WO2021147290A1 - 一种医学术语系统的构建方法、装置、设备及存储介质 - Google Patents

一种医学术语系统的构建方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2021147290A1
WO2021147290A1 PCT/CN2020/105394 CN2020105394W WO2021147290A1 WO 2021147290 A1 WO2021147290 A1 WO 2021147290A1 CN 2020105394 W CN2020105394 W CN 2020105394W WO 2021147290 A1 WO2021147290 A1 WO 2021147290A1
Authority
WO
WIPO (PCT)
Prior art keywords
medical
annotation
medical term
term
relationship
Prior art date
Application number
PCT/CN2020/105394
Other languages
English (en)
French (fr)
Inventor
何国平
王旭英
胡云苹
董驰
Original Assignee
医惠科技有限公司
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 医惠科技有限公司 filed Critical 医惠科技有限公司
Publication of WO2021147290A1 publication Critical patent/WO2021147290A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • 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
    • G16H70/00—ICT specially adapted for the handling or processing of medical references

Definitions

  • the present invention relates to the field of data processing, and in particular to a method, device, equipment and computer-readable storage medium for constructing a medical terminology system.
  • the purpose of the present invention is to provide a method for constructing a medical terminology system, which can improve the convenience of constructing a medical terminology system and the completeness and accuracy of the medical terminology system; another purpose of the present invention is to provide a medical terminology system.
  • the term system construction device, equipment, and computer-readable storage medium all have the above-mentioned beneficial effects.
  • the present invention provides a method for constructing a medical terminology system, including:
  • semantic relationships include: synonym relationships, ISA relationships, and association relationships;
  • the medical term entry is obtained, and a medical term system is constructed.
  • it further includes:
  • the method further includes:
  • the new fourth annotation is set in the target medical term entry.
  • the process of collecting medical terms and obtaining the semantic relationship between different medical terms specifically includes:
  • the natural language processing technology is used to collect the medical terms in the electronic medical record, and to obtain the semantic relationship between different medical terms.
  • the method further includes:
  • the medical terms are classified according to the framework of the top-level classification system.
  • the present invention also provides a device for constructing a medical terminology system, including:
  • the obtaining module is used to collect medical terms and obtain the semantic relationship between different medical terms; wherein, the semantic relationship includes: synonym relationship, ISA relationship and association relationship;
  • the first setting module is used to set a standard term in the medical term that has a synonym relationship as a medical term entity, and set a synonym of the standard term as the first comment of the medical term entity;
  • the second setting module is used to set the medical term entity that has an ISA relationship and an association relationship with the target medical term entity as the second annotation and the third annotation of the target medical term entity;
  • the third setting module is configured to use the ISA relationship between the association relationships of the medical term entities to set a fourth comment for the target medical term entity;
  • the construction module is used to obtain medical term entries based on the medical term entity, the first annotation and/or the second annotation and/or the third annotation and/or the fourth annotation, and construct a medical terminology system.
  • the present invention also provides a device for constructing a medical terminology system, including:
  • Memory used to store computer programs
  • the processor is used to implement the steps of any one of the above-mentioned methods for constructing a medical terminology system when the computer program is executed.
  • the present invention also provides a computer-readable storage medium with a computer program stored on the computer-readable storage medium, and when the computer program is executed by a processor, the construction of any one of the above-mentioned medical terminology systems is realized. Method steps.
  • the method for constructing a medical terminology system can use the ISA relationship between the association relationships of the medical terminology entities to set a fourth annotation for the target medical terminology entity, thus not only can improve the medical terminology entry Completeness, thereby improving the completeness of the constructed medical terminology system, and can relatively reduce the workload of medical terminology development, and reduce the complexity of the operation process of constructing medical terminology; and, because the fourth annotation is based on the association relationship of medical terminology entities Therefore, the accuracy of setting medical term entries can be further improved.
  • the present invention also provides a construction device, equipment and computer-readable storage medium of a medical terminology system, all of which have the above-mentioned beneficial effects.
  • Fig. 1 is a flowchart of a method for constructing a medical terminology system provided by an embodiment of the present invention
  • FIG. 2 is a structural diagram of a device for constructing a medical terminology system provided by an embodiment of the present invention
  • Fig. 3 is a structural diagram of a device for constructing a medical terminology system provided by an embodiment of the present invention.
  • the construction device, equipment, and computer-readable storage medium of all have the above-mentioned beneficial effects.
  • Fig. 1 is a flowchart of a method for constructing a medical terminology system provided by an embodiment of the present invention. As shown in Figure 1, a method for constructing a medical terminology system includes:
  • semantic relations include: synonym relations, ISA relations, and association relations;
  • the semantic relationship includes: synonym relationship, ISA relationship, and association relationship.
  • the synonym relationship means that multiple medical terms are synonymous with each other, that is, the semantics of medical terms are the same, but the textual expressions are different;
  • the ISA relationship refers to the parent-child relationship between medical terms;
  • the association relationship refers to the term The medical relationship between diseases, such as the therapeutic relationship between disease and surgery.
  • the medical terms in this embodiment include traditional Chinese medicine terms and Western medicine terms, that is, a collection of terms covering related concepts in the clinical medical field of Chinese medicine and Western medicine.
  • medical terms and semantic relations between medical terms can be directly obtained from existing clinical standard terminology databases such as ICD, or they can be implicitly included in the free text of electronic medical records through natural language processing technology (NLP). The medical terms and their semantic relations are extracted.
  • S20 Set a standard term in a medical term with a synonym relationship as a medical term entity, and set a synonym of the standard term as the first comment of the medical term entity;
  • S50 derive medical term entries based on the medical term entity, the first annotation and/or the second annotation and/or the third annotation and/or the fourth annotation, and construct a medical terminology system.
  • the standard terms in the medical term entity are transformed into the concept of the ontology; the medical term entity that has a synonym relationship with the target medical term entity is transformed into the annotation of the concept in the ontology; the ISA relationship between the medical term entities is transformed, Transform into the implication axioms of the concepts in the ontology; the association relations between medical term entities are transformed into the full inventory quantitative constraints and implication axioms of the concepts in the ontology; transform the ISA relations between the association relations of the medical term entities into the ontology The axiom of implication in relation.
  • each medical term entity is taken as the target medical term entity, and the second annotation and the third annotation are set for each target medical term entity.
  • the medical term entity that has an ISA relationship with the target medical term entity is set as the second annotation of the target medical term entity; the medical term entity that has an association relationship with the target medical term entity is set as the target medical term entity The third comment.
  • the ISA relationship between the association relationships of medical term entities is used to set the fourth annotation for the target medical term entity.
  • medical term entities gastritis
  • stomach upper abdominal disease
  • upper abdomen where the target medical term entity is “gastritis”
  • gastritis gastritis
  • stomach upper abdominal disease
  • ISA relationship between "stomach” and "upper abdomen”
  • the medical term entry refers to the standard medical term and its corresponding annotation.
  • the medical term is derived from the medical term entity, the first annotation and/or the second annotation and/or the third annotation and/or the fourth annotation. Terminology entries, build a medical terminology system.
  • the concept of complete connotation refers to the concept that the attribute restriction and hierarchical relationship of the concept can be clearly and uniquely defined. That is to say, in actual operation, according to the term entity and the corresponding annotation, the medical term entry that is the equivalence axiom can be determined.
  • the method for constructing a medical terminology system can use the ISA relationship between the association relationships of medical terminology entities to set a fourth annotation for the target medical terminology entity, thereby not only improving the completeness of medical terminology entries , Thereby improving the completeness of the constructed medical terminology system, and can relatively reduce the workload of medical terminology development, and reduce the complexity of the operation process of constructing medical terminology; and, because the fourth annotation is based on the association relationship between medical term entities Therefore, the accuracy of setting medical term entries can be further improved.
  • this embodiment further illustrates and optimizes the technical solution. Specifically, this embodiment further includes:
  • this embodiment is an operation performed on the basis of a medical terminology system.
  • a medical terminology system When it is necessary to detect the medical term entry to be tested, obtain the medical term entry to be tested and determine whether the medical term entry to be tested is consistent with the corresponding target medical term entry in the medical terminology system; If the target medical term entry in the terminology system is incorrect, modify the target medical term entry. Or, the medical term entry to be tested is consistent with the corresponding target medical term entry, and the second annotation and the third annotation in the medical term entry to be tested are further used to determine a new fourth annotation; and the new fourth annotation Set in the target medical term entry.
  • the consistency of the medical term definition is detected by further compatible reasoning on the medical term entry to be tested. If there is inconsistency, it returns to modify the ontology; uses categorical reasoning to determine whether a concept is a child of another concept. It is used for automatic classification of concepts, such as determining "whether the femoral neck fracture is a child of femoral neck fracture", etc., to explore the implicit ISA relationship between terms; thus, the accuracy and completeness of the medical terminology system can be further improved.
  • this embodiment further explains and optimizes the technical solution.
  • the process of collecting medical terms and obtaining the semantic relationship between different medical terms specifically includes:
  • the natural language processing technology is preferably used to collect the medical terms in the electronic medical record and the semantic relationship between the medical terms.
  • medical terms are extracted through named entity recognition technology; semantic relationships between terms are extracted through entity relationship extraction technology.
  • named entity recognition refers to the recognition of medical terms from the free text of electronic medical records, including the term boundaries and semantic types of medical terms; named entity recognition technology includes two specific methods, one is based on classification, each A word may have multiple tags. The tag with the highest classification probability is selected as the tag of the word.
  • the tags here include term boundary and semantic type; the other is based on serialized tagging method, tag multiple words at the same time, select The labeled sequence with the largest joint probability.
  • Entity relationship extraction refers to the identification of semantic relationships between medical terms based on the contextual features of the free text of electronic medical records. Specifically, first determine the type of semantic relationship to be extracted, that is, synonym relationship (equivalence relationship), ISA relationship, specific association relationship, such as "diagnosis", "treatment”, etc., mainly using classification methods, based on the two medical terms Contextual features predict the most probable semantic relationship between medical terms.
  • this embodiment further explains and optimizes the technical solution. Specifically, after collecting the semantic relationship between medical terms and medical terms, this embodiment further includes:
  • the medical terms are classified according to the framework of the top-level classification system.
  • the framework of the top-level classification system is preset, and then the medical terms are classified according to the framework of the top-level classification system to obtain the classified medical terms.
  • the top-level classification system framework covers the medical terms of Chinese medicine and Western medicine
  • the top-level classification system framework used in this embodiment includes 27 main axes, which are specimens, measurement units and limit values, phrases, environment and location, diseases, Test indicators, health management, theory and experience, connectives, clinical events, equipment, social terms, body substances, physiological structure and functional systems, laboratory operations, four inspection objects, special concepts, external substances, documents, physical factors, Drugs, drug processing, organisms, diagnosis, syndromes, symptoms and signs, treatment.
  • other types of top-level classification system frameworks may also be used to classify medical terms, which is not limited in this embodiment.
  • the embodiments of the method for constructing a medical terminology system provided by the present invention are described in detail above.
  • the present invention also provides a device, equipment and computer-readable storage medium for constructing a medical terminology system corresponding to the method. Since the embodiments of the device, equipment, and computer-readable storage medium part correspond to the embodiments of the method part, for the embodiments of the device, equipment, and computer-readable storage medium part, please refer to the description of the embodiments of the method part. Do not go into details.
  • Fig. 2 is a structural diagram of a device for constructing a medical terminology system provided by an embodiment of the present invention. As shown in Fig. 2, a device for constructing a medical terminology system includes:
  • the obtaining module 21 is used to collect medical terms and obtain the semantic relationship between different medical terms; wherein, the semantic relationship includes: synonym relationship, ISA relationship, and association relationship;
  • the first setting module 22 is configured to set a standard term in a medical term with a synonym relationship as a medical term entity, and set a synonym of the standard term as a first comment of the medical term entity;
  • the second setting module 23 is used to set the medical term entity that has an ISA relationship and an association relationship with the target medical term entity as the second annotation and the third annotation of the target medical term entity;
  • the third setting module 24 is configured to use the ISA relationship between the association relationships of medical term entities to set a fourth comment for the target medical term entity;
  • the construction module 25 is used for deriving medical term entries based on the medical term entity, the first annotation and/or the second annotation and/or the third annotation and/or the fourth annotation, and constructing the medical terminology system.
  • the device for constructing a medical terminology system provided by the embodiment of the present invention has the beneficial effects of the above-mentioned method for constructing a medical terminology system.
  • the construction device of the medical terminology system further includes:
  • the entry acquisition module is used to acquire the entry of the medical term to be tested
  • the judgment module is used to judge whether the medical term entry to be tested is consistent with the corresponding target medical term entry; if not, call the modification module;
  • the modification module is used to modify the target medical term entry.
  • it further includes:
  • the determining module is used to further determine the new fourth annotation by using the second annotation and the third annotation in the medical term entry to be tested when the medical term entry to be tested is consistent with the corresponding target medical term entry;
  • the setting module is used to set the new fourth annotation in the target medical term entry.
  • the acquisition module specifically includes:
  • the acquiring unit is used for acquiring the medical terms in the electronic medical record and the semantic relationship between the medical terms by using natural language processing technology.
  • it further includes:
  • the classification module is used to classify various medical terms according to the framework of the top-level classification system.
  • FIG. 3 is a structural diagram of a device for constructing a medical terminology system provided by an embodiment of the present invention. As shown in FIG. 3, a device for constructing a medical terminology system includes:
  • the memory 31 is used to store computer programs
  • the processor 32 is used to implement the steps of the above-mentioned method for constructing a medical terminology system when the computer program is executed.
  • the device for constructing a medical terminology system provided by the embodiment of the present invention has the beneficial effects of the above-mentioned method for constructing a medical terminology system.
  • the present invention also provides a computer-readable storage medium on which a computer program is stored.
  • the computer program is executed by a processor, the steps of the method for constructing a medical terminology system are implemented.
  • the computer-readable storage medium provided by the embodiment of the present invention has the beneficial effects of the construction method of the above-mentioned medical terminology system.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Epidemiology (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

一种医学术语系统的构建方法、装置、设备及存储介质，方法包括：采集医学术语，获取不同的医学术语之间的语义关系；将存在同义词关系的医学术语中的标准术语设置为医学术语实体，将标准术语的同义词设置为医学术语实体的第一注释；将与目标医学术语实体存在ISA关系和关联关系的医学术语实体，分别设置为目标医学术语实体的第二注释和第三注释；利用医学术语实体的关联关系之间的ISA关系，为目标医学术语实体设置第四注释；依据医学术语实体、第一注释和/或第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统；能够提高构建医学术语系统的便捷性和医学术语系统的完备性和准确度。

Description

一种医学术语系统的构建方法、装置、设备及存储介质
本申请要求于2020年01月20日提交中国专利局、申请号为202010067200.4、发明名称为“一种医学术语系统的构建方法、装置、设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及数据处理领域，特别涉及一种医学术语系统的构建方法、装置、设备及计算机可读存储介质。
背景技术
目前，针对医学术语数据量大、构词复杂，且存在一词多意、多词一意以及含义不清等情况，为了更便于医学术语的管理，现有技术构建了医学术语系统。具体的，现有技术在构建医学术语系统时，是在采集医学术语并获取不同的医学术语之间的语义关系之后，利用各语义关系为医学术语设置医学术语词条；再利用医学术语词条构建出医学术语系统。可见，按照现有技术构建出的医学术语系统，仅仅利用采集到的语义关系为医学术语设置医学词条，导致医学术语系统的完备性受到采集的语义关系的数量的限制，并且大量的采集工作导致构建医学术语医学的操作过程复杂，难以适应日渐增长的语义标准化的需求。
因此，如何提高构建医学术语系统的便捷性和医学术语系统的完备性，是本领域技术人员目前需要解决的技术问题。
发明内容
有鉴于此，本发明的目的在于提供一种医学术语系统的构建方法，能够提高构建医学术语系统的便捷性和医学术语系统的完备性和准确度；本发明的另一目的是提供一种医学术语系统的构建装置、设备及计算机可读存储介质，均具有上述有益效果。
为解决上述技术问题，本发明提供一种医学术语系统的构建方法，包 括：
采集医学术语，获取不同的医学术语之间的语义关系；其中，所述语义关系包括：同义词关系、ISA关系和关联关系；
将存在同义词关系的所述医学术语中的标准术语设置为医学术语实体，将所述标准术语的同义词设置为所述医学术语实体的第一注释；
将与目标医学术语实体存在ISA关系和关联关系的所述医学术语实体，分别设置为所述目标医学术语实体的第二注释和第三注释；
利用所述医学术语实体的关联关系之间的ISA关系，为所述目标医学术语实体设置第四注释；
依据所述医学术语实体、所述第一注释和/或所述第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
优选地，进一步包括：
获取待测医学术语词条；
判断所述待测医学术语词条是否与对应的目标医学术语词条一致；
若否，则修改所述目标医学术语词条。
优选地，在所述判断所述待测医学术语词条是否与对应的目标医学术语词条一致之后，进一步包括：
若是，则进一步利用所述待测医学术语词条中的第二注释和第三注释确定出新的第四注释；
将所述新的第四注释设置于所述目标医学术语词条中。
优选地，所述采集医学术语，获取不同的医学术语之间的语义关系的过程，具体包括：
利用自然语言处理技术采集电子病历中的所述医学术语，获取不同的所述医学术语之间的所述语义关系。
优选地，在所述采集医学术语和所述医学术语之间的语义关系之后，进一步包括：
依据顶层分类体系框架将各所述医学术语进行分类。
为解决上述技术问题，本发明还提供一种医学术语系统的构建装置，包括：
获取模块，用于采集医学术语，获取不同的医学术语之间的语义关系；其中，所述语义关系包括：同义词关系、ISA关系和关联关系；
第一设置模块，用于将存在同义词关系的所述医学术语中的标准术语设置为医学术语实体，将所述标准术语的同义词设置为所述医学术语实体的第一注释；
第二设置模块，用于将与目标医学术语实体存在ISA关系和关联关系的所述医学术语实体，分别设置为所述目标医学术语实体的第二注释和第三注释；
第三设置模块，用于利用所述医学术语实体的关联关系之间的ISA关系，为所述目标医学术语实体设置第四注释；
构建模块，用于依据所述医学术语实体、所述第一注释和/或所述第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
为解决上述技术问题，本发明还提供一种医学术语系统的构建设备，包括：
存储器，用于存储计算机程序；
处理器，用于执行所述计算机程序时实现上述任一种医学术语系统的构建方法的步骤。
为解决上述技术问题，本发明还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现上述任一种医学术语系统的构建方法的步骤。
本发明提供的一种医学术语系统的构建方法，能够利用所述医学术语实体的关联关系之间的ISA关系，为所述目标医学术语实体设置第四注释，因此不仅能提高医学术语词条的完备性，进而提高构建的医学术语系统的完备性，而且能够相对减少医学术语开发的工作量，降低构建医学术语医学的操作过程复杂度；并且，由于第四注释是根据医学术语实体的关联关系之间的ISA关系得出的，因此，能够进一步提高设置医学术语词条的准确度。
为解决上述技术问题，本发明还提供了一种医学术语系统的构建装置、设备及计算机可读存储介质，均具有上述有益效果。
附图说明
为了更清楚地说明本发明实施例或现有技术的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单的介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据提供的附图获得其他的附图。
图1为本发明实施例提供的一种医学术语系统的构建方法的流程图；
图2为本发明实施例提供的一种医学术语系统的构建装置的结构图；
图3为本发明实施例提供的一种医学术语系统的构建设备的结构图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明实施例的核心是提供一种医学术语系统的构建方法，能够提高构建医学术语系统的便捷性和医学术语系统的完备性和准确度；本发明的另一核心是提供一种医学术语系统的构建装置、设备及计算机可读存储介质，均具有上述有益效果。
为了使本领域技术人员更好地理解本发明方案，下面结合附图和具体实施方式对本发明作进一步的详细说明。
图1为本发明实施例提供的一种医学术语系统的构建方法的流程图。如图1所示，一种医学术语系统的构建方法包括：
S10：采集医学术语，获取不同的医学术语之间的语义关系；其中，语义关系包括：同义词关系、ISA关系和关联关系；
具体的，在本实施例中，首先需要采集医学术语，并获取不同的医学 术语之间的语义关系；其中，语义关系包括：同义词关系、ISA关系和关联关系。具体的，同义词关系指的是多个医学术语之间互为同义词，即，医学术语之间的语义相同，但文字表述不同；ISA关系，指医学术语之间的父子关系；关联关系，指术语之间的医学关联关系，如疾病与手术之间的治疗关系。
需要说明的是，本实施例中的医学术语包括中医医学术语和西医医学术语，即涵盖中医和西医的临床医学领域相关概念的称谓集合。在实际操作中，可以直接从现已有的临床标准术语库如ICD等中获取医学术语以及医学术语之间的语义关系，或者可以通过自然语言处理技术(NLP)对电子病历自由文本中隐含的医学术语及其语义关系进行抽取。
S20：将存在同义词关系的医学术语中的标准术语设置为医学术语实体，将标准术语的同义词设置为医学术语实体的第一注释；
S30：将与目标医学术语实体存在ISA关系和关联关系的医学术语实体，分别设置为目标医学术语实体的第二注释和第三注释；
S40：利用医学术语实体的关联关系之间的ISA关系，为目标医学术语实体设置第四注释；
S50：依据医学术语实体、第一注释和/或第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
需要说明的是，在实际操作中，对于不存在同义词关系的医学术语，将该医学术语设置为医学术语实体；对于存在同义词关系的医学术语，将互为同义词中的标准术语设置为医学术语实体，对应的，将该标准术语的同义词设置为医学术语实体的第一注释。需要说明的是，标准术语指的是预先确定出的标准医学术语，具体可以根据本领域技术人员的使用习惯设置，本实施例对此不做限定。
具体的，将医学术语实体中的标准术语，转化为本体的概念；将与目标医学术语实体存在同义词关系的医学术语实体，转化为本体中概念的注释；将医学术语实体之间的ISA关系，转化为本体中概念的蕴含公理；医学术语实体之间的关联关系，转化为本体中的概念的完全存量量化约束和蕴含公理；将医学术语实体的关联关系之间的ISA关系，转化为本体中关 系的蕴含公理。
在实际操作中，在确定出医学术语实体之后，将各医学术语实体作为目标医学术语实体，对各目标医学术语实体设置第二注释和第三注释。需要说明的是，将与目标医学术语实体存在ISA关系的医学术语实体，设置为目标医学术语实体的第二注释；将与目标医学术语实体存在关联关系的医学术语实体，设置为目标医学术语实体的第三注释。
在本实施例中，利用医学术语实体的关联关系之间的ISA关系，为目标医学术语实体设置第四注释。例如，假设存在医学术语实体“胃炎”、“胃”、“上腹部疾病”和“上腹部”，其中，目标医学术语实体为“胃炎”，当前已知“胃炎”与“胃”存在关联关系，“上腹部疾病”和“上腹部”存在关联关系，且“胃”与“上腹部”存在ISA关系，因此可以得出“胃炎”与“上腹部疾病”存在ISA关系，即得出目标医学术语实体的第四注释。
在实际操作中，医学术语词条指的是标准医学术语以及其对应的注释，依据医学术语实体、第一注释和/或第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
需要说明的是，对于内涵完备的概念，确定出等价公理；其中，内涵完备的概念指的是概念的属性限制和层级关系可以明确、唯一的定义这个概念。也就是说，在实际操作中，根据术语实体以及对应的注释的情况，可以确定出为等价公理的医学术语词条。
本发明实施例提供的一种医学术语系统的构建方法，能够利用医学术语实体的关联关系之间的ISA关系，为目标医学术语实体设置第四注释，因此不仅能提高医学术语词条的完备性，进而提高构建的医学术语系统的完备性，而且能够相对减少医学术语开发的工作量，降低构建医学术语医学的操作过程复杂度；并且，由于第四注释是根据医学术语实体的关联关系之间的ISA关系得出的，因此，能够进一步提高设置医学术语词条的准确度。
在上述实施例的基础上，本实施例对技术方案作了进一步的说明和优化，具体的，本实施例进一步包括：
获取待测医学术语词条；
判断待测医学术语词条是否与对应的目标医学术语词条一致；
若否，则修改目标医学术语词条。
需要说明的是，在判断待测医学术语词条是否与对应的目标医学术语词条一致之后，进一步包括：
若是，则进一步利用待测医学术语词条中的第二注释和第三注释确定出新的第四注释；
将新的第四注释设置于目标医学术语词条中。
具体的，本实施例是在已得出医学术语系统的基础上进行的操作。在需要对待测医学术语词条进行检测时，通过获取待测医学术语词条，并判断待测医学术语词条是否与医学术语系统中对应的目标医学术语词条一致；若不一致，则在医学术语系统中目标医学术语词条有误的情况下，修改该目标医学术语词条。或者，待测医学术语词条与对应的目标医学术语词条一致，进一步利用待测医学术语词条中的第二注释和第三注释确定出新的第四注释；并将新的第四注释设置于目标医学术语词条中。
可见，本实施例通过进一步对待测医学术语词条进行相容推理检测医学术语定义的一致性，如果出现不一致，返回修改本体；利用归类推理，判断一个概念是否为另一个概念的子级，用于概念的自动分类，如判定“股骨颈中段骨折是否是为股骨颈骨折的子级”等，挖掘术语之间隐含的ISA关系；从而可以进一步提高医学术语系统的准确性和完备性。
在上述实施例的基础上，本实施例对技术方案作了进一步的说明和优化，具体的，本实施例中，采集医学术语，获取不同的医学术语之间的语义关系的过程，具体包括：
利用自然语言处理技术采集电子病历中的医学术语，获取不同的医学术语之间的语义关系。
在本实施例中，优选地利用自然语言处理技术采集电子病历中的医学术语和医学术语之间的语义关系。具体的，通过命名实体识别技术抽取医学术语；通过实体关系抽取技术抽取术语之间的语义关系。具体的，命名 实体识别指的是从电子病历自由文本中识别出医学术语，包括医学术语的术语边界和所属的语义类型；命名实体识别技术具体包括两种，一种是基于分类的方法，每一个词可能有多个标记，选择分类概率最大的标记作为该词的标记，此处的标记包括术语边界和语义类型；另一种是基于序列化标注的方法，对多个词同时标记，选择联合概率最大的标注序列。
实体关系抽取(共指消解)指的是根据电子病历自由文本的上下文特征，识别医学术语之间的语义关系。具体的，首先确定待抽取的语义关系类型，即同义词关系(等价关系)、ISA关系、具体的关联关系，如“诊断”、“治疗”等，主要采用分类方法，根据两个医学术语的上下文特征预测医学术语间概率最大的语义关系。
可见，本实施例通过利用自然语言处理技术采集电子病历中的医学术语，获取不同的医学术语之间的语义关系，能够获取到更多的医学术语和医学术语之间的语义关系。
在上述实施例的基础上，本实施例对技术方案作了进一步的说明和优化，具体的，本实施例在采集医学术语和医学术语之间的语义关系之后，进一步包括：
依据顶层分类体系框架将各医学术语进行分类。
在本实施例中，预先设置顶层分类体系框架，然后依据顶层分类体系框架将各医学术语进行分类，得到分类后的医学术语。具体的，顶层分类体系框架涵盖中医与西医的医学术语，并且本实施例中所使用的顶层分类体系框架包括27个主轴，分别是标本、测量单位和限定值、短语、环境与定位、疾病、检测指标、健康管理、理论与经验、连接词、临床事件、设备、社会用语、身体物质、生理结构和功能系统、实验室操作、四个检查对象、特殊概念、外部物质、文档、物理因素、药物、药物加工、有机体、诊断、证候、症状和体征、治疗。需要说明的是，在其他实施例中，也可以采用其他类型的顶层分类体系框架对医学术语进行分类，本实施例对此不做限定。
可见，通过进一步依据顶层分类体系框架将各医学术语进行分类，能 够更便于用户查看各医学术语，进一步提升用户的使用体验。
上文对于本发明提供的一种医学术语系统的构建方法的实施例进行了详细的描述，本发明还提供了一种与该方法对应的医学术语系统的构建装置、设备及计算机可读存储介质，由于装置、设备及计算机可读存储介质部分的实施例与方法部分的实施例相互照应，因此装置、设备及计算机可读存储介质部分的实施例请参见方法部分的实施例的描述，这里暂不赘述。
图2为本发明实施例提供的一种医学术语系统的构建装置的结构图，如图2所示，一种医学术语系统的构建装置包括：
获取模块21，用于采集医学术语，获取不同的医学术语之间的语义关系；其中，语义关系包括：同义词关系、ISA关系和关联关系；
第一设置模块22，用于将存在同义词关系的医学术语中的标准术语设置为医学术语实体，将标准术语的同义词设置为医学术语实体的第一注释；
第二设置模块23，用于将与目标医学术语实体存在ISA关系和关联关系的医学术语实体，分别设置为目标医学术语实体的第二注释和第三注释；
第三设置模块24，用于利用医学术语实体的关联关系之间的ISA关系，为目标医学术语实体设置第四注释；
构建模块25，用于依据医学术语实体、第一注释和/或第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
本发明实施例提供的医学术语系统的构建装置，具有上述医学术语系统的构建方法的有益效果。
作为优选的实施方式，医学术语系统的构建装置进一步包括：
词条获取模块，用于获取待测医学术语词条；
判断模块，用于判断待测医学术语词条是否与对应的目标医学术语词条一致；若否，则调用修改模块；
修改模块，用于修改目标医学术语词条。
作为优选的实施方式，进一步包括：
确定模块，用于当待测医学术语词条与对应的目标医学术语词条一致时，进一步利用待测医学术语词条中的第二注释和第三注释确定出新的第四注释；
设置模块，用于将新的第四注释设置于目标医学术语词条中。
作为优选的实施方式，获取模块具体包括：
获取单元，用于利用自然语言处理技术采集电子病历中的医学术语和医学术语之间的语义关系。
作为优选的实施方式，进一步包括：
分类模块，用于依据顶层分类体系框架将各医学术语进行分类。
图3为本发明实施例提供的一种医学术语系统的构建设备的结构图，如图3所示，一种医学术语系统的构建设备包括：
存储器31，用于存储计算机程序；
处理器32，用于执行计算机程序时实现如上述医学术语系统的构建方法的步骤。
本发明实施例提供的医学术语系统的构建设备，具有上述医学术语系统的构建方法的有益效果。
为解决上述技术问题，本发明还提供一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，计算机程序被处理器执行时实现如上述医学术语系统的构建方法的步骤。
本发明实施例提供的计算机可读存储介质，具有上述医学术语系统的构建方法的有益效果。
以上对本发明所提供的医学术语系统的构建方法、装置、设备及计算机可读存储介质进行了详细介绍。本文中应用了具体实施例对本发明的原 理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想。应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明原理的前提下，还可以对本发明进行若干改进和修饰，这些改进和修饰也落入本发明权利要求的保护范围内。
说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。对于实施例公开的装置而言，由于其与实施例公开的方法相对应，所以描述的比较简单，相关之处参见方法部分说明即可。
专业人员还可以进一步意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。

Claims (8)

  1. 一种医学术语系统的构建方法，其特征在于，包括：
    采集医学术语，获取不同的医学术语之间的语义关系；其中，所述语义关系包括：同义词关系、ISA关系和关联关系；
    将存在同义词关系的所述医学术语中的标准术语设置为医学术语实体，将所述标准术语的同义词设置为所述医学术语实体的第一注释；
    将与目标医学术语实体存在ISA关系和关联关系的所述医学术语实体，分别设置为所述目标医学术语实体的第二注释和第三注释；
    利用所述医学术语实体的关联关系之间的ISA关系，为所述目标医学术语实体设置第四注释；
    依据所述医学术语实体、所述第一注释和/或所述第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
  2. 根据权利要求1所述的方法，其特征在于，进一步包括：
    获取待测医学术语词条；
    判断所述待测医学术语词条是否与对应的目标医学术语词条一致；
    若否，则修改所述目标医学术语词条。
  3. 根据权利要求2所述的方法，其特征在于，在所述判断所述待测医学术语词条是否与对应的目标医学术语词条一致之后，进一步包括：
    若是，则进一步利用所述待测医学术语词条中的第二注释和第三注释确定出新的第四注释；
    将所述新的第四注释设置于所述目标医学术语词条中。
  4. 根据权利要求1所述的方法，其特征在于，所述采集医学术语，获取不同的医学术语之间的语义关系的过程，具体包括：
    利用自然语言处理技术采集电子病历中的所述医学术语，获取不同的所述医学术语之间的所述语义关系。
  5. 根据权利要求1所述的方法，其特征在于，在所述采集医学术语和所述医学术语之间的语义关系之后，进一步包括：
    依据顶层分类体系框架将各所述医学术语进行分类。
  6. 一种医学术语系统的构建装置，其特征在于，包括：
    获取模块，用于采集医学术语，获取不同的医学术语之间的语义关系；其中，所述语义关系包括：同义词关系、ISA关系和关联关系；
    第一设置模块，用于将存在同义词关系的所述医学术语中的标准术语设置为医学术语实体，将所述标准术语的同义词设置为所述医学术语实体的第一注释；
    第二设置模块，用于将与目标医学术语实体存在ISA关系和关联关系的所述医学术语实体，分别设置为所述目标医学术语实体的第二注释和第三注释；
    第三设置模块，用于利用所述医学术语实体的关联关系之间的ISA关系，为所述目标医学术语实体设置第四注释；
    构建模块，用于依据所述医学术语实体、所述第一注释和/或所述第二注释和/或第三注释和/或第四注释得出医学术语词条，构建医学术语系统。
  7. 一种医学术语系统的构建设备，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于执行所述计算机程序时实现如权利要求1至5任一项所述的医学术语系统的构建方法的步骤。
  8. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1至5任一项所述的医学术语系统的构建方法的步骤。
PCT/CN2020/105394 2020-01-20 2020-07-29 一种医学术语系统的构建方法、装置、设备及存储介质 WO2021147290A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010067200.4A CN111274400B (zh) 2020-01-20 2020-01-20 一种医学术语系统的构建方法、装置、设备及存储介质
CN202010067200.4 2020-01-20

Publications (1)

Publication Number Publication Date
WO2021147290A1 true WO2021147290A1 (zh) 2021-07-29

Family

ID=70999001

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/105394 WO2021147290A1 (zh) 2020-01-20 2020-07-29 一种医学术语系统的构建方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN111274400B (zh)
WO (1) WO2021147290A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111274400B (zh) * 2020-01-20 2021-02-12 医惠科技有限公司 一种医学术语系统的构建方法、装置、设备及存储介质
CN112541056B (zh) * 2020-12-18 2024-05-31 卫宁健康科技集团股份有限公司 医学术语标准化方法、装置、电子设备及存储介质
CN113161011A (zh) * 2021-02-19 2021-07-23 中国中医科学院中医药信息研究所 药物间不良反应预警方法、预警装置及预警系统
CN113111191A (zh) * 2021-04-22 2021-07-13 崔蒙 中西医结合临床术语系统的构建方法及装置
CN113221543B (zh) * 2021-05-07 2023-10-10 中国医学科学院医学信息研究所 一种医学术语整合方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050021324A1 (en) * 2003-07-25 2005-01-27 Brants Thorsten H. Systems and methods for new event detection
CN102426578A (zh) * 2011-08-25 2012-04-25 华南理工大学 一种智能语义网中本体概念模糊相似度度量方法
CN107480131A (zh) * 2017-07-25 2017-12-15 李姣 中文电子病历症状语义提取方法及其系统
CN109086356A (zh) * 2018-07-18 2018-12-25 哈尔滨工业大学 大规模知识图谱的错误连接关系诊断及修正方法
CN110442869A (zh) * 2019-08-01 2019-11-12 腾讯科技（深圳）有限公司 一种医疗文本处理方法及其装置、设备和存储介质
CN111274400A (zh) * 2020-01-20 2020-06-12 医惠科技有限公司 一种医学术语系统的构建方法、装置、设备及存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040117127A1 (en) * 2002-12-11 2004-06-17 Affymetrix, Inc. Methods, computer software products and systems for clustering genes
CN104484845B (zh) * 2014-12-30 2019-03-05 天津迈沃医药技术股份有限公司 基于医学信息本体数据库的疾病自我分析平台
CN107844482A (zh) * 2016-09-17 2018-03-27 复旦大学 基于全局本体的多数据源模式匹配方法
CN108776706A (zh) * 2018-06-13 2018-11-09 北京信息科技大学 一种基于本体的专利技术主题聚类方法
CN109446340A (zh) * 2018-10-17 2019-03-08 长沙瀚云信息科技有限公司 一种医学标准术语本体管理系统及方法、设备和存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050021324A1 (en) * 2003-07-25 2005-01-27 Brants Thorsten H. Systems and methods for new event detection
CN102426578A (zh) * 2011-08-25 2012-04-25 华南理工大学 一种智能语义网中本体概念模糊相似度度量方法
CN107480131A (zh) * 2017-07-25 2017-12-15 李姣 中文电子病历症状语义提取方法及其系统
CN109086356A (zh) * 2018-07-18 2018-12-25 哈尔滨工业大学 大规模知识图谱的错误连接关系诊断及修正方法
CN110442869A (zh) * 2019-08-01 2019-11-12 腾讯科技（深圳）有限公司 一种医疗文本处理方法及其装置、设备和存储介质
CN111274400A (zh) * 2020-01-20 2020-06-12 医惠科技有限公司 一种医学术语系统的构建方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN111274400A (zh) 2020-06-12
CN111274400B (zh) 2021-02-12

Similar Documents

Publication Publication Date Title
WO2021147290A1 (zh) 一种医学术语系统的构建方法、装置、设备及存储介质
Pezoulas et al. Medical data quality assessment: On the development of an automated framework for medical data curation
CN109299239B (zh) 一种基于es的电子病历检索方法
CN110136788B (zh) 一种基于自动检测的病历质检方法、装置、设备及存储介质
García et al. Semantic similarity-based alignment between clinical archetypes and SNOMED CT: an application to observations
Matheny et al. Detection of infectious symptoms from VA emergency department and primary care clinical documentation
CN112562807B (zh) 医疗数据分析方法、装置、设备、存储介质和程序产品
CN112614565A (zh) 一种基于知识图谱技术的中药经典名方智能推荐方法
CN112541066B (zh) 基于文本结构化的医技报告检测方法及相关设备
TWI521467B (zh) Nursing decision support system
Demner-Fushman et al. A Knowledge-Based Approach to Medical Records Retrieval.
CN113808758B (zh) 一种检验数据标准化的方法、装置、电子设备和存储介质
Cross et al. Using semantic similarity in ontology alignment
Horng et al. Consensus development of a modern ontology of emergency department presenting problems—the Hierarchical Presenting Problem Ontology (HaPPy)
Kim Automating lexical cross-mapping of ICNP to SNOMED CT
CN112115697A (zh) 用于确定目标文本的方法、装置、服务器以及存储介质
CN113553840B (zh) 一种文本信息处理方法、装置、设备及存储介质
Friedlin et al. Knowledge discovery and data mining of free text radiology reports
Meystre et al. Evaluation of medical problem extraction from electronic clinical documents using MetaMap Transfer (MMTx)
Chondrogiannis et al. An Intelligent Ontology Alignment Tool Dealing with Complicated Mismatches.
CN112699669B (zh) 流行病学调查报告的自然语言处理方法、装置及存储介质
Park et al. Criteria2Query 3.0: Leveraging generative large language models for clinical trial eligibility query generation
Varzeshnezhad et al. Transcultural mapping and usability testing of the clinical care classification system for an Iranian neonatal ICU population
Dhombres et al. Developing a knowledge base to support the annotation of ultrasound images of ectopic pregnancy
Ming et al. AI assisted clinical diagnosis & treatment and development strategy

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20916021

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20916021

Country of ref document: EP

Kind code of ref document: A1