Medical knowledge graph construction method and device based on hybrid intelligent technology

Abstract

The invention provides a medical knowledge graph construction method and device based on a hybrid intelligent technology, and relates to the technical field of artificial intelligence and medical crossing knowledge graph construction, and the method comprises the steps: employing a hybrid intelligent model to recognize key entities in a medical knowledge material, extracting a semantic relation, and automatically constructing a knowledge graph; a preliminary proofreading suggestion is generated through an expert interaction proofreading mechanism driven by a large language model, an expert carries out adjustment and feedback on a visual interface, and meanwhile, a proofreading process is recorded and version management is carried out; accumulating expert proofreading information, re-importing the expert proofreading information into the training set, updating the hybrid intelligent model, circulating the steps, realizing man-machine collaboration to improve the model performance, and integrating expert experience into the knowledge graph. According to the method, the mixed intelligent technology is adopted to fuse artificial intelligence and expert cooperation, the high-precision medical knowledge graph is efficiently constructed, dynamic updating is supported, data dependence is reduced, and interpretability and cross-field applicability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of knowledge graph construction at the intersection of artificial intelligence and medicine, and in particular to a method and device for constructing a medical knowledge graph based on hybrid intelligent technology. Background Art

[0002] As an important method to express medical knowledge in a structured form, medical knowledge graphs are widely used in the medical field, such as supporting semantic query, reasoning, decision support, and intelligent question and answer. With the rapid development of artificial intelligence technology, the construction method of medical knowledge graphs has also evolved, from traditional manual and rule-driven to automation and intelligence, which has improved the construction efficiency to a certain extent, but there are still many problems.

[0003] The application of machine learning, natural language processing technology and deep learning models based on large-scale data enables medical knowledge graphs to automatically extract information from unstructured and semi-structured data, build entity relationships and semantic networks. However, medical data sources are complex and standards are not unified. These methods face huge challenges in data integrity, diversity and accuracy. In the construction of medical knowledge graphs, ensuring that the constructed knowledge graphs have high credibility and applicability has become an urgent problem to be solved.

[0004] At present, various single technical solutions have significant defects in the construction of medical knowledge graphs: 1. Rule-based methods: Extracting entities and relationships from structured and unstructured data through predefined rules and templates is relatively simple and interpretable, but it has many drawbacks. On the one hand, a large number of rules need to be manually defined, which is difficult to adapt to the rapid dynamic changes and diversified needs in the medical field; on the other hand, when faced with large-scale data, rule matching is inefficient, and the processing ability of unstructured data (such as clinical documents and scientific research papers) is also obviously insufficient.

[0005] 2. Machine learning-based methods: Using labeled data to train models to extract entities and relationships in text data, with high automation and strong adaptability. However, the high cost of obtaining labeled data in the medical field has severely limited the widespread application of this method. At the same time, the model is black-box and lacks transparency, making it difficult to understand and explain the decision-making process. In addition, the semantic understanding ability is insufficient, making it difficult to understand the complex semantics in medical texts, especially cross-domain or professional terms.

[0006] 3. Methods based on deep learning models (such as BERT and GPT): They are widely used in knowledge graph construction and have the advantages of strong semantic understanding and excellent large-scale data processing performance. However, these methods have extremely high training costs, require large-scale computing resources, and take a long time to train. In addition, the models are very sensitive to noise in the input data, easily spread false information, and lack supervision from domain knowledge. The output results may not be consistent with actual medical knowledge.

[0007] In general, the existing single technical solutions have laid the foundation for the construction of medical knowledge graphs, but their respective defects have limited their effectiveness and popularity in practical applications. The existing technology lacks an effective mechanism for human-computer collaboration, ignores the key role of human experts in verifying and supplementing medical knowledge, and is difficult to meet the strict requirements of the medical field for high data accuracy. At the same time, after the knowledge graph is built, it needs to be continuously updated to reflect the latest research results in the medical field, but the existing technology lacks effective means for automated updates and dynamic maintenance. Summary of the invention

[0008] To this end, an embodiment of the present invention provides a method and device for constructing a medical knowledge graph based on hybrid intelligent technology, which is used to solve the problems of data integrity and accuracy that exist in the prior art of constructing medical knowledge graphs with a single technology, as well as the lack of human-computer collaboration mechanism, difficulty in meeting high-precision requirements and insufficient dynamic update capabilities.

[0009] In order to solve the above problems, an embodiment of the present invention provides a method for constructing a medical knowledge graph based on hybrid intelligent technology, the method comprising: Use a hybrid intelligent model to identify medical entities and extract relationships, as well as build a knowledge graph. The hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model to identify key knowledge entities in medical knowledge materials and extract the semantic relationships between them, thereby realizing the automatic construction of a knowledge graph. Through the expert interactive proofreading mechanism driven by the big language model, the constructed knowledge graph is proofread and optimized. The big language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through the visual proofreading interface. At the same time, the proofreading process is recorded and versioned. The graph information proofread by experts is accumulated and re-imported into the training set, the hybrid intelligent model is updated, and the above steps are repeated. The performance of the hybrid intelligent model is continuously improved through human-computer collaboration, and the expert experience knowledge is absorbed into the knowledge graph.

[0010] Preferably, the use of a hybrid intelligent model for medical entity recognition and relationship extraction includes: Combine large language models with multimodal models to extract key medical entities from text data and image data, including disease names, drug information, onset sites, and abnormal descriptions; A mechanism based on the combination of time series models and text data is designed to achieve the entity extraction and alignment of key medical knowledge concepts, combine the descriptive relationships between medical entities mined from text, and achieve the alignment of medical entities and the association of their related relationships.

[0011] Preferably, the large language model generates preliminary proofreading suggestions, including: The large language model provides preliminary proofreading suggestions for each entity and relationship based on semantic analysis and relational reasoning, and marks low-confidence areas; The large language model outputs the logical basis for each proofreading suggestion, including the text source, reasoning path, and referenced knowledge base, providing decision-making support information for experts.

[0012] Preferably, the expert performs interactive adjustment and feedback through a visual proofreading interface, including: Experts can quickly understand the suggestions generated by the model and the basis behind them through a visual proofreading interface that displays nodes and relationships in a graphical knowledge graph; Experts directly adjust graph nodes and relationships, including adding, modifying or deleting entities and relationships, and provide feedback on each adjustment operation so that the hybrid intelligent model can absorb expert experience in subsequent training.

[0013] Preferably, the recording and proofreading process and performing version management include: Keep detailed records of every step and decision of proofreading, including operation time, knowledge entities or relationships involved, and revisions made by experts and their basis; After each proofreading, a new version of the knowledge graph is generated and compared with the previous version, marking the change points for subsequent tracking and review.

[0014] Preferably, the accumulation of the atlas information verified and confirmed by the experts and re-importing it into the training set to update the hybrid intelligent model includes: The expert-proofread graph information accumulated during the proofreading process is stored in the knowledge base as important data for further training the hybrid intelligent model; As feedback information accumulates, the hybrid intelligent model gradually optimizes its entity recognition and relationship extraction capabilities, reducing errors in subsequent construction and improving efficiency.

[0015] The embodiment of the present invention further provides a medical knowledge graph construction device based on hybrid intelligence technology, which is used to implement the above-mentioned medical knowledge graph construction method based on hybrid intelligence technology, specifically including: A hybrid intelligent model construction module is used to use a hybrid intelligent model to perform medical entity recognition and relationship extraction, as well as knowledge graph construction. The hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model to identify key knowledge entities in medical knowledge materials and extract the semantic relationships between them, thereby realizing the automatic construction of a knowledge graph. The expert interactive proofreading module is used to proofread and optimize the constructed knowledge graph through an expert interactive proofreading mechanism driven by a large language model. The large language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through a visual proofreading interface. At the same time, the proofreading process is recorded and versioned. The model updating module is used to accumulate the graph information proofread by experts and re-import it into the training set, update the hybrid intelligent model, repeat the above steps repeatedly, continuously improve the performance of the hybrid intelligent model through human-computer collaboration, and absorb expert experience knowledge into the knowledge graph.

[0016] Preferably, the hybrid intelligent model building module includes: Entity recognition and relationship extraction unit, which is used to extract key medical entities from text data and image data by combining large language models with multimodal models, and to design a mechanism based on the combination of time series models and text data to achieve entity extraction and alignment of key medical knowledge concepts; The knowledge graph construction unit is used to generate a structured knowledge graph after semantic classification and standardization of the extracted entities and relationships.

[0017] Preferably, the expert interactive proofreading module includes: The proofreading suggestion generation unit is used by the large language model to provide preliminary proofreading suggestions for each entity and relationship based on semantic analysis and relational reasoning, and to mark low-confidence areas, while outputting the logical basis for each proofreading suggestion; The visual proofreading interface unit is used to provide experts with a graphical display of knowledge graph nodes and relationships, allowing them to quickly understand the recommendations generated by the model and the basis behind them; An interactive adjustment and feedback unit, which allows experts to directly adjust graph nodes and relationships and provide feedback on each adjustment operation; The proofreading record and version management unit is used to record every step and decision of the proofreading in detail, and after each proofreading is completed, a new version of the knowledge graph is generated, compared with the previous version, and the change points are marked.

[0018] Preferably, the model updating module comprises: The knowledge accumulation unit is used to store the graph information proofread by experts accumulated during the proofreading process; The model optimization unit is used to gradually optimize the entity recognition and relationship extraction capabilities of the hybrid intelligent model as feedback information accumulates, reduce errors in subsequent construction, and improve efficiency.

[0019] It can be seen from the above technical solutions that the present invention has the following beneficial effects: (1) The present invention combines multiple artificial intelligence models and expert knowledge through hybrid intelligent technology to achieve automated information extraction and structured processing, significantly reducing manual operation time, while reducing dependence on large-scale labeled data, reducing data collection and labeling costs, and improving the scalability of the technology.

[0020] (2) The present invention introduces an expert feedback mechanism to optimize the model prediction results, reduce the impact of data noise and erroneous information, ensure the high consistency between the knowledge graph and medical field knowledge, and improve the credibility of the knowledge graph.

[0021] (3) The present invention combines the semantic understanding ability of deep learning models with the domain knowledge of experts to solve the problem of insufficient understanding of complex medical terminology and cross-domain knowledge, and supports the dynamic update and real-time maintenance of knowledge graphs, so that it can reflect the latest research progress in the medical field in real time and meet the ever-changing practical needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the implementation cases of the present invention or the technical solutions in the prior art, the following is a brief description of the drawings required for use in the embodiments. By referring to the drawings, the features and advantages of the present invention will be more clearly understood. The drawings are schematic and should not be understood as limiting the present invention in any way. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them: Figure 1 A flowchart of a method for constructing a medical knowledge graph based on hybrid intelligent technology provided in an embodiment; Figure 2 A block diagram of a medical knowledge graph construction device based on hybrid intelligent technology provided in an embodiment. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Embodiment 1

[0024] In order to solve the problems of data integrity and accuracy in the existing single technology for building medical knowledge graphs, as well as the lack of human-machine collaboration mechanism, difficulty in meeting high-precision requirements and insufficient dynamic update capabilities. Figure 1 As shown, the embodiment of the present invention proposes a method for constructing a medical knowledge graph based on hybrid intelligent technology, the method comprising: Use hybrid intelligent models to identify medical entities and extract relationships, as well as build knowledge graphs. The hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model. It identifies key knowledge entities in medical knowledge materials and extracts the semantic relationships between them to achieve automatic construction of knowledge graphs. Through the expert interactive proofreading mechanism driven by the big language model, the constructed knowledge graph is proofread and optimized. The big language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through the visual proofreading interface. At the same time, the proofreading process is recorded and versioned. Accumulate the graph information proofread by experts and re-import it into the training set, update the hybrid intelligent model, and repeat the above steps repeatedly. Through human-computer collaboration, the performance of the hybrid intelligent model is continuously improved, and the expert experience and knowledge is absorbed into the knowledge graph.

[0025] As can be seen from the above technical solution, the present invention proposes a method for constructing a medical knowledge graph based on hybrid intelligent technology, which uses a hybrid intelligent model to perform medical entity recognition and relationship extraction, and construct a knowledge graph. This step can efficiently identify key entities and extract semantic relationships from medical knowledge materials through the collaborative work of multiple AI models, realize the automated construction of knowledge graphs, significantly improve construction efficiency, and reduce the time consumption caused by manual operations in traditional methods. The constructed knowledge graph is proofread and optimized through an expert interactive proofreading mechanism driven by a large language model. The large language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through a visual proofreading interface, while recording the proofreading process and performing version management. This step introduces the participation of human experts to ensure the high consistency of the model output results with the knowledge in the medical field, optimize the model prediction results, reduce the errors caused by data noise and erroneous information, and ensure the high credibility of the knowledge graph. The graph information proofread by experts is accumulated and re-imported into the training set, the hybrid intelligent model is updated, and the above steps are repeated. The performance of the hybrid intelligent model is continuously improved through human-machine collaboration, and the expert experience knowledge is absorbed into the knowledge graph. This step uses a dynamic update mechanism to enable the knowledge graph to reflect the latest research progress in the medical field in real time, meet the ever-changing actual needs, and reduce the dependence on large-scale labeled data, reduce data collection and annotation costs, and improve the scalability of the technology. In summary, the present invention effectively overcomes many deficiencies in the prior art through the innovative application of hybrid intelligent technology, and provides a new technical solution for the efficient, accurate and dynamic construction of medical knowledge graphs, which has significant technical advancement and application value.

[0026] In this embodiment, the present invention utilizes a hybrid intelligent model to perform medical entity recognition and relationship extraction, as well as knowledge graph construction, wherein the hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model, which identifies key knowledge entities in medical knowledge materials and extracts the semantic relationships therebetween, thereby realizing the automatic construction of a knowledge graph.

[0027] Specifically, data preparation is first performed: text data (such as medical literature, clinical records) and image data (such as medical images, waveforms, etc.) in the medical field are collected. Text data is preprocessed, including operations such as word segmentation, part-of-speech tagging, and stop word removal; image data is preprocessed, including operations such as image cropping, normalization, and enhancement.

[0028] Furthermore, the model is constructed: a hybrid intelligent model is constructed, including a large language model, a multimodal model, a time series model and a large image model. The large language model is used to extract key medical entities and their semantic relationships from text data; the multimodal model is used to extract key medical entities and their features from image data; the time series model is used to analyze the time-varying features of medical data, such as the development of the disease, the changes in the efficacy of the drug, etc.; the large image model is used to extract the visual features of medical entities from image data.

[0029] Furthermore, a hybrid intelligent model is used for medical entity recognition and relationship extraction, including: Combine large language models with multimodal models to extract key medical entities from text data and image data, including disease names, drug information, onset sites, and abnormal descriptions; A mechanism based on the combination of time series models and text data is designed to achieve the entity extraction and alignment of key medical knowledge concepts, combine the descriptive relationships between medical entities mined from text, and achieve the alignment of medical entities and the association of their related relationships.

[0030] Specifically, by combining the large language model with the multimodal model, key medical entities are extracted from the preprocessed text data and image data. The large language model identifies key medical entities in the text, such as disease names, drug information, onset sites, abnormal descriptions, etc., through semantic analysis and context understanding, and extracts semantic relationships between entities, such as causal relationships, association relationships, etc. The multimodal model identifies key medical entities in the image, such as lesion areas, organ structures, etc., through feature extraction and fusion, and aligns them with entities in the text. The time series model analyzes the temporal variation characteristics of medical data and provides information support in the time dimension for entity recognition and relationship extraction. The large image model extracts the visual features of medical entities in the image data and provides information support in the visual dimension for entity recognition and relationship extraction. Finally, the medical entities and their relationships extracted from the text and images are integrated to form a preliminary medical knowledge graph structure.

[0031] Furthermore, semantic classification and standardization are performed: the extracted medical entities and relationships are semantically classified and classified into predefined medical knowledge systems, such as disease classification, drug classification, etc. At the same time, the entities and relationships are standardized, and their naming specifications and formats are unified to ensure the consistency and scalability of the knowledge graph.

[0032] Generate a knowledge graph: Organize the semantically classified and standardized medical entities and relationships into a structured knowledge graph. The knowledge graph represents medical entities with nodes and the semantic relationships between entities with edges. The knowledge graph is stored and managed through technologies such as graph databases to support efficient query and reasoning operations.

[0033] In this embodiment, the constructed knowledge graph is proofread and optimized through an expert interactive proofreading mechanism driven by a large language model. The large language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through a visual proofreading interface. At the same time, the proofreading process is recorded and versioned.

[0034] Specifically, the large language model generates preliminary proofreading suggestions, including: the large language model performs preliminary proofreading on each entity and relationship in the constructed knowledge graph based on semantic analysis and relational reasoning, and marks low-confidence areas. The model will judge the accuracy of entities and relationships based on existing knowledge and logical rules, and generate preliminary proofreading suggestions for parts that may be erroneous or uncertain. For example, when the model recognizes that the name of an entity is inconsistent with a known standard name, it will generate modification suggestions; when the model finds that there is doubt about the reasoning path of a relationship, it will propose possible correction directions.

[0035] Furthermore, the output of logical basis includes: the large language model simultaneously outputs the logical basis of each proofreading suggestion, including the text source, reasoning path, and referenced knowledge base. The logical basis is presented in a clear text form to provide experts with detailed decision-making support information. For example, for a suggestion to modify an entity name, the model will explain what information in the text data the suggestion is based on and how the conclusion is reached through semantic analysis; for a suggestion to modify a relationship, the model will display the knowledge rules and data evidence based on the reasoning process.

[0036] In addition, the present invention also includes interface design: designing an intuitive visual proofreading interface to display the nodes and relationships of the knowledge graph in a graphical manner. The proofreading interface adopts an interactive design so that experts can easily browse and operate the knowledge graph. The interface includes a detailed information display area for entity nodes, a visual display area for relationships, a prompt area for proofreading suggestions, and an interactive area for expert operations.

[0037] Information display and understanding assistance: In the visual proofreading interface, different entity types and relationship types are distinguished by different colors, shapes or icons, allowing experts to quickly identify and understand the structure and content of the knowledge graph. At the same time, the interface provides functions such as highlighting text information and pop-up prompts for related information to help experts better understand the proofreading suggestions generated by the model and the basis behind them.

[0038] Furthermore, experts make interactive adjustments and provide feedback through a visual proofreading interface, including: Experts can directly adjust nodes and relationships in the knowledge graph through a visual proofreading interface. Experts can modify the names and attributes of entities, adjust the types and directions of relationships, and add or delete entities and relationships. For example, if an expert finds that the translation of a disease name is incorrect, he can modify it directly on the interface; if an expert believes that the relationship between a drug and a disease is inaccurate, he can adjust the type of relationship or delete the relationship.

[0039] Experts provide detailed feedback for each adjustment operation, explaining the reasons and basis for the adjustment. Feedback is entered through the interactive function of the interface and stored in association with the corresponding adjustment operation. For example, when an expert modifies an entity name, he or she will enter a text to explain the correct source and standard specifications of the name; when an expert adjusts a relationship, he or she will explain the medical basis and reasoning process of the adjustment.

[0040] Furthermore, the proofreading process is recorded and versioned, including: Keep a detailed record of each proofreading operation and decision of the expert, including the operation time, the knowledge entities or relationships involved, the expert's revision content and its basis. The record is stored in a structured data form to facilitate subsequent query and analysis. For example, the record will detail when the expert modified which entity, what the specific content before and after the modification was, and what the basis for the modification was.

[0041] After each proofreading, a new version of the knowledge graph is generated and compared with the previous version, marking the change points. The new version of the knowledge graph is stored in the graph database and associated with the old version to facilitate subsequent tracking and review. For example, through the version comparison function, you can clearly see the differences between the new version and the old version, including newly added entities and relationships, modified entities and relationships, and deleted entities and relationships.

[0042] In this embodiment, the graph information proofread by experts is accumulated and re-imported into the training set, the hybrid intelligent model is updated, and the above steps are repeated to continuously improve the performance of the hybrid intelligent model through human-computer collaboration, and absorb the expert experience knowledge into the knowledge graph.

[0043] Specifically, the graph information proofread by experts is accumulated and re-imported into the training set to update the hybrid intelligent model, including: The expert confirmation information accumulated during the proofreading process is stored in the knowledge base. The expert confirmation information includes entity names, attributes, relationship types, semantic annotations, etc. that have been proofread and confirmed by experts. The knowledge base adopts a structured database form to classify, store and manage the expert confirmation information for subsequent query and use.

[0044] Integrate and annotate the expert-confirmed information and convert it into a data format suitable for model training. For example, the entity names and attributes confirmed by experts are used as annotated data to train the entity recognition part of the large language model; the relationship types and semantic annotations confirmed by experts are used as annotated data to train the relationship extraction part of the model.

[0045] As feedback information accumulates, the entity recognition and relationship extraction capabilities of the hybrid intelligent model are gradually optimized. The expert confirmation information in the knowledge base is used as new training data to retrain the hybrid intelligent model. During the training process, semi-supervised learning, transfer learning and other technologies are used to make full use of the expert's labeled data and the model's existing knowledge to improve the performance and accuracy of the model. For example, through semi-supervised learning, a small amount of expert-labeled data can be used to guide model learning, while a large amount of unlabeled data can be used for self-learning to improve the model's ability to recognize medical entities and relationships.

[0046] The performance of the updated hybrid intelligent model is evaluated, including the accuracy of entity recognition, recall rate, accuracy of relationship extraction, F1 value and other indicators. According to the evaluation results, the parameters and structure of the model are further adjusted for iterative optimization. For example, if it is found that the model has deficiencies in certain types of medical entity recognition, the feature extraction part of the model can be adjusted in a targeted manner to increase the weight of related features; if it is found that the model has errors in the extraction of certain types of relationships, the relationship reasoning part of the model can be optimized to improve the reasoning algorithm. Embodiment 2

[0047] like Figure 2 As shown, the present invention provides a medical knowledge graph construction device based on hybrid intelligent technology, which is used to implement the medical knowledge graph construction method based on hybrid intelligent technology in the above-mentioned embodiment 1, specifically comprising: The hybrid intelligent model construction module is used to use the hybrid intelligent model to perform medical entity recognition and relationship extraction, as well as knowledge graph construction. The hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model. It recognizes key knowledge entities in medical knowledge materials and extracts the semantic relationships between them to achieve automatic construction of knowledge graphs. The expert interactive proofreading module is used to proofread and optimize the constructed knowledge graph through an expert interactive proofreading mechanism driven by a large language model. The large language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through a visual proofreading interface. At the same time, the proofreading process is recorded and versioned. The model update module is used to accumulate the graph information proofread by experts and re-import it into the training set, update the hybrid intelligent model, and repeat the above steps repeatedly to continuously improve the performance of the hybrid intelligent model through human-computer collaboration and absorb expert experience knowledge into the knowledge graph.

[0048] Furthermore, the hybrid intelligent model building module includes: Entity recognition and relationship extraction unit, which is used to extract key medical entities from text data and image data by combining large language models with multimodal models, and to design a mechanism based on the combination of time series models and text data to achieve entity extraction and alignment of key medical knowledge concepts; The knowledge graph construction unit is used to generate a structured knowledge graph after semantic classification and standardization of the extracted entities and relationships.

[0049] Furthermore, the expert interactive proofreading module includes: The proofreading suggestion generation unit is used by the large language model to provide preliminary proofreading suggestions for each entity and relationship based on semantic analysis and relational reasoning, and to mark low-confidence areas, while outputting the logical basis for each proofreading suggestion; The visual proofreading interface unit is used to provide experts with a graphical display of knowledge graph nodes and relationships, allowing them to quickly understand the recommendations generated by the model and the basis behind them; An interactive adjustment and feedback unit, which allows experts to directly adjust graph nodes and relationships and provide feedback on each adjustment operation; The proofreading record and version management unit is used to record every step and decision of the proofreading in detail, and after each proofreading is completed, a new version of the knowledge graph is generated, compared with the previous version, and the change points are marked.

[0050] Furthermore, the model updating module includes: The knowledge accumulation unit is used to store the graph information proofread by experts accumulated during the proofreading process; The model optimization unit is used to gradually optimize the entity recognition and relationship extraction capabilities of the hybrid intelligent model as feedback information accumulates, reduce errors in subsequent construction, and improve efficiency.

[0051] A medical knowledge graph construction device based on hybrid intelligent technology in this embodiment is used to implement the aforementioned medical knowledge graph construction method based on hybrid intelligent technology. Therefore, the specific implementation method of the medical knowledge graph construction device based on hybrid intelligent technology can be found in the embodiment part of the medical knowledge graph construction method based on hybrid intelligent technology in the previous text. In order to avoid redundancy, it will not be repeated here.

[0052] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0053] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0054] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0055] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A method for constructing a medical knowledge graph based on hybrid intelligent technology, characterized in that: include: Use a hybrid intelligent model to identify medical entities and extract relationships, as well as build a knowledge graph. The hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model to identify key knowledge entities in medical knowledge materials and extract the semantic relationships between them, thereby realizing the automatic construction of a knowledge graph. Through the expert interactive proofreading mechanism driven by the big language model, the constructed knowledge graph is proofread and optimized. The big language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through the visual proofreading interface. At the same time, the proofreading process is recorded and versioned. The graph information proofread by experts is accumulated and re-imported into the training set, the hybrid intelligent model is updated, and the above steps are repeated. The performance of the hybrid intelligent model is continuously improved through human-computer collaboration, and the expert experience knowledge is absorbed into the knowledge graph.

2. The method for constructing a medical knowledge graph based on hybrid intelligent technology according to claim 1, characterized in that: The method of using a hybrid intelligent model to perform medical entity recognition and relationship extraction includes: Combine large language models with multimodal models to extract key medical entities from text data and image data, including disease names, drug information, onset sites, and abnormal descriptions; A mechanism based on the combination of time series models and text data is designed to achieve the entity extraction and alignment of key medical knowledge concepts, combine the descriptive relationships between medical entities mined from text, and achieve the alignment of medical entities and the association of their related relationships.

3. The method for constructing a medical knowledge graph based on hybrid intelligent technology according to claim 1, characterized in that: The large language model generates preliminary proofreading suggestions, including: The large language model provides preliminary proofreading suggestions for each entity and relationship based on semantic analysis and relational reasoning, and marks low-confidence areas; The large language model outputs the logical basis for each proofreading suggestion, including the text source, reasoning path, and referenced knowledge base, providing decision-making support information for experts.

4. The method for constructing a medical knowledge graph based on hybrid intelligent technology according to claim 1, characterized in that: The experts make interactive adjustments and provide feedback through a visual proofreading interface, including: Experts can quickly understand the suggestions generated by the model and the basis behind them through a visual proofreading interface that displays nodes and relationships in a graphical knowledge graph; Experts directly adjust graph nodes and relationships, including adding, modifying or deleting entities and relationships, and provide feedback on each adjustment operation so that the hybrid intelligent model can absorb expert experience in subsequent training.

5. The method for constructing a medical knowledge graph based on hybrid intelligent technology according to claim 1, characterized in that: The record proofreading process and version management include: Keep detailed records of every step and decision of proofreading, including operation time, knowledge entities or relationships involved, and revisions made by experts and their basis; After each proofreading, a new version of the knowledge graph is generated and compared with the previous version, marking the change points for subsequent tracking and review.

6. The method for constructing a medical knowledge graph based on hybrid intelligent technology according to claim 1, characterized in that: The accumulation of expert-proofread graph information and re-importing it into the training set to update the hybrid intelligent model includes: The expert-proofread graph information accumulated during the proofreading process is stored in the knowledge base as important data for further training the hybrid intelligent model; As feedback information accumulates, the hybrid intelligent model gradually optimizes its entity recognition and relationship extraction capabilities, reducing errors in subsequent construction and improving efficiency.

7. A medical knowledge graph construction device based on hybrid intelligent technology, characterized in that: The device is used to implement the medical knowledge graph construction method based on hybrid intelligent technology according to any one of claims 1 to 6, specifically comprising: A hybrid intelligent model construction module is used to use a hybrid intelligent model to perform medical entity recognition and relationship extraction, as well as knowledge graph construction. The hybrid intelligent model includes a large language model, a multimodal model, a time series model, and a large image model to identify key knowledge entities in medical knowledge materials and extract the semantic relationships between them, thereby realizing the automatic construction of a knowledge graph. The expert interactive proofreading module is used to proofread and optimize the constructed knowledge graph through an expert interactive proofreading mechanism driven by a large language model. The large language model generates preliminary proofreading suggestions, and experts make interactive adjustments and feedback through a visual proofreading interface. At the same time, the proofreading process is recorded and versioned. The model updating module is used to accumulate the graph information proofread by experts and re-import it into the training set, update the hybrid intelligent model, repeat the above steps repeatedly, continuously improve the performance of the hybrid intelligent model through human-computer collaboration, and absorb expert experience knowledge into the knowledge graph.

8. The medical knowledge graph construction device based on hybrid intelligent technology according to claim 1, characterized in that: The hybrid intelligent model building module includes: Entity recognition and relationship extraction unit, which is used to extract key medical entities from text data and image data by combining large language models with multimodal models, and to design a mechanism based on the combination of time series models and text data to achieve entity extraction and alignment of key medical knowledge concepts; The knowledge graph construction unit is used to generate a structured knowledge graph after semantic classification and standardization of the extracted entities and relationships.

9. The medical knowledge graph construction device based on hybrid intelligent technology according to claim 1, characterized in that: The expert interactive proofreading module comprises: The proofreading suggestion generation unit is used by the large language model to provide preliminary proofreading suggestions for each entity and relationship based on semantic analysis and relational reasoning, and to mark low-confidence areas, while outputting the logical basis for each proofreading suggestion; The visual proofreading interface unit is used to provide experts with a graphical display of knowledge graph nodes and relationships, allowing them to quickly understand the recommendations generated by the model and the basis behind them; An interactive adjustment and feedback unit, which allows experts to directly adjust graph nodes and relationships and provide feedback on each adjustment operation; The proofreading record and version management unit is used to record every step and decision of the proofreading in detail, and after each proofreading is completed, a new version of the knowledge graph is generated, compared with the previous version, and the change points are marked.

10. The medical knowledge graph construction device based on hybrid intelligent technology according to claim 1, characterized in that: The model updating module comprises: The knowledge accumulation unit is used to store the graph information proofread by experts accumulated during the proofreading process; The model optimization unit is used to gradually optimize the entity recognition and relationship extraction capabilities of the hybrid intelligent model as feedback information accumulates, reduce errors in subsequent construction, and improve efficiency.

Images