International-oriented traditional Chinese medicine operation standardized coding and multi-language mutual recognition method

Abstract

The invention relates to the technical field of medical information, in particular to an international-oriented traditional Chinese medicine operation standardized coding and multi-language mutual recognition method, which comprises the following steps of: constructing a multi-language parallel corpus; extracting traditional Chinese medicine nursing operation terms, generating hierarchical standardized codes, and associating multi-language translation names to form a term library; establishing a mapping relationship between the code and an international general medical term library; generating a standardized data packet through a multi-center mutual recognition service based on the mapping relationship; and carrying out block chain evidence storage on the key operation. According to the method, the problem that cross-border mutual recognition of traditional Chinese medicine nursing operation is difficult due to non-standard terms, lack of multi-language support and disjunction from international standards is solved, accurate understanding, efficient mutual recognition and credible tracing of traditional Chinese medicine nursing service in a cross-border scene are achieved, and internationalized popularization of traditional Chinese medicine service is powerfully supported.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, and in particular to a standardized coding and multilingual mutual recognition method for traditional Chinese medicine operations oriented towards international standards. Background Technology

[0002] The demand for cross-border TCM medical and nursing services is growing, especially in the Southeast Asian market.

[0003] However, the internationalization of TCM nursing faces three major obstacles: Firstly, terminology is non-standard and isolated. The names of TCM nursing procedures such as moxibustion, bloodletting, cupping, and medicated baths contain dialectal and habitual abbreviations, and lack a unified, machine-readable coding system. Existing standards are mostly focused on Chinese herbal medicine and diagnosis, with a very weak coding system specifically for "nursing procedures."

[0004] Lack of multilingual support: Most existing research or systems only support Chinese and English, lacking systematic support for less common languages. This fails to meet the understanding needs of medical staff and patients in the target market, becoming a core bottleneck for cross-border promotion.

[0005] Disconnected from international standards and difficult to mutually recognize: Traditional Chinese medicine nursing procedures lack effective mapping with mainstream international medical information standards such as SNOMEDCT and ICD-11. In scenarios such as cross-border consultations, referrals, and insurance claims, the lack of reliable, multilingual standardized codes and mutual recognition protocols makes it impossible to quantify, verify, and recognize services, resulting in inefficiency and a high risk of disputes.

[0006] Therefore, there is an urgent need to provide an internationally oriented standardized coding and multilingual mutual recognition method and system for TCM operations to solve the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide a standardized coding and multilingual mutual recognition system and method for TCM nursing operations oriented towards international certification, so as to solve the problems existing in the prior art.

[0008] To achieve the above objectives, this invention provides an internationally oriented method for standardized coding and multilingual mutual recognition of Traditional Chinese Medicine (TCM) procedures, comprising the following steps: constructing a multilingual parallel corpus, including acquiring multilingual medical knowledge data, performing semantic similarity calculations, assembling parallel corpora, and translating missing languages; extracting medical terms from the multilingual parallel corpus, generating hierarchical standardized codes, and associating them with multilingual translations to form a multilingual medical terminology database; establishing a mapping relationship between the hierarchical standardized codes and a general medical terminology database, and generating standardized data packets based on the mapping relationship through a multi-center mutual recognition service; and storing the above operational steps on a blockchain.

[0009] Furthermore, the multilingual medical knowledge data originates from the source database of the front-end server of the multi-center medical data center, open datasets in the medical field, and datasets from medical social platforms. The data formats include electronic medical records, medical operation records, and diagnostic reports. Before calculating semantic similarity, sentences that do not contain medical terms or whose proportion of medical terms is lower than a preset ratio are filtered out. The translation of the missing language includes: prioritizing the direct translation of a translation model with a translation accuracy greater than a first preset value; if no suitable translation model exists, one of the remaining missing languages ​​is selected as an intermediate language, and the intermediate language sentence is first translated using existing language sentences in the parallel corpus, and then the currently selected missing language sentence is translated using the intermediate language sentence; it is determined whether the semantic similarity between the translated missing language sentence and the original existing language sentence is greater than a second preset value, and sentences with semantic similarity that meet the requirements are retained.

[0010] Furthermore, when extracting medical terms from the multilingual parallel corpus, the sentences in each parallel corpus are segmented into words, and non-medical terms are removed according to the corresponding language's medical terminology dictionary to extract medical terms. The word segmentation process adapts to different language characteristics using appropriate tools: Chinese uses the Jieba word segmentation tool for precise segmentation, while English and French use word segmentation tools based on root and suffix features. When removing non-medical terms, if the similarity between the segmented word and the medical terminology dictionary is lower than a third preset value, it is determined to be a non-medical term and removed. Moreover, for less commonly spoken languages, a unified character encoding conversion algorithm is used to convert different format character encodings to UTF-8 standard encoding, and a semantic feature-based word segmentation and synonym expansion algorithm is used to achieve fuzzy matching and synonym expansion retrieval of medical terms in less commonly spoken languages.

[0011] Furthermore, when establishing mapping relationships, a mapping rule base is established. Multidimensional features of hierarchical standardized coding of medical terms are extracted through a semantic mapping engine. At the same time, semantic features of corresponding terms in a general medical terminology database are extracted. A feature matching model is established, and feature similarity is calculated to determine the optimal mapping relationship. The optimal mapping relationship is then corrected by combining the prior knowledge of medical experts. Medical terminology mapping relationships that are not recognized by the model are manually added. After being jointly reviewed and approved by multiple medical experts, they are included in the mapping rule base.

[0012] Furthermore, in the multi-center mutual recognition service, the multi-center medical business scenarios include cross-border emergency consultations, emergency referrals, and cross-border insurance claims, which are divided into regular medical data sharing. Priorities are set for mutual recognition requests in different scenarios: cross-border emergency consultations and emergency referrals are set to the highest priority, are processed first, and the response time does not exceed the preset time; requests in other scenarios are set to the regular priority and are processed in the order of request reception.

[0013] Furthermore, it also includes a data cleaning step: formulating cleaning rules, assigning weights to each data element in the multilingual parallel corpus and the multilingual medical terminology database, and filtering out severely missing structural hierarchical dirty data and instance hierarchical dirty data.

[0014] Furthermore, it also includes an incremental update step: for the source database that has completed the terminology standardization mapping, when incremental medical data is generated, the historical mapping relationship records are retrieved to complete the terminology standardization mapping of the incremental data; if the medical terms in the incremental data cannot match the historical mapping relationship, the steps of extracting medical terms and generating hierarchical standardization codes and establishing mapping relationships are repeated to complete the encoding generation and mapping of new terms.

[0015] Furthermore, it also includes exception handling steps: recording the execution process of each module, classifying and saving error logs and unmapped medical terms, generating a document of failed terms; setting timestamps in the database to support backtracking the matched database to data from a specified date.

[0016] This invention also discloses a standardized coding and multilingual mutual recognition system for TCM nursing operations oriented towards international certification. The system includes: a multilingual corpus and coding management module, used to construct a multilingual parallel corpus and manage the hierarchical standardized coding and multilingual terminology of TCM nursing operation terms; and a general terminology mapping management module, used to maintain the mapping relationship between the hierarchical standardized coding and the general medical terminology database. The multi-center mutual recognition service module is used to receive and process terminology mutual recognition requests and generate standardized data packets; the blockchain evidence storage module is used to store evidence of key operations; the data support module includes a multilingual TCM nursing terminology database for storing core data and an exception handling submodule for handling errors.

[0017] Furthermore, it also includes a data statistics and analysis module, which is used to statistically analyze the number of term recognition requests, response time, and data accuracy, generate visual statistical reports, and optimize retrieval algorithms and processing rules based on the statistical results.

[0018] This invention provides a standardized coding and multilingual recognition method for TCM operations oriented towards international standards, which has the following beneficial effects: 1. By constructing a multilingual parallel corpus, designing a hierarchical standardized coding system, and adopting character encoding conversion and synonym expansion algorithms for minority languages, we have achieved unified, accurate expression and efficient retrieval of TCM nursing operation terms in a multilingual environment, directly overcoming the bottleneck of cross-border promotion caused by non-standard terminology and lack of support for minority languages.

[0019] 2. By establishing a semantic mapping engine based on multi-dimensional feature matching, combined with an expert knowledge review mechanism, and constructing a multi-center mutual recognition service based on the FHIR standard, dynamic, accurate association and instant mutual recognition of TCM nursing operation codes and international standard terminology have been achieved, effectively breaking down the barriers of disconnection from international standards and significantly improving the efficiency of cross-border business processing.

[0020] 3. By introducing a blockchain-based evidence storage mechanism to record key operations and designing incremental update and anomaly handling processes, the entire process of terminology standardization and mutual recognition is made traceable, tamper-proof, and continuously optimized. This ensures the data credibility and long-term operational robustness of the system, laying a solid technical trust foundation for the international recognition of TCM nursing services. Attached Figure Description

[0021] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 The present invention provides an overall flowchart of a standardized coding and multilingual mutual recognition method for TCM operations oriented towards internationalization. Figure 2 This invention provides a schematic diagram of a standardized coding and multilingual mutual recognition system for traditional Chinese medicine operations oriented towards international standards. Detailed Implementation

[0022] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0023] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0024] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0025] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0026] This embodiment provides an internationally oriented method for standardized coding and multilingual mutual recognition of Traditional Chinese Medicine (TCM) procedures, comprising the following steps: constructing a multilingual parallel corpus, including acquiring multilingual medical knowledge data, performing semantic similarity calculations, assembling parallel corpora, and translating missing languages; extracting medical terms from the multilingual parallel corpus, generating hierarchical standardized codes, and associating them with multilingual translations to form a multilingual medical terminology database; establishing a mapping relationship between the hierarchical standardized codes and a general medical terminology database, and generating standardized data packets based on the mapping relationship through a multi-center mutual recognition service; and storing the above operational steps on a blockchain.

[0027] The following is in conjunction with the appendix Figure 1 The complete workflow of this invention is illustrated using a cross-border insurance claims scenario between a hospital in Hainan and a Thai insurance company as an example: Specifically, the multilingual medical knowledge data originates from the source database of the front-end server of a multi-center medical data center, open datasets in the medical field, and datasets from medical social platforms. The data formats include electronic medical records, medical operation records, and diagnostic reports. Before semantic similarity calculation, sentences that do not contain medical terms or whose proportion of medical terms is lower than a preset ratio are filtered out. The translation of the missing language includes: prioritizing the use of a translation model with a translation accuracy greater than a first preset value for direct translation; if no suitable translation model exists, one of the remaining missing languages ​​is selected as an intermediate language, and the intermediate language sentence is first translated using existing language sentences in the parallel corpus, and then the currently selected missing language sentence is translated using the intermediate language sentence; it is then determined whether the semantic similarity between the translated missing language sentence and the original existing language sentence is greater than a second preset value, and sentences with semantic similarity that meet the requirements are retained.

[0028] Therefore, the multilingual medical knowledge data originates from the source database of the front-end server of the Hainan Traditional Chinese Medicine Hospital's data center. This database includes nearly three years of electronic medical records, TCM nursing operation records such as "medicinal ironing therapy" and "moxibustion therapy" operation records, open datasets in the international TCM nursing field such as the World Health Organization's Traditional Medicine Database, and relevant datasets from Southeast Asian medical social platforms such as the Thailand Medical and Health Forum. The data format includes text records in Chinese, English, Thai, and Vietnamese. The first preset value for translation accuracy is 85%, the second preset value for semantic similarity is 90%, and the third preset value for terminology similarity is 80%. The response time for emergency scenarios such as cross-border emergency consultations and emergency referrals is preset to 3 seconds.

[0029] It is easy to understand that through the above steps, the construction of a multilingual parallel corpus can be completed, data preparation can be achieved, and a foundation can be provided for identifying TCM nursing operation terms in it through natural language processing technology and performing structured analysis based on a preset multidimensional feature system.

[0030] Specifically, when extracting medical terms from the multilingual parallel corpus, the sentences in each parallel corpus are segmented into words, and non-medical terms are removed according to the corresponding language's medical terminology dictionary before extracting medical terms. The word segmentation process adapts to different language characteristics using appropriate tools: Chinese uses the Jieba word segmentation tool for precise segmentation, while English and French use word segmentation tools based on root and suffix features. When removing non-medical terms, if the similarity between the segmented word and the medical terminology dictionary is lower than a third preset value, it is determined to be a non-medical term and removed. Furthermore, for less commonly spoken languages, a unified character encoding conversion algorithm is used to convert different format character encodings to UTF-8 standard encoding, and a semantic feature-based word segmentation and synonym expansion algorithm is used to achieve fuzzy matching and synonym expansion retrieval of medical terms in less commonly spoken languages.

[0031] Therefore, the acquired multilingual medical knowledge data is initially screened to remove sentences that do not contain TCM nursing terminology or whose terminology accounts for less than 30%, such as medical record fragments that only record basic patient information. The BERT semantic model is used to calculate the semantic similarity of the remaining sentences. The semantic similarity between Chinese sentences and English sentences is calculated. If the result is greater than the second preset value of 90%, the two are combined into a parallel corpus.

[0032] If the relevant parallel corpus only contains Chinese and English, lacking Thai and Vietnamese, the translation model library is queried first. If the translation model's accuracy is greater than the first preset value of 85%, the Chinese sentence is directly translated into Thai using that model. If the accuracy of the "Chinese-Vietnamese" translation model is 82% or less, English is selected as the intermediate language. The English sentence is first translated into Vietnamese, and then the semantic similarity between the Vietnamese sentence and the original Chinese sentence is calculated. If the similarity is greater than 90%, the Vietnamese sentence is retained. Finally, parallel corpora containing Chinese, English, Thai, and Vietnamese are constructed to form a multilingual parallel corpus.

[0033] Specifically, when establishing mapping relationships, a mapping rule base is established. Multidimensional features of hierarchical standardized coding of medical terms are extracted through a semantic mapping engine. At the same time, semantic features of corresponding terms in a general medical terminology database are extracted. A feature matching model is established, and feature similarity is calculated to determine the optimal mapping relationship. The optimal mapping relationship is then corrected by combining the prior knowledge of medical experts. Medical terminology mapping relationships that are not recognized by the model are manually added and incorporated into the mapping rule base after being jointly reviewed and approved by multiple medical experts.

[0034] Therefore, during the mapping relationship establishment phase, the semantic mapping engine extracts the multi-dimensional features encoded in "medicinal ironing therapy" and simultaneously searches for similar concepts in SNOMED CT. The feature matching model calculates that the highest similarity is found with "Hot pack treatment (procedure)," reaching 0.91. After joint review and confirmation of the mapping relationship's validity by TCM experts and standardization experts, the mapping relationship "TNE-001-2023→SNOMED CT:312124000" is stored in the mapping rule base. The creator, creation time, review records, and other information of this mapping relationship are generated into a hash value and uploaded to the consortium blockchain for evidence storage.

[0035] In one preferred embodiment, three experts in the field of traditional Chinese medicine nursing, including one expert in traditional Thai medicine, are invited to review the above mapping relationship. If the experts suggest that "the difference between the materials of 'medicated ironing therapy' and 'Herbal pack therapy' needs to be supplemented", the feature similarity is recalculated after correction and is 92%. After the three experts unanimously approve the review, the mapping relationship is then included in the mapping rule base, completing the expert review and mapping entry into the base.

[0036] Specifically, in the multi-center mutual recognition service, the multi-center medical business scenarios include cross-border emergency consultations, emergency referrals, and cross-border insurance claims, which are divided into regular medical data sharing. Priorities are set for mutual recognition requests in different scenarios: cross-border emergency consultations and emergency referrals are set to the highest priority, are processed first, and the response time does not exceed the preset time; requests in other scenarios are set to the regular priority and are processed in the order of request reception.

[0037] Therefore, during the business processing phase, when a hospital in Hainan performs "medicinal ironing therapy" on a Thai patient, it records the operation code "TNE-001-2023" in the EMR system. The Thai insurance company's claims system sends a request containing the operation code, target language, and business scenario to the multilingual mutual recognition service API of this invention. After receiving the request, the API gateway identifies the business scenario as insurance claims and enters the processing queue. The system queries the multilingual medical terminology database, obtains the corresponding Thai definition and SNOMED CT code, and generates a Bundle resource conforming to the FHIR standard, returning it to the requester. The request content, response result, processing timestamp, and other information of this mutual recognition operation generate a new hash value, which is appended to the blockchain. The system also records the frequency of term usage in this operation, and automatically optimizes the relevant retrieval algorithm when the frequency reaches a threshold.

[0038] Specifically, it also includes data cleaning steps: formulating cleaning rules, assigning weights to each data element in the multilingual parallel corpus and the multilingual medical terminology database, and filtering out severely missing structural and instance-level dirty data.

[0039] Therefore, the system regularly performs cleaning rules on the multilingual parallel corpus, including correcting issues such as field type errors and constraint violations at the structural level, removing records with a missing rate exceeding 15% at the instance level, merging duplicate medical records, and retaining data with a quality score higher than 0.8 based on data element weight scoring. When a hospital adds a new nursing operation, the system automatically triggers a terminology extraction process and generates new codes when it detects the absence of a term. For cases where international standards cannot be automatically matched, an expert review process is initiated, and a new mapping relationship is established after review by the Traditional Chinese Medicine Nursing Expert Committee. The system fully records the entire business processing log, including API reception time, database query time, mapping matching results, and blockchain storage status. If a dispute arises later, the data can be traced back to the state at the time of processing through timestamps.

[0040] It is easy to understand that this embodiment improves the accuracy of terminology mutual recognition, cross-border claims processing time, and multilingual terminology coverage, and establishes a responsibility traceability mechanism through blockchain evidence storage.

[0041] This invention also provides a system for implementing an internationally oriented standardized coding and multilingual mutual recognition method for Traditional Chinese Medicine (TCM) procedures. The system includes a multilingual corpus and coding management module for constructing a multilingual parallel corpus and managing hierarchical standardized coding and multilingual terminology for TCM nursing procedures; a general terminology mapping management module for maintaining the mapping relationship between the hierarchical standardized coding and a general medical terminology database; a multi-center mutual recognition service module for receiving and processing terminology mutual recognition requests and generating standardized data packets; a blockchain evidence storage module for storing evidence of key operations; and a data support module, including a multilingual TCM nursing terminology database for storing core data and an exception handling submodule for handling errors. It also includes a data statistical analysis module for statistically analyzing the number of terminology mutual recognition requests, response time, and data accuracy, generating visual statistical reports, and optimizing retrieval algorithms and processing rules based on the statistical results.

[0042] This demonstrates that the various system modules work collaboratively. The multilingual corpus and encoding management module provides an expert interface, supporting the addition, review, and multilingual translation management of terms; the general terminology mapping management module maintains mapping relationships and records all version change history; the multi-center mutual recognition service module processes API requests and prioritizes resources according to business scenarios; the blockchain evidence storage module ensures the immutability of all critical operations; and the data support module provides persistent storage and exception handling capabilities to guarantee high system availability.

[0043] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A standardized coding and multilingual recognition method for TCM procedures oriented towards international standards, characterized in that: Includes the following steps: Constructing a multilingual parallel corpus includes acquiring multilingual medical knowledge data, calculating semantic similarity, assembling parallel corpora, and translating missing languages; Medical terms are extracted from the multilingual parallel corpus, hierarchical standardized codes are generated, and multilingual translations are associated to form a multilingual medical terminology database. Establish a mapping relationship between the hierarchical standardized coding and the general medical terminology database, and generate standardized data packets based on the mapping relationship through a multi-center mutual recognition service; The above operational steps are recorded on the blockchain.

2. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, The multilingual medical knowledge data comes from the source database of the front-end server of the multi-center medical data center, open datasets in the medical field, and datasets from medical social platforms. The data formats include electronic medical records, medical operation records, and diagnostic reports. Before performing semantic similarity calculation, sentences that do not contain medical terms or whose proportion of medical terms is lower than the preset ratio are filtered out. The translation of the missing language includes: prioritizing the use of a translation model with a translation accuracy greater than a first preset value for direct translation; if no suitable translation model exists, then selecting one of the remaining missing languages ​​as an intermediate language, first translating the intermediate language sentence using existing language sentences in the parallel corpus, and then translating the currently selected missing language sentence using the intermediate language sentence; determining whether the semantic similarity between the translated missing language sentence and the original existing language sentence is greater than a second preset value, and retaining sentences with semantic similarity that meet the requirements.

3. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, When extracting medical terms from the multilingual parallel corpus, the sentences in each parallel corpus are segmented, and non-medical terms are removed according to the corresponding language's medical terminology dictionary to extract the medical terms; among which, The word segmentation process adapts to different language characteristics and corresponding tools: Chinese uses the Jieba word segmentation tool for accurate segmentation, while English and French use word segmentation tools based on root and suffix features; When removing non-medical terms, if the similarity between the segmented word and the medical terminology dictionary is lower than the third preset value, it is determined to be a non-medical term and removed. Furthermore, for minority languages, a unified character encoding conversion algorithm is used to convert different format character encodings to UTF-8 standard encoding, and a semantic feature-based segmentation and synonym expansion algorithm is used to achieve fuzzy matching and synonym expansion retrieval of medical terms in minority languages.

4. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, When establishing mapping relationships, a mapping rule base is established, and multi-dimensional features of hierarchical standardized coding of medical terms are extracted through a semantic mapping engine. At the same time, semantic features of corresponding terms in a general medical terminology base are extracted, a feature matching model is established, and feature similarity is calculated to determine the optimal mapping relationship. The optimal mapping relationship is modified by combining the prior knowledge of medical experts. Medical terminology mapping relationships that are not recognized by the model are manually added and incorporated into the mapping rule base after being jointly reviewed and approved by medical experts from multiple centers.

5. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, In the multi-center mutual recognition service, the multi-center medical business scenarios include cross-border emergency consultations, emergency referrals, and cross-border insurance claims, which are divided into regular medical data sharing. Priorities are set for mutual recognition requests in different scenarios. Among them, cross-border emergency consultation and emergency referral requests are set to the highest priority, are processed first, and the response time does not exceed the preset time. Requests in other scenarios are set to the regular priority and are processed in the order of receiving the requests.

6. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, It also includes a data cleaning step: formulating cleaning rules, assigning weights to each data element in the multilingual parallel corpus and the multilingual medical terminology database, and filtering out severely missing structural and instance-level dirty data.

7. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, It also includes an incremental update step: For the source database that has completed the terminology standardization mapping, when incremental medical data is generated, the historical mapping relationship records are retrieved to complete the terminology standardization mapping of the incremental data; if the medical terms in the incremental data cannot match the historical mapping relationship, the steps of extracting medical terms and generating hierarchical standardization codes and establishing mapping relationships are repeated to complete the coding generation and mapping of new terms.

8. The internationally oriented standardized coding and multilingual mutual recognition method for TCM operations according to claim 1, characterized in that, It also includes exception handling steps: recording the execution process of each module, classifying and saving error logs and unmapped medical terms, and generating a document of failed terms; setting timestamps in the database to support backtracking the matched database to data from a specified date.

9. A multi-center medical terminology standardization system based on a multilingual parallel corpus and a general medical terminology database for implementing the method of any one of claims 1-8, characterized in that, include: The multilingual corpus and coding management module is used to build a multilingual parallel corpus and manage the hierarchical standardized coding and multilingual terminology of TCM nursing operation terms. The general terminology mapping management module is used to maintain the mapping relationship between the hierarchical standardized codes and the general medical terminology database; The multi-center mutual recognition service module is used to receive and process terminology mutual recognition requests and generate standardized data packets; The blockchain-based evidence storage module is used to store evidence of key operations. The data support module includes a multilingual TCM nursing terminology database for storing core data and an exception handling submodule for handling errors.

10. The multi-center medical terminology standardization system based on a multilingual parallel corpus and a general medical terminology database according to claim 9, characterized in that, It also includes a data statistics and analysis module, which is used to count the number of term recognition requests, response time and data accuracy, generate visual statistical reports, and optimize retrieval algorithms and processing rules based on the statistical results.

Images