Medical term mapping method based on knowledge graph, electronic equipment and medium

Abstract

The invention discloses a medical term mapping method based on a knowledge graph, electronic equipment and a medium. The method can comprise the steps of determining a knowledge graph basic architecture type of medical terms, and then performing correlation assignment to obtain a knowledge graph; performing format specification processing on the mapped term set, and matching the mapped term set with the knowledge graph to obtain a mapping result of the mapped terms; and after the mapping result is obtained, carrying out semi-automatic mapping on the new entry in the mapped term set based on the knowledge graph to obtain the mapping result corresponding to the new entry. According to the method, the knowledge graph and the medical term mapping are combined, the correlation of term attributes is increased, the term mapping compatibility is improved, and the mapping effect of newly added terms is automatically corrected based on the knowledge graph.

Description

Technical Field

[0001] The present invention relates to the field of medical terminology mapping, and more specifically, to a medical terminology mapping method, electronic device and medium based on a knowledge graph. Background Art

[0002] Medical terminology is a special vocabulary used by medical professionals in the process of diagnosis, treatment and care, aiming to achieve effective and accurate communication. Medical terminology is an important component of clinical information systems, including diagnostic terminology, drug terminology, laboratory terminology, examination terminology, operation data, surgical data, traditional Chinese medicine terminology, etc. Foreign medical terminology is relatively standardized, including the International Classification of Diseases and Codes (ICD-10), the International Classification of Diseases and Codes (ICD-11), the International Classification of Diseases, Ninth Edition, Clinical Revision, Volume 3 (ICD-9-CM-3), the Standard Nomenclature of Clinical Drugs (RxNorm), the Systematic Nomenclature of Medical Terminology-Clinical Terminology (SNOMED-CT), the Logical Nomenclature Precoding System for Observation Indicator Identifiers (LOINC), the Unified Medical Language System (UMLS), etc. There are also Common Clinical Medical Terms (2023 Edition) in China, the National Clinical Edition 2.0 Disease Diagnosis Code (ICD-10), the National Medical Insurance Edition 2.0 Disease Diagnosis Code (ICD-10), etc.

[0003] However, there is a lack of standards related to medical terminology in China, and the system is not yet complete. Most of them are for the convenience of statistics and charging, and there is a certain gap with clinical business expression. It is mainly manifested in: 1. There is a lack of a unified organization. Each health administrative agency, medical institution and clinical information system manufacturer maintains it separately, and the dictionary types vary greatly; 2. The system connection is inefficient. Due to the inconsistency of terminology among manufacturers, data sharing and interoperability between systems are difficult; 3. The terminology maintenance of each medical institution is too casual. There is no special department for maintenance, and there will be multiple terminology entries according to business needs, which makes it difficult to maintain terminology in the later stage. 4. There is a gap between existing terminology and clinical application, especially in the process of writing electronic medical records, there are often differences between terminology and clinical guideline standards.

[0004] Therefore, it is necessary to develop a medical terminology mapping method, electronic device and medium based on knowledge graph.

[0005] The information disclosed in the background technology section of the present invention is only intended to deepen the understanding of the general background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention proposes a medical terminology mapping method, electronic device and medium based on knowledge graph, which combines the knowledge graph with medical terminology mapping, increases the relevance of term attributes, improves the compatibility of term mapping, and automatically corrects the mapping effect of newly added terms based on the knowledge graph.

[0007] In a first aspect, the present disclosure provides a medical terminology mapping method based on a knowledge graph, comprising: Determine the basic architecture type of the knowledge graph of medical terms, and then assign relevance values ​​to obtain the knowledge graph; Perform format standardization on the mapped term set, match it with the knowledge graph, and obtain the mapping results of the mapped terms; After obtaining the mapping results, semi-automatic mapping is performed on the new terms in the mapped term set based on the knowledge graph to obtain the mapping results corresponding to the new terms.

[0008] Preferably, the basic architecture types of the knowledge graph include term type, term encoding, term name, term relationship, and attribute value; Among them, the term types include disease, medicine, examination, test, and surgery.

[0009] Preferably, the relevance is the degree of match between the term name and the attribute value.

[0010] Preferably, the corresponding correlations are assigned to category 1, category 2, and category 3 according to the matching degree from high to low, wherein the matching degree corresponding to category 1 is greater than the matching degree corresponding to category 2 and the matching degree corresponding to category 3.

[0011] Preferably, the format of the mapped term set is standardized and matched with the knowledge graph to obtain the mapping results of the mapped terms, including: Match the mapped terms with the terms with a correlation of category 1 in the knowledge graph to obtain the mapping results; If no term with a correlation of Class 1 is matched, the mapped term is segmented, and the segmentation result is matched with the terms with a correlation of Class 1 and Class 2 in the knowledge graph to obtain the mapping result.

[0012] Preferably, the word segmentation processing method includes a forward maximum matching method, a reverse maximum matching algorithm, a bidirectional maximum matching method, and a segmentation mark setting method.

[0013] Preferably, the word segmentation results are matched with the terms in the knowledge graph with the correlation of category 1 and category 2, and the mapping results obtained include: Each entry in the word segmentation result is matched with the terms in the knowledge graph with correlation of category 1 and category 2, and the matched results are weighted to obtain the relevance. The matching result with the highest relevance is taken as the final matching result.

[0014] Preferably, after obtaining the mapping result, semi-automatic mapping is performed on the new terms in the mapped term set based on the knowledge graph, and the mapping result corresponding to the new terms is obtained, including: Match the new entry with the term ID in the mapping result and display the matching mapping results; If there is no matching result with the term id, it will be matched with the term name in the mapping result, and the matching mapping result will be displayed; If there is no matching result with the term id or term name, the new term is segmented, and the segmentation result is matched with the terms with a correlation of category 1 in the knowledge graph, and the matching mapping results are displayed.

[0015] In a second aspect, an embodiment of the present disclosure further provides an electronic device, the electronic device comprising: A memory storing executable instructions; A processor runs the executable instructions in the memory to implement the medical terminology mapping method based on the knowledge graph.

[0016] In a third aspect, an embodiment of the present disclosure further provides a computer-readable storage medium storing a computer program, which implements the medical terminology mapping method based on the knowledge graph when executed by a processor.

[0017] The methods and apparatus of the present invention have other features and advantages that will be apparent from, or will be described in detail in, the accompanying drawings and subsequent detailed descriptions incorporated herein, which together serve to explain the specific principles of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.

[0019] Figure 1 A flowchart showing the steps of a medical terminology mapping method based on a knowledge graph according to an embodiment of the present invention.

[0020] Figure 2 A flowchart of medical terminology mapping based on a knowledge graph according to an embodiment of the present invention is shown.

[0021] Figure 3 A schematic diagram showing semi-automatic mapping of medical terms according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0022] The preferred embodiments of the present invention will be described in more detail below. Although the preferred embodiments of the present invention are described below, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0023] To facilitate understanding of the solutions and effects of the embodiments of the present invention, three specific application examples are given below. Those skilled in the art should understand that the examples are only for facilitating understanding of the present invention, and any specific details thereof are not intended to limit the present invention in any way. Example 1

[0024] Figure 1 A flowchart showing the steps of a medical terminology mapping method based on a knowledge graph according to an embodiment of the present invention.

[0025] like Figure 1 As shown, the medical terminology mapping method based on knowledge graph includes: Step 101, determining the basic architecture type of the knowledge graph of medical terms, and then performing relevance assignment to obtain the knowledge graph; Step 102, formatting the mapped term set, matching it with the knowledge graph, and obtaining the mapping results of the mapped terms; Step 103, after obtaining the mapping results, semi-automatic mapping is performed on the new terms in the mapped term set based on the knowledge graph to obtain the mapping results corresponding to the new terms.

[0026] In one example, the basic architecture types of the knowledge graph include term type, term encoding, term name, term relationship, and attribute value; Among them, the term types include disease, medicine, examination, test, and surgery.

[0027] In one example, relevance is how closely a term name matches an attribute value.

[0028] In one example, the corresponding correlations are assigned to category 1, category 2, and category 3 according to the matching degree from high to low, wherein the matching degree corresponding to category 1 is greater than the matching degree corresponding to category 2 and the matching degree corresponding to category 3.

[0029] In one example, the format of the mapped term set is standardized and matched with the knowledge graph to obtain the mapping results of the mapped terms, including: Match the mapped terms with the terms with a correlation of category 1 in the knowledge graph to obtain the mapping results; If no term with a correlation of Class 1 is matched, the mapped term is segmented, and the segmentation result is matched with the terms with a correlation of Class 1 and Class 2 in the knowledge graph to obtain the mapping result.

[0030] In one example, word segmentation processing methods include a forward maximum matching method, a reverse maximum matching algorithm, a bidirectional maximum matching method, and a segmentation mark setting method.

[0031] In one example, the word segmentation results are matched with the terms in the knowledge graph with the correlation of category 1 and category 2, and the mapping results obtained include: Each entry in the word segmentation result is matched with the terms in the knowledge graph with correlation of category 1 and category 2, and the matched results are weighted to obtain the relevance. The matching result with the highest relevance is taken as the final matching result.

[0032] In one example, after obtaining the mapping results, semi-automatic mapping is performed on the new terms in the mapped term set based on the knowledge graph, and the mapping results corresponding to the new terms include: Match the new entry with the term ID in the mapping result and display the matching mapping results; If there is no matching result with the term id, it will be matched with the term name in the mapping result, and the matching mapping result will be displayed; If there is no matching result with the term id or term name, the new term is segmented, and the segmentation result is matched with the terms with a correlation of category 1 in the knowledge graph, and the matching mapping results are displayed.

[0033] Specifically, the current medical terminology is scattered, and the differences in the structures of various types of terms make it difficult to maintain, resulting in uneven mapping results. When the in-hospital terminology dictionary changes, the new terms cannot automatically complete the mapping. A unified knowledge graph is conducive to the terminology system used in all systems. The current term mapping is mainly based on the mapping of existing term sets, and a corresponding result will be returned in the end, but the premise is that the structure of the mapped term and the mapped term is basically the same. Considering the obvious differences between multiple systems (such as the large differences between knowledge base terms and clinical terms), the traditional method does not solve the problems of big and small concepts, set tests, and combined concepts, and returns empty when there is no complete mapping result. The internal logical relationship of the knowledge graph can partially solve the above problems. In the application of clinical information systems, if the terms change, such as additions, modifications, etc., the original mapping will become invalid or the new mapping cannot be found, and the term mapping needs to be manually maintained again. The mapping relationship based on the knowledge graph, in addition to using coding as the basis for mapping, also uses semantics, knowledge relationships and other methods to supplement, thereby realizing semi-automatic technology for graph mapping and reducing the labor cost of term mapping maintenance.

[0034] Determine the basic architecture type of the knowledge graph, including term type, term code, term name, term relationship, attribute value, and relevance. Construct different types of term types, including disease, medicine, examination, test, surgery, etc. Construct different types of term relationships and attribute values, including standard names, aliases, categories, and unique term relationships.

[0035] The importance of term relationship values ​​is divided. Category 1 attributes are important attributes. If they match, it means a complete match. Category 2 attributes are auxiliary attributes. They need to be matched after the terms are disassembled. Relevance mapping is performed according to the matching degree of different parts. Category 3 attributes are related attributes, which are further supplementary and extended to the mapping results. As shown in Table 1, a knowledge graph is obtained.

[0036] Table 1 .

[0037] Figure 2 A flowchart of medical terminology mapping based on a knowledge graph according to an embodiment of the present invention is shown.

[0038] Based on the knowledge graph, the terms in the mapped term set (term set A) are compared, and the word segmentation technology is used to split the terms in the term set A, and then compared with the terms and term attributes in the knowledge graph. The most relevant matching items are obtained through feature word matching. If there is no final matching item, the related or similar items are determined by associating with the superordinate category (disease, set of tests, examination site). Finally, the mapping results are manually reviewed and used for knowledge graph optimization, such as Figure 2 shown.

[0039] The term set A is formatted and standardized to remove the term interference structure and special punctuation marks. The term is matched with the 1st attribute of the knowledge graph. If the feedback result is true, the mapping term code and term name are marked.

[0040] If there is no direct mapping, the term set A is segmented. The segmentation methods include forward maximum matching method, reverse maximum matching algorithm, bidirectional maximum matching method, and setting segmentation mark method. Considering that medical terms are not short texts, this method adopts bidirectional maximum matching method. For example, carotid endarterectomy can be divided into "carotid artery\endarterectomy". The segmentation results are compared with the 1st and 2nd attributes of the knowledge graph. The 1st attribute is prioritized to determine the subject concept of the term, and then compared with the 2nd attribute. The results of the 1st and 2nd attributes are weighted and finally a relevance is obtained. The relevance is sorted and the term mapping result with the highest relevance is taken. For example, carotid endarterectomy can be directly mapped with the 1st attribute, and the corresponding matching degree is 100. After segmentation, carotid endarterectomy can be divided into two terms, carotid artery and stripping, with matching degrees of 10 respectively, and 20 after weighting. Therefore, carotid endarterectomy is directly mapped with the 1st attribute, as shown in Table 2.

[0041] Table 2 type Hospital ID Hospital Name Term ID Term Name disease Yyjb0001 Cerebral infarction JB0001 Cerebral infarction disease Cerebral infarction Cerebral infarction JB0001 Cerebral infarction drug Yyyp0001 Aspirin tablets YP0001 Aspirin tablets drug Aspirin tablets Aspirin tablets YP0001 Aspirin tablets examine Yyjc0001 CT scan of the head JC0001 CT scan of the head examine CT scan of the head CT scan of the head JC0001 CT scan of the head Operation Yyss0001 Carotid endarterectomy SS0001 Carotid endarterectomy Operation Carotid endarterectomy Carotid endarterectomy SS0001 Carotid endarterectomy .

[0042] If there is still no mapping result, consider the upper and lower classes according to different term types, such as diseases, and use the upper class filling method to map. Manual review is performed on the mapping results, and the wrong mapping results are used for knowledge graph optimization.

[0043] Figure 3 A schematic diagram showing semi-automatic mapping of medical terms according to an embodiment of the present invention is shown.

[0044] like Figure 3 As shown, in the mapping term application, based on the knowledge graph and the mapping results, semi-automatic mapping can be performed for the dynamic term set A, so as to ensure that automatic matching can be performed after the term changes, additions or modifications, thereby increasing the fault tolerance rate.

[0045] The medical terminology mapping results are imported into the dictionary management system backend, and the results are output by calling the terms during the clinical information system call process.

[0046] During the system call, the match is performed based on the id of term set A. If there is a mapping relationship between the id of term set A and the knowledge graph in the dictionary background, the mapping result is called for content display. If there is no direct mapping relationship, the name of term set A is used for matching. If there is a mapping relationship between the name of term set A and the knowledge graph in the dictionary background, the mapping result is called for content display. This method is compatible with the addition of new standardized terms and can output the corresponding mapping results without manual operation.

[0047] If there is still no mapping relationship, the term name is converted into a search term, and the keyword stems are extracted, combined and searched, and the search results are returned.

[0048] If the mapping result is still empty, feedback can be provided through the feedback page, and manual judgment can be made to optimize the knowledge graph. After the knowledge graph is updated, it can take effect directly.

[0049] The present disclosure provides an electronic device, which includes: a memory storing executable instructions; and a processor executing the executable instructions in the memory to implement the above-mentioned medical terminology mapping method based on knowledge graph.

[0050] An electronic device according to an embodiment of the present disclosure includes a memory and a processor.

[0051] The memory is used to store non-temporary computer-readable instructions. Specifically, the memory may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0052] The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of the present disclosure, the processor is used to execute the computer-readable instructions stored in the memory.

[0053] Those skilled in the art should be able to understand that in order to solve the technical problem of how to obtain a good user experience, the present embodiment may also include well-known structures such as a communication bus and an interface, and these well-known structures should also be included in the protection scope of the present disclosure.

[0054] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0055] An embodiment of the present disclosure provides a computer-readable storage medium storing a computer program, which implements the medical terminology mapping method based on the knowledge graph when executed by a processor.

[0056] According to the computer-readable storage medium of the embodiment of the present disclosure, non-transitory computer-readable instructions are stored thereon. When the non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the above-mentioned methods of each embodiment of the present disclosure are executed.

[0057] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (e.g., memory card) and media with built-in ROM (e.g., ROM box).

[0058] Those skilled in the art should understand that the purpose of the above description of the embodiments of the present invention is only to exemplarily illustrate the beneficial effects of the embodiments of the present invention, and is not intended to limit the embodiments of the present invention to any given examples.

[0059] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A medical term mapping method based on knowledge graph, characterized in that: include: Determine the basic architecture type of the knowledge graph of medical terms, and then assign relevance values ​​to obtain the knowledge graph; Perform format standardization on the mapped term set, match it with the knowledge graph, and obtain the mapping results of the mapped terms; After obtaining the mapping results, semi-automatic mapping is performed on the new terms in the mapped term set based on the knowledge graph to obtain the mapping results corresponding to the new terms.

2. The medical terminology mapping method based on knowledge graph according to claim 1, wherein: The basic architecture types of knowledge graphs include term type, term encoding, term name, term relationship, and attribute value; Among them, the term types include disease, medicine, examination, test, and surgery.

3. The medical terminology mapping method based on knowledge graph according to claim 2, wherein: Relevance is the degree of match between the term name and the attribute value.

4. The medical terminology mapping method based on knowledge graph according to claim 3, wherein: According to the matching degree from high to low, the corresponding correlations are divided into category 1, category 2, and category 3, wherein the matching degree corresponding to category 1 is greater than the matching degree corresponding to category 2 and the matching degree corresponding to category 3.

5. The medical terminology mapping method based on knowledge graph according to claim 4, wherein: The mapped term set is formatted and matched with the knowledge graph to obtain the mapping results of the mapped terms, including: Match the mapped terms with the terms with a correlation of category 1 in the knowledge graph to obtain the mapping results; If no term with a correlation of Class 1 is matched, the mapped term is segmented, and the segmentation result is matched with the terms with a correlation of Class 1 and Class 2 in the knowledge graph to obtain the mapping result.

6. The medical terminology mapping method based on knowledge graph according to claim 5, wherein: Word segmentation processing methods include forward maximum matching method, reverse maximum matching algorithm, bidirectional maximum matching method, and setting segmentation mark method.

7. The medical terminology mapping method based on knowledge graph according to claim 5, wherein: According to the word segmentation results, the terms with correlations of category 1 and category 2 in the knowledge graph are matched to obtain the mapping results including: Each entry in the word segmentation result is matched with the terms in the knowledge graph with correlation of category 1 and category 2, and the matched results are weighted to obtain the relevance. The matching result with the highest relevance is taken as the final matching result.

8. The medical terminology mapping method based on knowledge graph according to claim 1, wherein: After obtaining the mapping results, semi-automatic mapping is performed based on the knowledge graph for the new terms in the mapped term set, and the mapping results corresponding to the new terms include: Match the new entry with the term ID in the mapping result and display the matching mapping results; If there is no matching result with the term id, it will be matched with the term name in the mapping result, and the matching mapping result will be displayed; If there is no matching result with the term id or term name, the new term is segmented, and the segmentation result is matched with the terms with a correlation of category 1 in the knowledge graph, and the matching mapping results are displayed.

9. An electronic device, characterized in that: The electronic device comprises: A memory storing executable instructions; A processor, wherein the processor runs the executable instructions in the memory to implement the medical terminology mapping method based on the knowledge graph as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the medical terminology mapping method based on a knowledge graph as described in any one of claims 1 to 8.

Images