyword intersection E in the second keyword union D2 as a keyword complement F;
the first confidence level condition is that the confidence level of the image keyword set C is equal to 1, and the second confidence level condition is that the confidence level of the image keyword set C is not equal to 1.
Further, the knowledge retrieval unit sets recommendation priorities of the retrieval results obtained based on the first keyword union D1 according to matching degrees of the retrieval results obtained based on the first keyword union D1 and the first keyword union D1 under a first retrieval condition, and the interaction unit transmits the retrieval results obtained based on the first keyword union D1 to an external information source and marks the recommendation priorities of the transmitted retrieval results;
The first search condition is that the knowledge search unit searches medical knowledge in the database according to the first keyword union set D1.
Further, the knowledge retrieval unit acquires a union of the first retrieval result set, the second retrieval result set and the third retrieval result set as an alternative retrieval result set under the second retrieval condition;
the second search condition is that the knowledge search unit searches medical knowledge in the database according to the second keyword union set D2 to obtain the first search result set, searches medical knowledge in the database according to the keyword intersection set E to obtain the second search result set, and searches medical knowledge in the database according to the keyword complement set F to obtain the third search result set;
the knowledge retrieval unit screens out retrieval results with the matching degree with the second keyword union set D2 being greater than or equal to a preset matching degree from the candidate retrieval result set to serve as a reference medical knowledge set, recommendation priorities for the retrieval results in the reference medical knowledge set are respectively set according to the matching degree of the retrieval results in the reference medical knowledge set and the second keyword union set D2, and the interaction unit transmits the retrieval results in the reference medical knowledge set to an external information source and marks the recommendation priorities of the transmitted retrieval results.
Further, the image information extraction module respectively converts effective information elements which are identified and extracted according to image feedback information into corresponding keywords under a third search result condition, and marks each keyword as an image keyword set C, the knowledge search unit searches medical knowledge in the database according to a keyword union H obtained by the image keyword set C and the text keyword set A, and sets recommendation priority of each search result obtained based on the keyword union H according to matching degree of each search result obtained based on the keyword union H with the keyword union H, and the interaction unit transmits each search result obtained based on the keyword union H to an external information source and marks the recommendation priority of each transmitted search result;
the keyword union H is a union of the image keyword set C and the text keyword set A integrated by the information integration unit, and the image keyword set C comprises a plurality of keywords extracted by the image information extraction module according to image feedback information;
the third search result condition is that the number of search results obtained by the knowledge search unit is smaller than a minimum threshold value of the number of search results, and the interaction unit sends an image information obtaining request to an external information source to obtain image feedback information.
Compared with the prior art, the invention has the beneficial effects that the database is arranged, the stored knowledge graph can be updated in real time, the medical knowledge after the update of the knowledge graph is increased in real time, and each stored training set can be updated in real time according to an external information source, so that the accuracy rate of the medical knowledge finally obtained is higher; the invention is provided with the identification unit and the effective information judgment unit, can extract, identify and match keywords of the external information source and the search result, and realize the guidance of information correction of the external information source so as to increase the information reliability of the external information source; the invention sets the knowledge retrieval unit, can retrieve the medical knowledge according to the keyword of the external information source, and set the recommendation priority of each retrieval result according to the matching degree of the keyword and each retrieval result, the consulting property of the recommendation information of the external information source is more objective; the information integration unit is arranged, so that keywords provided by the external information source at different stages can be integrated, and the accuracy of the keywords of the external information source is improved.
In particular, the invention sets that the number of the search results or the number of the search result categories of the medical knowledge searched according to the keywords is too large or the number of the search results is too small, the current keywords are judged to be invalid, when the number of the search results or the number of the search result categories is too large, the current keywords are too fuzzy and have no characteristics, the acquired medical knowledge has lower referential property, and when the number of the search results is too small, the current keywords are inaccurate, so that the accurate medical knowledge cannot be searched.
Particularly, when the number of the search results is too large or the number of the search result categories is too large and the number of the keywords is enough, the images provided by the external information sources are adopted to carry out auxiliary screening on medical knowledge, and the traditional Chinese medicine inquiry comprises 'inspection inquiry', the inquiry is converted into text information as a leading part, the inspection inquiry is converted into image information as an auxiliary judging condition, and the credibility of the external information sources can be increased.
In particular, when the number of the search results is too large or the number of the search result categories is too large and the number of the keywords is small, the external information source can further provide effective information for the system by extracting the characteristic keywords of each current search result as an information group for further confirming the external information source, so that further screening of medical knowledge is realized, and recommended medical knowledge meeting the current needs of the external information source is further obtained.
In particular, the image information provided by the external information source is influenced by subjective conditions of the information source and objective conditions of the environment, the validity and the authenticity of the information contained in the image information cannot be guaranteed, the reliability of the image information can be roughly judged by comparing the effective information elements in the image information with text keywords, and the influence degree of the image information on a final search result is determined according to the reliability degree of the image information.
Particularly, when the confidence of the image information is high, the effective information elements contained in the image can be used as conditions for screening search results, so that the effective information elements are integrated into keywords of the text information to be used as screening conditions for searching medical knowledge, when the confidence of the image information is low, the reliability of the effective information elements contained in the image is low, different screening conditions for searching medical knowledge are acquired through different integration modes of the keywords of the text information and the keywords of the image information, the search results acquired through the integration modes of the information are integrated, and the search results are screened and recommended to an external information source according to the matching degree of the keywords of the text information and the keywords of the image information.
Drawings
Fig. 1 is a diagram of an overall architecture of a knowledge-based auxiliary decision making system of traditional Chinese medicine in an embodiment of the invention;
fig. 2 is a detailed diagram of a traditional Chinese medicine auxiliary decision making system architecture based on a knowledge graph in the embodiment of the invention;
fig. 3 is a first operation logic diagram of a traditional Chinese medicine auxiliary decision-making system based on a knowledge graph according to an embodiment of the invention;
fig. 4 is a second operational logic diagram of the auxiliary decision making system of traditional Chinese medicine based on a knowledge graph according to the embodiment of the invention.
Detailed Description
In order that the objects and advantages of the invention will become more apparent, the invention will be further described with reference to the following examples; it should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.
Referring to fig. 1 and 2, fig. 1 is a diagram of an overall architecture of a knowledge-based auxiliary decision system of a traditional Chinese medicine according to an embodiment of the present invention, and fig. 2 is a detailed diagram of an architecture of a knowledge-based auxiliary decision system of a traditional Chinese medicine according to an embodiment of the present invention, including:
a database for storing knowledge-graph based medical knowledge, an effective image element training set, and an effective keyword training set;
the identification unit comprises a text information extraction module connected with the database for extracting keywords in text information from an external information source according to the effective keyword training set and an image information extraction module connected with the database for identifying and extracting effective information elements in image feedback information according to the effective image element training set and sequentially converting each effective information element into corresponding keywords; the identification unit marks the acquired keywords as keyword sets, and comprises a text keyword set A extracted by the text information extraction module according to the text information and an image keyword set C extracted by the image information extraction module according to the effective information elements;
the information integration unit is connected with the identification unit and is used for integrating the keyword sets to generate integrated keyword sets meeting corresponding conditions;
The knowledge retrieval unit is respectively connected with the database, the identification unit and the information integration unit and is used for retrieving medical knowledge in the database according to the keyword set or the integrated keyword set to obtain a plurality of retrieval results, wherein the retrieval results comprise keyword retrieval results obtained by using the keyword set for retrieval and integrated retrieval results obtained by using the integrated keyword set for retrieval; the knowledge retrieval unit is further configured to set recommendation priorities of the integrated retrieval results according to matching degrees of the integrated retrieval results and the integrated keyword sets;
an effective information judging unit connected with each module in the identifying unit and used for judging whether the keywords in the text keyword set A are effective or not according to the number of the search results acquired by the knowledge searching unit based on the text keyword set A and the category number of the search results acquired by the text keyword set A, and acquiring corresponding keywords from each search result acquired by the text keyword set A based on the text keyword set A and the effective keyword training set when judging that the keywords in the text keyword set A are invalid so as to establish a reference set and sequentially determine the priority of each keyword in the reference set;
The interaction unit is connected with the identification unit and is used for acquiring text information of an external information source, transmitting the text information to the text information extraction module, transmitting keywords belonging to the reference set to the external information source to acquire feedback information of the external information source on the keywords, transmitting an image information acquisition request to the external information source to acquire image feedback information of the external information source, and transmitting the image feedback information to the image information extraction module; the interaction unit is also used for transmitting each search result to an external information source and labeling the recommendation priority of each transmitted search result.
Specifically, the invention sets the database, can update the stored knowledge graph in real time, increase the medical knowledge after the knowledge graph update in real time, and can update each stored training set in real time according to the external information source, so that the accuracy rate of the medical knowledge finally obtained is higher; the invention is provided with the identification unit and the effective information judgment unit, can extract, identify and match keywords of the external information source and the search result, and realize the guidance of information correction of the external information source so as to increase the information reliability of the external information source; the invention sets the knowledge retrieval unit, can retrieve the medical knowledge according to the keyword of the external information source, and set the recommendation priority of each retrieval result according to the matching degree of the keyword and each retrieval result, the consulting property of the recommendation information of the external information source is more objective; the information integration unit is arranged, so that keywords provided by the external information source at different stages can be integrated, and the accuracy of the keywords of the external information source is improved.
The text information extraction module extracts keywords in text information of an external information source according to the effective keyword training set to obtain a text keyword set A, the knowledge retrieval unit retrieves medical knowledge in a database according to the text keyword set A, the effective information judgment unit judges whether the keywords in the text keyword set A are effective according to the number of retrieval results and the number of retrieval result categories, wherein,
if the number of the search results is larger than the maximum threshold value of the number of the search results, or the number of the search results is smaller than the minimum threshold value of the number of the search results, or the number of the search result categories is larger than the maximum threshold value of the number of the search result categories, the effective information judging unit judges that the keywords in the text keyword set A are invalid;
if the number of the search results is greater than or equal to the minimum threshold value of the number of the search results and is less than or equal to the maximum threshold value of the number of the search results, and the number of the search result categories is less than or equal to the maximum threshold value of the number of the search result categories, the effective information judging unit judges that the keywords in the text keyword set A are effective, the knowledge searching unit sets the recommendation priority of each search result acquired based on the text keyword set A according to the matching degree of each search result acquired based on the text keyword set A and the text keyword set A, and the interaction unit transmits each search result acquired based on the text keyword set A to an external information source and marks the recommendation priority of each transmitted search result.
Specifically, in this embodiment, the keyword algorithm based on the statistical feature extracts the text keyword, and the matching degree α=m/M of the keyword set and the search result obtained based on the keyword set, where M is the number of keywords in the keyword set included in the search result, M is the number of keywords in the keyword set, and the larger the matching degree value of the search result, the higher the recommendation priority to the search result.
Specifically, the present embodiment does not limit the classification method of the search result, and may be a therapy, a pharmaceutical agent, a human body part, or the like; the number of the search result categories in this embodiment indicates the number of parallel categories in which the search result is classified based on a certain classification mode, for example, the therapy category includes acupuncture moxibustion, massage, scraping, cupping, inspection, medicinal tea, medicated wine, and the like; the medicament category comprises cardiovascular diseases, respiratory diseases, digestive diseases and the like; the human body part comprises: spleen, stomach, kidneys, etc.
In this embodiment, the minimum number of search results threshold 20, the maximum number of search results threshold 500, and the maximum number of search results category threshold 8 are set without limiting the minimum number of search results threshold, the maximum number of search results threshold, and the maximum number of search results category threshold.
Specifically, the invention sets that the number of the search results or the number of the search result categories of the medical knowledge searched according to the keywords is too large or the number of the search results is too small, the current keywords are judged to be invalid, when the number of the search results or the number of the search result categories is too large, the current keywords are too fuzzy and have no characteristics, the acquired medical knowledge has lower referential property, and when the number of the search results is too small, the current keywords are inaccurate, so that the accurate medical knowledge cannot be searched.
Referring to fig. 3, which is a first operation logic diagram of the auxiliary decision making system of traditional Chinese medicine based on a knowledge graph according to the embodiment of the invention, when the number of search results obtained by the knowledge search unit is greater than a maximum threshold value of the number of search results or the number of search result categories is greater than a maximum threshold value of the number of search result categories, if the number of keywords in the text keyword set a is greater than or equal to a minimum threshold value of the number of keywords, the interaction unit sends an image information obtaining request to an external information source.
Specifically, the minimum threshold value of the number of keywords is set to 5 in this embodiment.
Specifically, when the number of the search results is too large or the number of the search result categories is too large and the number of the keywords is enough, the images provided by the external information sources are adopted to carry out auxiliary screening on medical knowledge, and the traditional Chinese medicine inquiry comprises 'inspection inquiry', the inquiry is converted into text information as a leading part, the inspection inquiry is converted into image information as an auxiliary judging condition, and the credibility of the external information sources can be increased.
When the number of the search results obtained by the knowledge search unit is larger than the maximum threshold value of the number of the search results or the number of the search result categories is larger than the maximum threshold value of the number of the search result categories, if the number of the keywords in the text keyword set A is smaller than the minimum threshold value of the number of the keywords, the knowledge search unit calculates the matching degree of the text keyword set A and each search result, screens the search result with the matching degree larger than or equal to the preset matching degree as a reference information set, the effective information judgment unit extracts each keyword which does not contain the text keyword set A in the reference information set as a reference set, sets the priority of each keyword in the reference set according to the descending order of the number of each keyword in the reference set in the reference information set,
the effective information judging unit respectively sets each keyword in the reference set as a first priority reference keyword, a second priority reference keyword, … … and an Nth priority reference keyword, wherein N is the number of keywords in the reference set;
when the number of the keywords in the reference set is the same in the reference information group, the effective information judging unit sets the keywords with the same number in the reference information group as priority according to an initial ordering rule.
Specifically, the preset matching degree in this embodiment is 0.6.
Specifically, in this embodiment, the first letter sorting rule indicates that the keywords with the same occurrence frequency in the reference information set are preferentially sorted according to the first letter sequence of the chinese pinyin, and when the first letters are the same, the keywords are sorted according to the second letter sequence, so as to sequentially form the priorities of the keywords.
Specifically, when the number of the search results is too large or the number of the search result categories is too large and the number of the keywords is small, the external information source can further provide effective information for the system by extracting the characteristic keywords of each current search result as the information group which is further confirmed to the external information source, so that further screening of medical knowledge is realized, and further recommended medical knowledge meeting the current needs of the external information source is obtained.
The interaction unit sequentially transmits the keywords in the reference set to an external information source according to the sequence of the first priority reference keyword, the second priority reference keyword, … … and the Nth priority reference keyword to acquire feedback information of the external information source on the keywords in the reference set, the information integration unit supplements keywords in the text keyword set A according to the feedback information of the external information source to form a new supplement keyword set B, the knowledge retrieval unit retrieves medical knowledge in the database according to the supplement keyword set B, the interaction unit judges whether to send an image information acquisition request to the external information source according to the retrieval result number acquired by the knowledge retrieval unit based on the supplement keyword set B,
If the number of the search results obtained by the knowledge search unit based on the supplementary keyword set B is larger than the maximum threshold value of the number of the search results, the interaction unit judges that an image information obtaining request is sent to an external information source;
if the number of the search results acquired by the knowledge search unit based on the supplementary keyword set B is smaller than or equal to the maximum threshold of the number of the search results, the knowledge search unit sets recommendation priorities of the search results acquired based on the supplementary keyword set B according to the matching degree of the search results acquired based on the supplementary keyword set B and the supplementary keyword set B, and the interaction unit transmits the search results acquired based on the supplementary keyword set B to an external information source and marks the recommendation priorities of the transmitted search results.
Specifically, the image information provided by the external information source comprises human body characterization information such as a look, a complexion, a tongue body, a tongue fur and the like; the external information source can confirm each keyword in the reference set and can select the continuous confirm keyword or the stop confirm keyword by itself.
The interaction unit sends an image information acquisition request to an external information source, acquires image feedback information of the external information source and transmits the image feedback information to the image information extraction module, the image information extraction module identifies effective information elements for extracting the image feedback information, converts each effective information element into a plurality of keywords, sets each keyword converted by each effective information element as an image keyword set C, the knowledge retrieval unit retrieves medical knowledge in a database according to the image keyword set C, the effective information judgment unit acquires confidence of the image keyword set C according to a coincidence ratio of a retrieval result acquired by the knowledge retrieval unit based on the image keyword set C and a retrieval result acquired by the knowledge retrieval unit based on a supplementary keyword set B,
If the coincidence ratio of the search result obtained based on the image keyword set C and the search result obtained based on the supplementary keyword set B is smaller than the coincidence ratio of the preset search result, the confidence degree of the effective information judging unit obtained by the image keyword set C is a first confidence degree;
if the coincidence ratio of the search result obtained based on the image keyword set C and the search result obtained based on the supplementary keyword set B is more than or equal to the preset search result coincidence ratio, the effective information judging unit obtains the confidence coefficient of the image keyword set C as a second confidence coefficient;
the first confidence level b1=η0 and the second confidence level b2=1 are set, where η is a superposition ratio of a search result set obtained based on the image keyword set C and a search result obtained based on the keyword B, and η0 is a preset search result superposition ratio.
Specifically, in this embodiment, the coincidence ratio a=z/min { Z1, Z2} of the search results obtained based on the image keyword set C and the search results obtained based on the supplementary keyword set B, where Z is the same number of search results as the search results obtained based on the image keyword set C and the search results obtained based on the supplementary keyword set B, Z1 is the number of search results obtained based on the supplementary keyword set B, and Z2 is the number of search results obtained based on the image keyword set C.
Specifically, the preset search result overlap ratio in this embodiment is 0.8.
Specifically, the image information extraction module in this embodiment performs recognition extraction on the effective information elements in the image feedback information based on a machine learning algorithm.
Specifically, the image information provided by the external information source is influenced by subjective conditions of the information source and objective conditions of the environment, the validity and the authenticity of the information contained in the image information cannot be guaranteed, the reliability of the image information can be roughly judged by comparing the effective information elements in the image information with text keywords, and the influence degree of the image information on a final search result is determined according to the reliability degree of the image information.
The information integration unit integrates each keyword in the supplementary keyword set B and each keyword in the image keyword set C according to the confidence coefficient of the image keyword set C and the number of keywords in the supplementary keyword set B, wherein,
if the confidence coefficient of the image keyword set C is equal to 1, the information integration unit sets the union of the supplementary keyword set B and the image keyword set C as a first keyword union D1;
if the confidence of the image keyword set C is not equal to 1, the information integration unit sets the union of the supplemental keyword set B and the image keyword set C as a second keyword union D2, sets the intersection of the supplemental keyword set B and the image keyword set C as a keyword intersection E, and sets the relative complement of the keyword intersection E in the keyword set D2 as a keyword complement F.
Specifically, when the confidence of the image information is high, the effective information elements contained in the image can be used as conditions for screening search results, so that the effective information elements are integrated into keywords of the text information to be used as screening conditions for searching medical knowledge, when the confidence of the image information is low, the reliability of the effective information elements contained in the image is low, different screening conditions for searching medical knowledge are acquired through different integration modes of the keywords of the text information and the keywords of the image information, the search results acquired through the integration modes of the information are integrated, and the search results are screened and recommended to an external information source according to the matching degree of the keywords of the text information and the keywords of the image information.
When the information integration unit acquires the first keyword union set D1, the knowledge retrieval unit retrieves medical knowledge in the database according to the first keyword union set D1, sets recommendation priorities of the retrieval results based on the medical knowledge acquired by the first keyword union set D1 according to matching degrees of the retrieval results based on the medical knowledge acquired by the first keyword union set D1 and the first keyword union set D1, and the interaction unit transmits the retrieval results based on the medical knowledge acquired by the first keyword union set D1 to an external information source and marks the recommendation priorities of the retrieval results.
When the information integration unit respectively acquires a second keyword union set D2, a keyword intersection set E and a keyword complement set F, the knowledge retrieval unit retrieves medical knowledge in the database according to the second keyword union set D2 to acquire a first retrieval result set, retrieves medical knowledge in the database according to the keyword intersection set E to acquire a second retrieval result set, retrieves medical knowledge in the database according to the keyword complement set F to acquire a third retrieval result set, and the knowledge retrieval unit acquires the union sets of the first retrieval result set, the second retrieval result set and the third retrieval result set as alternative retrieval result sets;
the knowledge retrieval unit screens out retrieval results with the matching degree with the second keyword union set D2 being larger than or equal to a preset matching degree from the candidate retrieval result set to serve as a reference medical knowledge set, recommendation priority of each retrieval result in the reference medical knowledge set is set according to the matching degree of each retrieval result in the reference medical knowledge set and the second keyword union set D2, and the interaction unit transmits each retrieval result in the reference medical knowledge set to an external information source and marks the recommendation priority of each transmitted retrieval result.
Referring to fig. 4, which is a second operation logic diagram of the auxiliary decision making system of traditional Chinese medicine based on a knowledge graph according to the embodiment of the invention, when the number of the search results obtained by the knowledge search unit is smaller than the minimum threshold value of the number of the search results, the interaction unit sends an image information obtaining request to an external information source, the interaction unit obtains image feedback information of the external information source and transmits the image feedback information to the image information extraction module, the image information extraction module identifies effective information elements for extracting the image feedback information and converts each effective information element into a plurality of keywords, each keyword converted by each effective information element is set as an image keyword set C, the information integration unit sets a union of the image keyword set C and a text keyword set a as a keyword union H, the knowledge search unit searches medical knowledge in a database according to the keyword union H, sets priorities of each search result based on medical knowledge obtained by the keyword union H according to matching degrees of each search result with the keyword union H, and transmits each recommended result based on the recommendation result to each keyword union H, and the information recommendation unit transmits each recommended result based on the priority to each recommendation result.
Specifically, if the external information source refuses to provide the image information, the image information extraction module determines that no effective information element exists, and the converted keyword set is an empty set.
Specifically, the recommendation priority of the search results in this embodiment determines the recommendation order of the search results to the external information sources, that is, the higher the priority of the search results, the higher the recommendation order of the search results to the external information sources.
Examples:
the interaction unit acquires external information source text information of 'stomach is not digested, tiredness is easy to fatigue', the text information extraction module identifies and extracts keywords of 'stomach', 'digestion' and 'fatigue', the knowledge retrieval unit retrieves medical knowledge according to each keyword, and records retrieval logs, wherein the log content comprises: the number of search results is 1019, the number of search results with a degree of matching equal to or greater than 0.6 is 517, the search result categories include "spleen and stomach deficiency cold, liver and stomach qi stagnation, liver and stomach stagnated heat, stomach yin deficiency", the effective information determination unit extracts keywords (keywords do not include "stomach", "digestion" and "fatigue") in each search result, the extraction results are { pale and thirst, pale complexion, hypodynamia, pale tongue, …, thin and white tongue coating, soft and weak pulse }, wherein the number of occurrences of pale and thirst in the search result with a degree of matching equal to or greater than 0.6 is 403, the number of occurrences of face bloom in the search result with a matching degree of 0.6 or more is 389, the number of occurrences of hypodynamia in the search result with a matching degree of 0.6 or more is 371, … …, the number of occurrences of pulse soft in the search result with a matching degree of 0.6 or more is 57, the effective information determination unit sets that the face bloom is a first priority reference keyword, the face bloom is a second priority reference keyword, … …, and the pulse soft is a first hundred-zero seven priority reference keyword. The interaction unit sends keywords in { pale mouth, pale complexion, hypodynamia, pale tongue, …, thin white coating and soft pulse } to an external information source according to the descending order of priority, the interaction unit obtains feedback information of the external information source on { pale mouth, pale complexion, hypodynamia, pale tongue, …, thin white coating and soft pulse } to obtain new keywords of 'pale mouth, not thirsty' and 'hypodynamia', and the knowledge retrieval unit retrieves medical knowledge according to { pale mouth, not thirsty, hypodynamia, stomach, digestion and fatigue }, records retrieval logs, and the log content comprises: the number of search results is 503. The interaction unit sends an image information acquisition request to an external information source, the interaction unit acquires a facial image, the image information extraction module acquires keywords such as "sallow complexion", "eye reddish swelling", "pale tongue", "thin white tongue coating" and "emaciation" according to the facial image, the knowledge retrieval unit retrieves medical knowledge according to { sallow complexion, eye reddish swelling, pale tongue, thin white tongue coating, emaciation }, and records retrieval logs, and the log content comprises: the search results according to { sallow complexion, red and swollen eyes, thin and white tongue coating, emaciation } include all search results according to { oral thin and thirst, hypodynamia, stomach, digestion, fatigue }. The information integrating unit acquires { sallow complexion, red and swollen eyes, pale tongue, thin and white coating, emaciation, light mouth, no thirst, hypodynamia, stomach, digestion and fatigue } as a final retrieval keyword set, the knowledge retrieving unit retrieves medical knowledge according to the final retrieval keyword set, sets recommendation priorities of retrieval results of the medical knowledge acquired based on the final retrieval keyword set according to matching degrees of the retrieval results of the medical knowledge acquired by the final retrieval keyword set and the final retrieval keyword set respectively, and the interaction unit transmits the retrieval results of the medical knowledge acquired based on the final retrieval keyword set to an external information source, and marks the recommendation priorities of the transmitted retrieval results.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
The foregoing description is only of the preferred embodiments of the invention and is not intended to limit the invention; various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (8)

1. The traditional Chinese medicine decision-making assisting system based on the knowledge graph is characterized by comprising the following components:
a database for storing knowledge-graph based medical knowledge, an effective image element training set, and an effective keyword training set;
the identification unit comprises a text information extraction module connected with the database for extracting text keywords in text information from an external information source according to the effective keyword training set and an image information extraction module connected with the database for identifying and extracting effective information elements in image feedback information according to the effective image element training set and sequentially converting each effective information element into corresponding image keywords;
The information integration unit is connected with the identification unit and is used for integrating the keywords to generate an integrated keyword set meeting the corresponding conditions;
the knowledge retrieval unit is respectively connected with the database, the identification unit and the information integration unit and is used for retrieving medical knowledge in the database according to the keywords or the integrated keyword set to obtain a plurality of retrieval results;
an effective information judging unit connected with each module in the identifying unit and used for judging whether the text keyword is effective or not according to the number of the search results obtained by the knowledge searching unit based on the text keyword and the category number of the search results, and obtaining corresponding keywords based on each search result obtained by the text keyword and secondary search of the effective keyword training set when judging that the text keyword is invalid so as to establish a reference set;
the interaction unit is connected with the identification unit and is used for acquiring text information of an external information source, transmitting the text information to the text information extraction module, transmitting keywords belonging to the reference set to the external information source to acquire feedback information of the external information source on the keywords, transmitting an image information acquisition request to the external information source to acquire image feedback information of the external information source, and transmitting the image feedback information to the image information extraction module; the interaction unit is also used for transmitting each search result to an external information source and labeling the recommendation priority of each transmitted search result;
The knowledge retrieval unit marks the set of text keywords as a text keyword set A, and retrieves medical knowledge in the database according to the text keyword set A, wherein:
the effective information judging unit judges that the keywords in the text keyword set A are invalid under a first judging condition;
the effective information judging unit judges that the keywords in the text keyword set A are effective under a second judging condition, the knowledge searching unit sets recommendation priorities of the search results acquired based on the text keyword set A according to the matching degree of the search results acquired based on the text keyword set A and the text keyword set A, and the interaction unit transmits the search results acquired based on the text keyword set A to an external information source and marks the recommendation priorities of the transmitted search results;
the first judging condition is that the number of the search results obtained based on the text keyword set A is larger than a maximum threshold value of the number of the search results, or the number of the search results is smaller than a minimum threshold value of the number of the search results, or the number of the search result categories is larger than a maximum threshold value of the number of the search result categories; the second judging condition is that the number of the search results obtained based on the text keyword set A is larger than or equal to a minimum threshold value of the number of the search results and smaller than or equal to a maximum threshold value of the number of the search results, and the number of the search result categories is smaller than or equal to a maximum threshold value of the number of the search result categories;
The information integration unit forms a new supplementary keyword set B by supplementing keywords into the text keyword set A according to feedback information of an external information source, the knowledge retrieval unit retrieves medical knowledge in a database according to the supplementary keyword set B, the interaction unit judges whether to send an image information acquisition request to the external information source according to the number of retrieval results acquired by the knowledge retrieval unit based on the supplementary keyword set B,
if the number of the search results obtained by the knowledge search unit based on the supplementary keyword set B is larger than the maximum threshold value of the number of the search results, the interaction unit judges that an image information obtaining request is sent to an external information source;
if the number of the search results acquired by the knowledge search unit based on the supplementary keyword set B is smaller than or equal to the maximum threshold value of the number of the search results, the knowledge search unit sets recommendation priorities of the search results acquired based on the supplementary keyword set B according to the matching degree of the search results acquired based on the supplementary keyword set B and the supplementary keyword set B, and the interaction unit transmits the search results acquired based on the supplementary keyword set B to an external information source and marks the recommendation priorities of the transmitted search results;
The information integration unit integrates each keyword in the supplemental keyword set B with each keyword in the image keyword set C according to the confidence of the image keyword set C, wherein,
the information integration unit sets the union of the supplemental keyword set B and the image keyword set C as a first keyword union D1 under the first confidence level condition;
the information integration unit sets the union of the supplementary keyword set B and the image keyword set C as a second keyword union D2 under the condition of second confidence, sets the intersection of the supplementary keyword set B and the image keyword set C as a keyword intersection E, and sets the relative complement of the keyword intersection E in the second keyword union D2 as a keyword complement F;
the first confidence coefficient condition is that the confidence coefficient of the image keyword set C is equal to 1, and the second confidence coefficient condition is that the confidence coefficient of the image keyword set C is not equal to 1;
when the information integration unit obtains a second keyword union set D2, a keyword intersection set E and a keyword complement set F respectively, the knowledge retrieval unit retrieves medical knowledge in the database according to the second keyword union set D2 to obtain a first retrieval result set, retrieves medical knowledge in the database according to the keyword intersection set E to obtain a second retrieval result set, retrieves medical knowledge in the database according to the keyword complement set F to obtain a third retrieval result set, and the knowledge retrieval unit obtains the union sets of the first retrieval result set, the second retrieval result set and the third retrieval result set as alternative retrieval result sets.
2. The knowledge-based auxiliary decision making system of traditional Chinese medicine according to claim 1, wherein the interaction unit sends an image information acquisition request to an external information source under the condition of a first search result;
the first search result condition is that the knowledge search unit is based on the fact that the number of search results obtained by the text keyword set A is larger than a maximum threshold value of the number of search results and the number of keywords in the text keyword set A is larger than or equal to a minimum threshold value of the number of keywords, or based on the fact that the number of search result categories obtained by the text keyword set A is larger than the maximum threshold value of the number of search result categories and the number of keywords in the text keyword set A is larger than or equal to the minimum threshold value of the number of keywords.
3. The knowledge-graph-based auxiliary decision system of traditional Chinese medicine according to claim 1, wherein the knowledge retrieval unit calculates the matching degree of the text keyword set A and each retrieval result obtained based on the text keyword set A under the condition of the second retrieval result, and uses a plurality of retrieval results with the matching degree larger than or equal to a preset matching degree as a reference information set, the effective information judgment unit extracts keywords which belong to the effective keyword training set and do not belong to the text keyword set A in the reference information set as a reference set, and sets the priority for each keyword in the reference set according to the descending order of the number of each keyword in the reference set in the reference information set,
The second search result condition is that the knowledge search unit obtains the number of search results based on the text keyword set A, wherein the number of the search results is larger than a maximum threshold value of the number of the search results, and the number of keywords in the text keyword set A is smaller than a minimum threshold value of the number of the keywords, or the number of the search result categories is larger than a maximum threshold value of the number of the search result categories, and the number of the keywords in the text keyword set A is smaller than a minimum threshold value of the number of the keywords;
the effective information judging unit sets each keyword in the reference set as a plurality of priority reference keywords according to the priority, and when a plurality of keywords exist in the reference set and the number of the keywords in the reference information set is the same, the effective information judging unit sets the priority of each keyword according to an initial ordering rule.
4. The knowledge-graph-based auxiliary decision-making system of traditional Chinese medicine according to claim 3, wherein the interaction unit sequentially transmits the keywords in the reference set to an external information source according to the sequence of a plurality of priority reference keywords to acquire feedback information of the external information source on the keywords in the reference set, the information integration unit selects the corresponding keywords from the reference set according to the feedback information of the external information source and adds the keywords to the text keyword set A to form a supplementary keyword set B, the knowledge retrieval unit retrieves the medical knowledge in the database according to the supplementary keyword set B, and the interaction unit judges whether to send an image information acquisition request to the external information source according to the retrieval result number acquired by the knowledge retrieval unit based on the supplementary keyword set B.
5. The knowledge-graph-based Chinese medicine decision-making aid system according to claim 4, wherein the image information extraction module converts the effective information elements recognized and extracted from the image feedback information into corresponding keywords respectively under a first interaction condition, and marks each keyword as an image keyword set C, the knowledge retrieval unit obtains a confidence level of the image keyword set C based on a coincidence ratio of an image retrieval result obtained by the image keyword set C and a retrieval result obtained by the supplementary keyword set B, wherein,
the effective information identification module acquires the confidence coefficient of the image keyword set C as a first confidence coefficient under the first coincidence rate condition; the effective information identification module acquires the confidence coefficient of the image keyword set C as a second confidence coefficient under the second coincidence rate condition;
the first interaction condition is that the interaction unit sends an image information acquisition request to an external information source and acquires image feedback information of the external information source, the first coincidence rate condition is that the coincidence rate of a search result acquired based on the image keyword set C and a search result acquired based on the supplementary keyword set B is smaller than a preset search result coincidence rate, and the second coincidence rate condition is that the coincidence rate of the search result acquired based on the image keyword set C and the search result acquired based on the supplementary keyword set B is larger than or equal to the preset search result coincidence rate.
6. The knowledge-based auxiliary decision-making system of traditional Chinese medicine according to claim 5, wherein the knowledge retrieval unit sets recommendation priorities of the retrieval results obtained by the first keyword union D1 according to matching degrees of the retrieval results obtained by the first keyword union D1 and the first keyword union D1 under a first retrieval condition, and the interaction unit transmits the retrieval results obtained by the first keyword union D1 to an external information source and marks the recommendation priorities of the transmitted retrieval results;
the first search condition is that the knowledge search unit searches medical knowledge in the database according to the first keyword union set D1.
7. The knowledge-based auxiliary decision making system of traditional Chinese medicine according to claim 6, wherein the knowledge retrieval unit obtains a union of the first retrieval result set, the second retrieval result set and the third retrieval result set as an alternative retrieval result set under a second retrieval condition;
the second search condition is that the knowledge search unit searches medical knowledge in the database according to the second keyword union set D2 to obtain the first search result set, searches medical knowledge in the database according to the keyword intersection set E to obtain the second search result set, and searches medical knowledge in the database according to the keyword complement set F to obtain the third search result set;
The knowledge retrieval unit screens out retrieval results with the matching degree with the second keyword union set D2 being greater than or equal to a preset matching degree from the candidate retrieval result set to serve as a reference medical knowledge set, recommendation priorities for the retrieval results in the reference medical knowledge set are respectively set according to the matching degree of the retrieval results in the reference medical knowledge set and the second keyword union set D2, and the interaction unit transmits the retrieval results in the reference medical knowledge set to an external information source and marks the recommendation priorities of the transmitted retrieval results.
8. The knowledge-based auxiliary decision-making system of traditional Chinese medicine according to claim 1, wherein the image information extraction module respectively converts effective information elements which are identified and extracted according to image feedback information into corresponding keywords under a third search result condition, marks each keyword as an image keyword set C, the knowledge search unit searches medical knowledge in the database according to a keyword union H acquired by the image keyword set C and the text keyword set a, sets recommendation priority of each search result acquired based on the keyword union H according to matching degree of each search result acquired based on the keyword union H with the keyword union H, and the interaction unit transmits each search result acquired based on the keyword union H to an external information source and marks the recommendation priority of each transmitted search result;
The keyword union H is a union of the image keyword set C and the text keyword set A integrated by the information integration unit, and the image keyword set C comprises a plurality of keywords extracted by the image information extraction module according to image feedback information;
the third search result condition is that the number of search results obtained by the knowledge search unit is smaller than a minimum threshold value of the number of search results, and the interaction unit sends an image information obtaining request to an external information source to obtain image feedback information.
CN202310593821.XA 2023-05-25 2023-05-25 Traditional Chinese medicine auxiliary decision-making system based on knowledge graph Active CN116344041B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310593821.XA CN116344041B (en) 2023-05-25 2023-05-25 Traditional Chinese medicine auxiliary decision-making system based on knowledge graph

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310593821.XA CN116344041B (en) 2023-05-25 2023-05-25 Traditional Chinese medicine auxiliary decision-making system based on knowledge graph

Publications (2)

Publication Number Publication Date
CN116344041A CN116344041A (en) 2023-06-27
CN116344041B true CN116344041B (en) 2023-09-08

Family

ID=86877445

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310593821.XA Active CN116344041B (en) 2023-05-25 2023-05-25 Traditional Chinese medicine auxiliary decision-making system based on knowledge graph

Country Status (1)

Country Link
CN (1) CN116344041B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3575984A1 (en) * 2018-06-01 2019-12-04 Accenture Global Solutions Limited Artificial intelligence based-document processing
CN112420191A (en) * 2020-11-23 2021-02-26 北京麦岐科技有限责任公司 Traditional Chinese medicine auxiliary decision making system and method
CN114512228A (en) * 2022-02-08 2022-05-17 吾征智能技术(北京)有限公司 Traditional Chinese medicine disease auxiliary diagnosis system, equipment and storage medium
CN115238064A (en) * 2022-09-20 2022-10-25 大安健康科技(北京)有限公司 Keyword extraction method of traditional Chinese medicine medical record based on clustering

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3575984A1 (en) * 2018-06-01 2019-12-04 Accenture Global Solutions Limited Artificial intelligence based-document processing
CN112420191A (en) * 2020-11-23 2021-02-26 北京麦岐科技有限责任公司 Traditional Chinese medicine auxiliary decision making system and method
CN114512228A (en) * 2022-02-08 2022-05-17 吾征智能技术(北京)有限公司 Traditional Chinese medicine disease auxiliary diagnosis system, equipment and storage medium
CN115238064A (en) * 2022-09-20 2022-10-25 大安健康科技(北京)有限公司 Keyword extraction method of traditional Chinese medicine medical record based on clustering

Also Published As

Publication number Publication date
CN116344041A (en) 2023-06-27

Similar Documents

Publication Publication Date Title
CN107403068B (en) Merge the intelligence auxiliary way of inquisition and system of clinical thinking
CN110619962B (en) Doctor-patient sharing network medical service system
CN110289095B (en) Clinical intelligent auxiliary decision-making method and system for femoral neck fracture
Sager et al. Medical language processing: applications to patient data representation and automatic encoding
WO2020124856A1 (en) Diagnosis standardization method and device based on word vectors
CN106875941A (en) A kind of voice method for recognizing semantics of service robot
CN110111884A (en) A kind of man-machine coordination intelligent medical treatment aid decision-making system based on CMKMC
CN106845147A (en) Medical practice summarizes method for building up, device and the data assessment method of model
WO2019000852A1 (en) Apparatus for automatically triaging patient and automatic triage method
CN111370102A (en) Department diagnosis guiding method, device and equipment
CN109378077A (en) The machine readable storage medium of history-taking method and execution this method before examining
CN107705853A (en) Clinical nutrition risk screening method and system
CN115376648A (en) Intelligent prescription generating device and intelligent prescription generating system
CN116910172A (en) Follow-up table generation method and system based on artificial intelligence
CN113705477B (en) Medical image recognition method, system, equipment and medium
CN116344041B (en) Traditional Chinese medicine auxiliary decision-making system based on knowledge graph
Nasiri et al. A medical case-based reasoning approach using image classification and text information for recommendation
CN109686453A (en) Medical aid decision-making system based on big data
Ong et al. The colorectal cancer recurrence support (CARES) system
CN117334290A (en) Critical disease evidence-based nursing system
CN115565655A (en) Enhanced auxiliary inquiry method
CN117095835A (en) Intelligent medical health consultation system
Taboada et al. An automated approach to mapping external terminologies to the UMLS
US20220028511A1 (en) Systems and methods for initiating an updated user ameliorative plan
CN115910300A (en) Medical virtual platform system based on artificial intelligence and information processing 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