Medical record digital generation management method, system and product

Abstract

The invention discloses a medical record digital generation management method, a medical record digital generation management system and a medical record digital generation management product, and solves the problems of limited automation degree, insufficient data processing capability, limited verification range and lack of full-process management in the prior art. According to the application, the medical record records in the hospital information system and the electronic medical record system are automatically obtained by using the RPA technology, and the basic information in the medical record is extracted; automatically creating the extracted basic information as to-be-scanned medical record processing data, and filling a predefined field of a medical record management system with the to-be-scanned medical record processing data; receiving a digital image of a paper medical record material, extracting recognizable text information from the digital image by using an OCR technology, generating a recognition result, extracting an image name through LLM and NLP technologies, and performing cataloguing by using the image name; and carrying out first-time comprehensive verification on the eye-classified medical record by utilizing the AIAgent according to legal specifications and pre-processing configuration, and filing and storing digital data of the medical record after the two-time verification is passed.

Description

Technical Field

[0001] The present application relates to the technical field of medical record management, and in particular to a method for managing the digital generation of medical records. Background Art

[0002] In the healthcare industry, medical record management is a critical component in ensuring healthcare quality and patient safety. Medical records provide detailed records of a patient's diagnosis and treatment, including medical history, diagnosis, treatment measures, and test results. This information is crucial for medical decision-making, patient treatment, medical research, and health insurance reimbursement. With the surge in medical data volumes, traditional medical record management approaches face significant challenges.

[0003] In recent years, with the development of information technology, new technologies have emerged in the field of medical record management, such as optical character recognition (OCR), natural language processing (NLP), and robotic process automation (RPA). The inventors have discovered that while these technologies have improved the efficiency and accuracy of medical record management to a certain extent, they still have some problems and limitations.

[0004] For example, Patent Document 1CN118430730A proposes a medical record management method, system, computer device, and storage medium. This method primarily uses OCR technology, natural language processing (NLP), and pre-set medical record quality control rules to scan medical records, extract text information, perform structured processing, and perform quality control analysis. Although this method improves the efficiency and accuracy of medical record quality control to a certain extent, the following issues remain:

[0005] Limited automation: Despite the introduction of OCR and NLP technologies, the system relies on manual scanning of paper medical records and image generation during the medical record collection stage, and manual review and modification are still required after quality control analysis. The degree of automation is not high enough to completely eliminate the errors and inefficiencies caused by manual intervention.

[0006] Insufficient data processing capabilities: This method primarily focuses on extracting and structuring textual information from medical records. It does not adequately optimize medical record images, which may affect the accuracy of subsequent text recognition and data extraction. Furthermore, the method lacks the ability to process large amounts of medical record data, potentially failing to meet the needs of large medical institutions.

[0007] Patent document 2CN115620855A discloses a medical record integrity verification method and system based on artificial intelligence and machine learning. This method primarily forms a medical record rule library by organizing medical record rules, and then uses artificial intelligence to analyze patient admissions and conditions, automatically verifying medical record integrity based on these rules. While this method has made some contributions to medical record integrity verification, it also has the following issues:

[0008] The scope of verification is limited: it mainly focuses on the verification of medical record integrity, and does not involve other aspects such as the reliability, accuracy and compliance of medical records, and cannot fully guarantee the quality of medical record data.

[0009] Insufficient data processing capabilities: During the medical record processing process, medical record images are not fully optimized and processed, and text information is not extracted. This may affect the accuracy and availability of medical record data.

[0010] Furthermore, the inventors recognized that both Patent Document 1 and Patent Document 2 lack full-process management. Patent Document 1 primarily focuses on the extraction and structural processing of medical record text information, but fails to address the full-process management of medical record collection and processing. This makes it impossible to achieve a complete conversion and efficient management of medical records from paper to digital. Patent Document 2 focuses solely on verifying medical record integrity, but fails to address the full-process management of medical record collection and processing. This also fails to achieve a complete conversion and efficient management of medical records from paper to digital. Summary of the Invention

[0011] This application provides a medical record digital generation management method, system and product, aiming to solve the problems of limited automation, insufficient data processing capabilities, limited verification scope, and lack of full-process management in existing technologies.

[0012] In the first aspect, a method for managing digital generation of medical records is provided, comprising:

[0013] Medical record collection: Utilize RPA technology to automatically obtain medical records from hospital information systems and electronic medical record systems, and extract basic information from the medical records. This extracted basic information is automatically created as medical record processing data to be scanned and populated into predefined fields in the medical record management system.

[0014] Medical record processing: Receive digital images of paper medical records, extract recognizable text information from the digital images using OCR technology, generate recognition results, extract image names using LLM and NLP technologies, and use image names for cataloging; simultaneously, use NLP technology to extract key entity information, and store the medical record images in the corresponding directory based on the extracted key entity information;

[0015] Medical record verification: After the medical records are classified, the AIAgent is used to conduct the first comprehensive verification of the classified medical records according to legal regulations and pre-processing configurations, marking unqualified items, feeding back the unqualified items and their problems to the medical record processing step, and reminding relevant personnel to deal with them; the medical records that have completed the first comprehensive verification are submitted to relevant staff for a second verification; after receiving the second verification results, if any problems are found, the medical records will be returned to the medical record processing stage;

[0016] Medical record storage and management: After passing two verifications, the digitized medical record data is archived and stored; the recognition results extracted using OCR technology, as well as the basic information and key entity information, are integrated through AIAgent to build a medical record retrieval library; this retrieval library allows doctors and patients to access and retrieve the required medical record information.

[0017] In the above solution, optionally, when the user searches for medical record information through the search library, the finished digital medical record archive can be directly located and used through the search results.

[0018] In the above solution, optionally, in the medical record processing step, different catalog compilation standards for the same material in different medical institutions are flexibly configured;

[0019] In the medical record storage and management step, the medical record retrieval library is initialized and customized according to the different catalog compilation standards of different medical institutions for the same material.

[0020] In the above solution, the AI ​​Agent can be optionally maintained and updated according to the needs of the medical institution and changes in laws and regulations.

[0021] In the above solution, optionally, after receiving the digital image of the paper medical record material and before extracting the recognizable text information from the digital image using OCR technology, the method further includes:

[0022] The digital image information is preprocessed using CV technology, including image enhancement, image noise reduction, text and seal deepening, stain removal, black edge removal, binding hole removal, tilt correction and rotation correction.

[0023] In the above solution, optionally, the search library supports a flexible combination of multiple search methods, including keyword search, patient information search and medical record type search.

[0024] In the above solution, optionally, the medical record collection stage can seamlessly connect to various types of hospital information systems and electronic medical record systems.

[0025] Secondly, a medical record digital generation and management system is provided, including:

[0026] The medical record collection module uses RPA technology to automatically obtain medical records from hospital information systems and electronic medical record systems and extract basic information from them. The extracted basic information is automatically created as medical record processing data to be scanned and populated into predefined fields in the medical record management system.

[0027] The medical record processing module is used to receive digital images of paper medical record materials, extract recognizable text information from the digital images using OCR technology, generate recognition results, extract image names using LLM and NLP technologies, and catalog the image names; at the same time, use NLP technology to extract key entity information and store the medical record images in the corresponding directory based on the extracted key entity information;

[0028] The medical record verification module is used to perform a first comprehensive verification of the classified medical records using AIAgent according to legal regulations and pre-processing configuration after the medical records are classified, mark unqualified items, and feedback the unqualified items and their problems to the medical record processing step, and remind relevant personnel to handle them; submit the medical records that have completed the first comprehensive verification to relevant staff for a second verification; receive the second verification results, and if any problems are found, return the medical records to the medical record processing stage;

[0029] The medical record storage and management module is used to archive and store the digitized medical record data after two verifications. The recognition results extracted by OCR technology, as well as the basic information and key entity information, are integrated through AIAgent to build a medical record retrieval library. The retrieval library allows doctors and patients to access and retrieve the required medical record information.

[0030] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0031] In a fourth aspect, a computer program product is provided, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0032] Compared with the prior art, this application has at least the following beneficial effects:

[0033] Based on further analysis and research on existing technical problems, this application recognizes that the existing technology has problems such as limited automation, insufficient data processing capabilities, limited verification scope, and lack of full-process management. This application introduces RPA (robotic process automation) technology to automatically obtain medical records in hospital information systems and electronic medical record systems, and accurately extract basic information therein. This process does not require manual intervention, greatly improving the efficiency and accuracy of medical record collection, reducing errors caused by manual operations, and effectively solving the problem of limited automation in the existing technology. In the medical record processing stage, this application uses OCR technology to extract text information from digital images of paper medical record materials, which can quickly and accurately identify the text content in the image. At the same time, with the help of LLM and NLP technology, not only can the image name be extracted for cataloging, but also the key entity information can be accurately extracted, and the medical record image can be stored in the corresponding directory accordingly. The application of this series of technologies has greatly improved the data processing capabilities and efficiency, enabling medical record data to be better utilized, thereby solving the problem of insufficient data processing capabilities in the existing technology. Through AI Agent (artificial intelligence agent), the first comprehensive verification of the classified medical records is carried out in accordance with legal regulations and pre-processing configurations. It can automatically mark unqualified items and feedback the unqualified items and their problems to the medical record processing steps, reminding relevant personnel to deal with them. This not only improves the efficiency of verification, but also expands the scope of verification, and can deeply check whether the content of the medical records meets the various requirements, making up for the defect of limited verification scope in the existing technology. In addition, this application further ensures the quality of medical records by conducting a secondary verification of medical records. This application closely links the collection, processing, verification, storage and management of medical records to form a complete full-process management system. An effective connection and feedback mechanism is set up between each link, so that the various links of medical record management can work together, improving the overall efficiency and quality of medical record management. Finally, the digitized medical record data is archived and stored, and the medical record retrieval library is built through AI Agent integration, realizing efficient storage and convenient retrieval of medical record information, providing better services for doctors and patients, thereby solving the problem of lack of full-process management in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A flowchart of a method for digitally generating and managing medical records provided in the first embodiment of this application.

[0035] Figure 2 A flowchart of a medical record digital generation and management method provided in the second embodiment of the present application. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0037] In the description of the present application, unless otherwise specified, expressions such as “include”, “comprising”, “having”, etc. also mean “not limited to” (certain units, components, materials, steps, etc.).

[0038] In one embodiment, reference Figure 1 and Figure 2 , provides a medical record digital generation and management method, which specifically includes the following steps S1-S4.

[0039] Step S1, medical record collection: Utilize RPA technology to automatically obtain medical records from the hospital information system and electronic medical record system, and extract basic information from the medical records; automatically create the extracted basic information into medical record processing data to be scanned, and fill it into predefined fields in the medical record management system;

[0040] In one embodiment, the medical record collection stage can seamlessly connect to various types of hospital information systems and electronic medical record systems.

[0041] In this embodiment, using RPA (Robotic Process Automation) technology, the system can automatically access archivable medical records from hospital information systems (HIS) and electronic medical records (EMR). By seamlessly integrating with these systems, RPA can efficiently extract basic information from medical records, including important data such as medical record number, hospitalization number, patient name, admission date, and discharge date. After extraction, this basic information is automatically created as scanned medical record processing data and populated into the corresponding fields, ensuring that the basic information of each medical record is complete and accurate. This prepares the system for subsequent scanning and processing, optimizes the medical record processing process, improves work efficiency, reduces the need for manual intervention, and provides medical personnel with a more efficient medical record management experience.

[0042] The medical record collection stage supports seamless connection with various types of hospital information systems (HIS) and electronic medical record systems (EMR), which can quickly obtain medical record information, meet the needs of different medical institutions, and achieve efficient response.

[0043] The medical record verification module implements an intelligent verification mechanism through the built-in AI Agent (intelligent agent), and its pre-processing configuration is highly flexible and scalable. This module can be individually adjusted and customized according to the medical record management standards and business needs of different medical institutions, thereby meeting diverse compliance requirements. This series of functional designs ensures the system's adaptability and efficiency in various complex scenarios and requirements.

[0044] Step S2, medical record processing: receiving a digital image of a paper medical record material, extracting recognizable text information from the digital image using OCR technology, generating a recognition result, extracting the image name using LLM and NLP technology, and cataloging the image name; simultaneously, extracting key entity information using NLP technology, and storing the medical record image in the corresponding directory based on the extracted key entity information;

[0045] In one embodiment, after receiving the digital image of the paper medical record material and before extracting recognizable text information from the digital image using OCR technology, the method further includes:

[0046] The digital image information is preprocessed using CV technology, including image enhancement, image noise reduction, text and seal deepening, stain removal, black edge removal, binding hole removal, tilt correction and rotation correction.

[0047] In this embodiment, the processing personnel perform the scanning of medical record images based on the medical record processing data to be scanned. Using professional equipment such as scanners or high-definition cameras, the processing personnel convert the medical record paper materials into digital image information and enter it into the system. To ensure the clarity and readability of the image, the system uses CV (computer vision) technology to perform a series of optimization processes on the scanned images. These processes include pre-processing technologies such as image enhancement, image noise reduction, text and seal deepening, decontamination, black edge removal, binding hole tilt correction, and rotation correction. Through the application of these technologies, the system can significantly improve the image quality and ensure that the subsequent text recognition and data extraction stages have higher accuracy and reliability.

[0048] After the image preprocessing is completed, the system will enter the cataloging and categorization process, the ultimate goal of which is to name the image and store it in the corresponding directory to complete the cataloging and categorization operations. During this process, the system first uses OCR (optical character recognition) technology to extract all recognizable text information from the image and generate recognition results. Then, combined with LLM (large language model) and NLP (natural language processing) technology, the system can extract the name of the image from the recognition results to ensure that each image can be named accurately and clearly. Subsequently, based on the extracted image name, the system will perform the cataloging operation.

[0049] During the cataloging process, NLP technology is also used to extract key entity information, such as department information, medical record type, and patient information. Extracting these key entities not only provides a crucial basis for image cataloging but also ensures systematic and categorized management of medical records. Based on the extracted key entity information, the system catalogs the medical records accordingly, ensuring that each record's storage location in the electronic system accurately reflects its content and attributes, thereby improving the efficiency and accuracy of medical record management.

[0050] In one embodiment, in the medical record processing step, different catalog compilation specifications for the same material are flexibly configured according to different medical institutions; in the medical record storage and management step, the medical record retrieval library is initialized and customized according to different catalog compilation specifications for the same material by different medical institutions.

[0051] During the medical record processing phase, computer vision (CV) technology provides flexible configuration options for image preprocessing, enabling tailored processing based on the quality of paper medical records and supporting the combined use of multiple preprocessing functions. Cataloging and grouping functions can also be flexibly configured to accommodate the varying cataloging standards used by different medical institutions for the same document, as well as initializing and customizing the knowledge base to meet the specific requirements of each hospital.

[0052] By introducing RPA (Robotic Process Automation) technology, the entire medical record collection process is now fully automated. The system automates the entire process, from collecting and organizing medical record information to scanning and processing paper materials, significantly reducing the complexity and error rate of manual operations. Combined with OCR (Optical Character Recognition) technology, the system efficiently and accurately extracts text information from paper medical records, enabling the digital conversion of data. Furthermore, leveraging LLM (Large Language Model) and NLP (Natural Language Processing) technologies, the system intelligently catalogs and groups medical record images, significantly improving classification accuracy and automation.

[0053] This comprehensive technical solution not only significantly reduces manual intervention and lowers labor costs, but also greatly improves the efficiency and accuracy of data processing, enabling hospitals to respond to patient needs more quickly, optimize medical service processes, promote the modernization of medical informatization, and provide patients with a safer, more efficient, and intelligent diagnosis and treatment experience.

[0054] Step S3, medical record verification: After the medical records are classified, the AIAgent is used to conduct a first comprehensive verification of the classified medical records according to legal regulations and pre-processing configurations, marking unqualified items, feeding back the unqualified items and their problems to the medical record processing step, and reminding relevant personnel to handle them; the medical records that have completed the first comprehensive verification are submitted to relevant staff for a second verification; after receiving the second verification results, if any problems are found, the medical records are returned to the medical record processing stage;

[0055] In this embodiment, an AI Agent (intelligent agent) first conducts a comprehensive verification of the medical record's reliability, completeness, and accuracy based on legal regulations and pre-processing configurations. This intelligent agent system uses an automated process to flag unqualified items and relay these issues back to the medical record processing step, requiring the processor to correct them. Once the processor completes the necessary corrections, the system resubmits the records for verification.

[0056] After verification is complete, dedicated verification personnel will conduct a second review of each medical record to ensure the authenticity and reliability of all information. This double verification process not only improves the quality of medical records and ensures compliance with relevant laws and regulations, but also reduces potential errors and provides a safer and more reliable guarantee for hospital medical record management. This rigorous verification mechanism helps maintain the integrity and availability of medical record data, ultimately improving the quality of medical services.

[0057] In one embodiment, the AI ​​Agent is maintained and updated according to the needs of the medical institution and changes in laws and regulations.

[0058] The intelligent verification mechanism built on AIAgent (intelligent agent) relies on relevant laws and regulations and pre-processing configuration to achieve multi-dimensional and comprehensive automatic review of medical record information. This mechanism strictly controls the reliability, integrity and accuracy of data to ensure high quality assurance and compliance of medical record data. At the same time, the pre-processing configuration is highly flexible and scalable, and can be personalized and customized according to the medical record management standards and business needs of different medical institutions, thereby meeting diverse regulatory requirements and improving the system's applicability and practical value.

[0059] The medical record verification phase utilizes an AI Agent (intelligent agent) to implement an intelligent verification mechanism, providing highly flexible and scalable pre-processing configuration. This module can be individually adjusted and customized based on the medical record management standards and business needs of different medical institutions, thereby meeting diverse compliance requirements. This comprehensive set of functional designs ensures the system's adaptability and efficiency in various complex scenarios and requirements.

[0060] Step S4, medical record storage and management: After passing two verifications, the digitized medical record data is archived and stored; the recognition results extracted using OCR technology, as well as the basic information and key entity information, are integrated through AI Agent to build a medical record retrieval library; this retrieval library allows doctors and patients to access and retrieve the required medical record information.

[0061] In this embodiment, the verified digitized medical record data is archived and stored. During this process, the system uses OCR (Optical Character Recognition) technology to extract recognition results and related medical record metadata, which are then integrated and constructed by the AI ​​Agent (Intelligent Agent) to form an efficient search library. This search library not only contains key information about the medical record images, but also covers rich metadata such as basic patient information, disease type, treatment department, and medical records.

[0062] Through this search library, doctors and patients can quickly and conveniently access and retrieve the medical record information they need during subsequent medical treatment. This efficient search mechanism greatly improves information accessibility, helping doctors make diagnostic and treatment decisions more quickly while also providing patients with a better service experience. Furthermore, a comprehensive medical record management and storage system ensures data security and compliance, laying a solid foundation for the hospital's digital transformation and information development.

[0063] In one embodiment, when a user searches for medical record information through a search library, the finished digitized medical record archive can be directly located and used through the search results.

[0064] In one embodiment, the search library supports a flexible combination of multiple search methods, including keyword search, patient information search, and medical record type search.

[0065] Based on advanced technologies such as AI Agent (intelligent agent), OCR (optical character recognition), and NLP (natural language processing), a highly automated medical record digital management and retrieval database has been constructed. This system not only offers the advantages of small storage volume and fast retrieval speed, but also achieves high accuracy in search results. It supports flexible combinations of multiple search methods, facilitating convenient and efficient medical record information query and retrieval based on the needs of medical staff and patients, significantly improving the utilization efficiency of medical data and the service experience.

[0066] This application uses artificial intelligence (AI) technology, robotic process automation (RPA) and intelligent agents (AIAgent) to realize the digital generation and management method of medical records from medical record scanning and processing to digital utilization. This application is based on AI (artificial intelligence technology) and uses LLM (large language model), NLP (natural language processing), OCR (optical character recognition), CV (computer vision) and other technologies to realize the digitization of medical records such as scanning, image processing, and automatic cataloging of medical record materials. Based on RPA (robotic process automation) technology, medical record basic metadata information collection, entry, verification and other medical record information digitization is realized. Based on AIAgent (intelligent agent) technology, medical record verification, retrieval and many other functions are realized, making full use of the data resources after the medical record is digitized.

[0067] The following is a complete description of the process of a medical record digital generation and management method of this application:

[0068] First, RPA technology is used to collect archivable medical record data from external systems such as HIS / EMR systems. This basic information is then entered into the medical record to be processed. Processing personnel then scan the paper materials and assign the electronic files to the corresponding medical records. After scanning, the system uses computer vision (CV) technology to optimize and pre-process the images. The optimized images are then entered into the medical record to be processed and output as scanned, completed medical records.

[0069] Next, combining technologies such as LLM (Large Language Model), NLP (Natural Language Processing), and OCR (Optical Character Recognition), and in accordance with catalog compilation specifications, the medical records are intelligently cataloged and classified to generate medical records to be verified. Subsequently, the AIAgent (intelligent agent) automatically verifies the medical records according to the verification specifications. If the verification fails, the defect is marked and the medical record is returned to the processing link for reprocessing; if it passes the verification, the system submits the medical record to the manual verification personnel for a second review. If there are still problems, it will be returned to the medical record to be processed for correction; if it passes smoothly, the final digital medical record file will be generated.

[0070] Subsequently, AIAgent combined NLP and OCR technologies to establish a medical record retrieval database, supporting medical record managers, patients and other users to conveniently query medical record information through the retrieval database, and can directly locate and use the finished digital medical record archives through the retrieval results, improving the efficiency of obtaining medical record information and the user experience.

[0071] This application has at least the following advantages:

[0072] 1. Significantly improve work efficiency: Automated medical record collection, medical record image optimization processing, medical record cataloging and classification, medical record quality verification, and the construction of a medical record retrieval database have significantly reduced labor costs and greatly improved the work efficiency of medical record digital management.

[0073] 2. Reduce human errors: Through technologies such as LLM (large language model), NLP (natural language processing), OCR (optical character recognition), CV (computer vision), RPA (robotic process automation), and AIAgent (intelligent agent), the system effectively reduces errors that may occur during manual collection, entry, and analysis, thereby significantly improving the accuracy and consistency of data.

[0074] 3. Improve information utilization: The digital medical record retrieval database, built using intelligent agents (AI Agents), natural language processing (NLP), and optical character recognition (OCR), ensures the comprehensiveness of search results and the accuracy of key information, while also improving retrieval efficiency. This not only increases information utilization but also significantly enhances the value of digital medical record information.

[0075] 4. Optimize resource allocation: Efficient automated processing reduces dependence on human resources, allowing medical record managers to devote more energy to other important tasks, thereby improving the efficiency of overall resource allocation.

[0076] 5. Enhance the flexibility and adaptability of the system: The medical record digitization process supports multiple customizations and flexible configurations, making the system highly adaptable and able to meet the medical record digitization standard requirements of different medical institutions, thereby achieving rapid response and rapid deployment.

[0077] In summary, the proposed method for digital medical record generation and management based on AI, RPA, and AIAgent significantly outperforms existing digital medical record management solutions in terms of automation, intelligent verification mechanisms, rapid retrieval and access, and high scalability and flexibility. This system not only possesses high technological advancement and application value, but also demonstrates broad application prospects and promotional potential.

[0078] In one embodiment, a medical record digital generation and management system includes:

[0079] The medical record collection module uses RPA technology to automatically obtain medical records from hospital information systems and electronic medical record systems and extract basic information from them. The extracted basic information is automatically created as medical record processing data to be scanned and populated into predefined fields in the medical record management system.

[0080] The medical record processing module is used to receive digital images of paper medical record materials, extract recognizable text information from the digital images using OCR technology, generate recognition results, extract image names using LLM and NLP technologies, and catalog the image names; at the same time, use NLP technology to extract key entity information and store the medical record images in the corresponding directory based on the extracted key entity information;

[0081] The medical record verification module is used to perform a first comprehensive verification of the classified medical records using AIAgent according to legal regulations and pre-processing configuration after the medical records are classified, mark unqualified items, and feedback the unqualified items and their problems to the medical record processing step, and remind relevant personnel to handle them; submit the medical records that have completed the first comprehensive verification to relevant staff for a second verification; receive the second verification results, and if any problems are found, return the medical records to the medical record processing stage;

[0082] The medical record storage and management module is used to archive and store the digitized medical record data after two verifications. The recognition results extracted by OCR technology, as well as the basic information and key entity information, are integrated through AIAgent to build a medical record retrieval library. The retrieval library allows doctors and patients to access and retrieve the required medical record information.

[0083] The specific implementation content of each module can be found in the above definition of a medical record digital generation and management method, which will not be repeated here.

[0084] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method of the above embodiment.

[0085] In one embodiment, a computer program product is also provided, including a computer program / instruction, which implements the steps of the above-mentioned embodiment method when executed by a processor.

[0086] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method for digital generation and management of medical records, characterized in that: include: Medical record collection: Utilize RPA technology to automatically obtain medical records from hospital information systems and electronic medical record systems, and extract basic information from the medical records. This extracted basic information is automatically created as medical record processing data to be scanned and populated into predefined fields in the medical record management system. Medical record processing: Receive digital images of paper medical records, extract recognizable text information from the digital images using OCR technology, generate recognition results, extract image names using LLM and NLP technologies, and use image names for cataloging; simultaneously, use NLP technology to extract key entity information, and store the medical record images in the corresponding directory based on the extracted key entity information; Medical record verification: After the medical records are classified, AIAgent is used to conduct the first comprehensive verification of the classified medical records according to legal regulations and pre-processing configurations, marking unqualified items, and feeding back the unqualified items and their problems to the medical record processing step, and reminding relevant personnel to handle them; Submit the medical records that have completed the first comprehensive verification to the relevant staff for a second verification; receive the second verification results, and if any problems are found, return the medical records to the medical record processing stage; Medical record storage and management: After passing two verifications, the digital medical record data is archived and stored; the recognition results extracted by OCR technology, as well as the basic information and key entity information, are integrated through AIAgent to build a medical record retrieval library; The search library allows doctors and patients to access and retrieve the medical record information they need.

2. The medical record digital generation and management method according to claim 1, characterized in that: When users search for medical record information through the search library, they can directly locate and use the finished digital medical record archive through the search results.

3. The medical record digital generation and management method according to claim 1, characterized in that: In the medical record processing step, different catalog compilation standards for the same material in different medical institutions are flexibly configured; In the medical record storage and management step, the medical record retrieval library is initialized and customized according to the different catalog compilation standards of different medical institutions for the same material.

4. The medical record digital generation and management method according to claim 1, characterized in that: Maintain and update AIAgent according to the needs of medical institutions and changes in laws and regulations.

5. The medical record digital generation and management method according to claim 1, characterized in that: After receiving the digital image of the paper medical record material and before extracting recognizable text information from the digital image using OCR technology, the method further includes: The digital image information is preprocessed using CV technology, including image enhancement, image noise reduction, text and seal deepening, stain removal, black edge removal, binding hole removal, tilt correction and rotation correction.

6. The medical record digital generation and management method according to claim 1, characterized in that: The search library supports flexible combinations of multiple search methods, including keyword search, patient information search, and medical record type search.

7. The medical record digital generation and management method according to claim 1, characterized in that: During the medical record collection stage, various types of hospital information systems and electronic medical record systems can be seamlessly connected.

8. A medical record digital generation and management system, characterized by: include: The medical record collection module uses RPA technology to automatically obtain medical records from hospital information systems and electronic medical record systems and extract basic information from them. The extracted basic information is automatically created as medical record processing data to be scanned and populated into predefined fields in the medical record management system. The medical record processing module is used to receive digital images of paper medical record materials, extract recognizable text information from the digital images using OCR technology, generate recognition results, extract image names using LLM and NLP technologies, and catalog the image names; at the same time, use NLP technology to extract key entity information and store the medical record images in the corresponding directory based on the extracted key entity information; The medical record verification module is used to perform the first comprehensive verification of the medical records after they are classified using AIAgent according to legal regulations and pre-processing configurations, mark unqualified items, and feed back the unqualified items and their problems to the medical record processing step, and remind relevant personnel to handle them; Submit the medical records that have completed the first comprehensive verification to the relevant staff for a second verification; receive the second verification results, and if any problems are found, return the medical records to the medical record processing stage; The medical record storage and management module is used to archive and store the digitized medical record data after two verifications. The recognition results extracted by OCR technology, as well as the basic information and key entity information, are integrated through AIAgent to build a medical record retrieval library. The search library allows doctors and patients to access and retrieve the medical record information they need.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to claim 1.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to claim 1 are implemented.

Images