Medical settlement sheet information verification processing method and system
Through multi-dimensional verification of image acquisition and deep learning intelligent models, the problems of low efficiency and poor accuracy of traditional medical settlement statements are solved, automated processing and data analysis are realized, operating costs are significantly reduced, and medical service quality is improved.
Patent Information
- Application Number
- CN202411095841.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-22
AI Technical Summary
Traditional medical settlement statement processing relies on a large number of manual operations, resulting in information entry errors and omissions, low efficiency and high cost, making it difficult to adapt to the needs of modern medical big data analysis.
Image acquisition equipment combined with OCR technology to convert paper settlement statements into electronic data, and multi-dimensional verification is performed based on deep learning intelligent model, including project matching, amount verification and patient identity verification, and through error analysis and correction suggestions mechanisms, it can achieve Automatic processing and structured finishing, and periodically back up to remote servers.
It significantly improves data accuracy and processing efficiency, reduces disputes, enhances the credibility of the settlement process, provides strong data analysis capabilities, reduces operational costs, and improves patient satisfaction and medical service quality.
Smart Images

Figure CN119007950A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of medical information technology, and specifically relates to a method and system for verifying and processing medical settlement form information. Background Art
[0002] In the context of the rapid advancement of medical informatization, medical settlement is an important part of the medical service process, and its efficiency and accuracy directly affect hospital management and patient experience. Traditional medical settlement form processing relies on a large number of manual operations, which is not only time-consuming and labor-intensive, but also prone to information entry errors, verification omissions and other problems, resulting in settlement delays and frequent cost disputes. In addition, paper settlement forms are inconvenient to store and manage, and information retrieval is difficult, making it difficult to adapt to the needs of modern medical big data analysis. Therefore, in the face of increasingly complex medical settlement processes and massive data processing challenges, there is an urgent need for a technical solution that can automatically, efficiently and accurately process medical settlement form information.
[0003] How to design and implement a method that can automatically collect, verify and manage medical settlement information, ensure the accuracy of data entry, the efficiency of processing and the security of storage, and provide in-depth analysis and visualization of data to support the refined management and decision-making optimization of medical institutions is an issue that needs to be urgently addressed. Summary of the invention
[0004] The purpose of the present invention is to provide a method and system for verifying and processing medical settlement form information, which significantly reduces the operating costs of medical institutions, improves patient satisfaction, and promotes the overall improvement of medical service quality. In order to solve the problems in the prior art mentioned in the above background technology.
[0005] To achieve the above object, the present invention provides a method for verifying and processing medical settlement form information, comprising the following steps:
[0006] The image information of the paper medical settlement form is obtained through the image acquisition device, and the text information in the image is converted into editable electronic data using the OCR technology, and the electronic data is converted into a medical settlement form file;
[0007] Based on the intelligent model of deep learning, it can identify and supplement the missing information in the medical settlement documents, and verify them through multiple dimensions such as item matching, amount verification and patient identity verification;
[0008] For settlement orders that fail verification, analyze the error type and provide error reports and correction suggestions, and guide the backend manual review and intervention for correction;
[0009] Structuralize and organize the verified settlement form information to form standardized data, classify and summarize the standardized data according to different dimensions of medical projects, departments, and time, and use charts and graphs to show the distribution characteristics and time series trends of medical settlement data;
[0010] Standardized data is backed up regularly to remote servers or cloud storage, and support query, export and printing of medical settlement information.
[0011] The use of the image acquisition device and the information processing process include: preprocessing the image obtained by scanning with a scanner to remove noise and shadows in the image, using OCR software to convert the preprocessed high-definition image into electronic text, and matching the data recognized by OCR with a preset medical settlement form template.
[0012] The verification rules for project matching include: establishing a medical project code database; using OCR technology and an intelligent information recognition module to extract all project codes and related information from the medical settlement form, and then standardizing the project codes: standardizing the extracted project codes; comparing the standardized project codes with the medical project coding system database in real time, and then obtaining the verification results.
[0013] During the real-time comparison process, determine whether each project code on the settlement form has a corresponding record in the database; if there is a match, then confirm whether the project name, executing department, and cost information are consistent; if not, it will be marked as a non-compliant project.
[0014] The verification rules for the amount verification include: reading the total amount listed on the medical settlement form through the intelligent verification module; parsing the settlement form item by item, adding up each medical service item and its corresponding fee; and comparing the total of the accumulated item fees with the total amount on the settlement form.
[0015] The patient identity verification rules include: establishing a data interface with the hospital information system, collecting patient information during the medical settlement form information entry stage, and transmitting the patient information on the settlement form to the hospital information system through the data interface in real time. After the HIS system receives the information, it immediately compares it with the patient registration information.
[0016] The analyzing error types and providing error reports and correction suggestions include: checking the medical settlement form information according to a plurality of preset verification rules in the intelligent verification module, and when an error is detected, classifying the error under a preset error type label, and providing error reports and correction suggestions for medical settlement forms that fail the verification.
[0017] The structured organization of the verified settlement form information includes: converting the image or non-text format of the medical settlement form into processable text data, then extracting key information, structuring the extracted key information according to a preset template, and outputting the structured data in a unified format.
[0018] The standardized data is backed up regularly to a remote server or cloud storage, including: structuring the verified medical settlement form information to form a standardized data format, and performing advanced encryption processing on the standardized data; creating storage nodes at multiple remote servers or cloud storage service providers, each node storing a copy of the standardized data; dividing the standardized data set into multiple data blocks, and saving multiple copies on different storage nodes.
[0019] On the other hand, the present invention provides a method for verifying and processing medical settlement form information, comprising:
[0020] The image acquisition and OCR conversion module is used to obtain the image information of the paper medical settlement form through the image acquisition device, and use the OCR technology to convert the text information in the image into editable electronic data, and convert the electronic data into a medical settlement form file;
[0021] The information completion and verification module is used to identify and supplement the missing information in the medical settlement form based on the intelligent model of deep learning, and verify it through multiple dimensions such as item matching, amount verification, and patient identity verification;
[0022] The error analysis and correction suggestion module is used to analyze the error type and provide error reports and correction suggestions for settlement orders that fail verification, and guide the back-end manual review and intervention correction;
[0023] The data structuring and analysis display module is used to structure and organize the verified settlement form information to form standardized data, classify and summarize the standardized data according to different dimensions of medical projects, departments, and time, and use charts and graphs to display the distribution characteristics and time series trends of medical settlement data;
[0024] The data backup and management module is used to regularly back up standardized data to a remote server or cloud storage, and supports the query, export and printing of medical settlement form information.
[0025] Technical effects and advantages of the present invention: Compared with the prior art, the present invention provides a method and system for verifying and processing medical settlement form information, which has the following advantages:
[0026] The present invention realizes the rapid conversion from paper documents to electronic data, greatly shortening the information processing cycle. Through deep learning models and multiple verification rules, it greatly improves data accuracy, reduces disputes caused by errors, and enhances the credibility of the settlement process. The introduced error analysis and correction suggestion mechanism provides precise guidance for manual review and accelerates the processing speed of problematic documents. The structured processing and multi-dimensional analysis of settlement order information not only facilitates data management and retrieval, but also provides medical institutions with powerful data analysis capabilities, which helps to optimize resource allocation and cost control. The data visualization function enables management to intuitively understand the distribution and trend of settlement data, and provides data support for policy formulation and process improvement. Combining the above technical improvements, the operating costs of medical institutions have been significantly reduced, patient satisfaction has been improved, and the overall quality of medical services has been promoted. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a flow chart of the method for verifying and processing medical settlement form information of the present invention;
[0028] Figure 2 The present invention is a block diagram of a medical settlement form information verification and processing system. DETAILED DESCRIPTION
[0029] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0030] The present invention provides Figure 1 A method for verifying and processing medical settlement form information is shown, comprising the following steps:
[0031] The image information of the paper medical settlement form is obtained through the image acquisition device, and the text information in the image is converted into editable electronic data using the OCR technology, and the electronic data is converted into a medical settlement form file;
[0032] Based on the intelligent model of deep learning, it can identify and supplement the missing information in the medical settlement documents, and verify them through multiple dimensions such as item matching, amount verification and patient identity verification;
[0033] For settlement orders that fail verification, analyze the error type and provide error reports and correction suggestions, and guide the backend manual review and intervention for correction;
[0034] Structuralize and organize the verified settlement form information to form standardized data, classify and summarize the standardized data according to different dimensions of medical projects, departments, and time, and use charts and graphs to show the distribution characteristics and time series trends of medical settlement data;
[0035] Standardized data is backed up regularly to remote servers or cloud storage, and support query, export and printing of medical settlement information.
[0036] Furthermore, the use of the image acquisition device and the information processing process are specifically implemented as follows:
[0037] Image acquisition equipment configuration: Use a high-precision scanner as the main device for image acquisition. The scanner should have high-resolution scanning capabilities, at least 600 DPI (dots per inch) or above, to ensure that all details on the medical bill are captured, including small text and barcodes. The scanner should have a document feeding (ADF) function to batch process multiple pages of bills and improve work efficiency. Configure color correction and denoising functions to ensure that clear, high-quality images can be obtained even in poor lighting or when the bill is slightly stained.
[0038] Image preprocessing: Preprocess the scanned image, including but not limited to grayscale and binarization, remove noise and shadows in the image, and optimize OCR recognition effect. Apply image enhancement technology, such as contrast adjustment and tilt correction, to ensure that the text part is more prominent and easy to recognize.
[0039] Application of OCR technology: Use OCR software to convert pre-processed high-definition images into electronic text. The software should support multiple fonts and character sets, including special medical symbols and abbreviations, to accommodate the diversity of medical bills. Implement a multi-layer recognition strategy, recognize the full text of the first line, and then perform refined recognition for specific areas (such as amount, drug name, patient information, etc.) to improve recognition accuracy.
[0040] Data post-processing and verification: Match the OCR-recognized data with the preset medical settlement form template and fill it into the corresponding spreadsheet or database fields.
[0041] Manual review assistance: For fields that are uncertain or erroneous in OCR recognition, mark and prompt manual verification and correction to ensure the final accuracy of the information. Provide a user interface that allows operators to view the original image and recognition results directly on the interface to facilitate rapid location and correction of errors. Through this series of steps, the solution can effectively ensure that the medical settlement form images obtained by high-precision scanners are accurately converted into editable electronic data through OCR technology, thus laying a solid foundation for subsequent intelligent verification, processing, storage and management.
[0042] Further, regarding the specific implementation of the intelligent model, the process of deep learning training using a combination of a convolutional neural network (CNN) and a recurrent neural network (RNN) is as follows:
[0043] Sample collection: Collect a large number of annotated medical billing form samples from medical institutions, including correct and incorrect examples, to ensure that the data covers various formats and contents of billing forms.
[0044] Data preprocessing: Clean the collected samples, including denoising, format unification, data enhancement, etc., to improve the generalization ability of the model.
[0045] Feature engineering: Extract key features of medical billing statements, such as project code, service description, cost, patient information, etc., as structured data for model input.
[0046] Model building phase
[0047] CNN module design: Use convolutional neural network (CNN) to process the visual features of settlement form images, such as recognizing text, barcodes and table structures through convolutional layers, and extracting local features and patterns.
[0048] RNN module design: Combined with recurrent neural network (RNN) or long short-term memory network (LSTM), it processes sequence data such as continuous lists of items or text descriptions, captures contextual dependencies, and understands the overall logical structure of the settlement form.
[0049] Deep learning model training
[0050] Model integration: The image features processed by CNN and the sequence features processed by RNN are integrated to form an end-to-end deep learning model to jointly learn the structure and content features of medical bills.
[0051] Training strategy: Use the back-propagation algorithm with a loss function (such as cross entropy loss) to perform multiple iterative training on a large-scale labeled dataset to optimize model parameters.
[0052] Hyperparameter tuning: Adjust hyperparameters such as learning rate, batch size, number of layers, number of nodes, etc. through grid search, random search, etc. to find the optimal model configuration.
[0053] Model deployment and application
[0054] Integration into the system: Embed the trained model into the medical settlement form information processing system to realize real-time or batch intelligent verification and supplement of settlement form information.
[0055] Continuous optimization: After the model is launched, feedback data is continuously collected, and the model is adjusted according to new data through online learning or regular offline retraining to continuously improve the accuracy of recognition and supplementation.
[0056] Through the above steps, the solution effectively combines the advantages of CNN and RNN, specifically improves the accuracy of medical settlement form information recognition and supplementation, and provides technical support for the standardization and intelligence of the medical settlement process.
[0057] Furthermore, the implementation steps of the project matching verification rules are as follows:
[0058] Establish a medical project coding system: Establish or update a detailed medical project code database. This database should contain detailed information such as the unique code, project name, service description, and fee standard for each medical project.
[0059] Settlement form information extraction: Utilize OCR technology and intelligent information recognition module to extract all project codes and related information from medical settlement forms, including project number, project name, executing department, cost, etc.
[0060] Project code standardization: Standardize the extracted project codes to remove possible prefixes, suffixes or format differences to ensure consistency with the coding format in the database.
[0061] Real-time comparison: The standardized project code is compared with the medical project coding system database in real time. This process needs to ensure fast response and reduce system delays.
[0062] Matching verification: During the comparison process, the system needs to determine whether each project code on the settlement form has a corresponding record in the database. If there is a match, it will further confirm whether other information such as the project name, execution department, and cost are consistent; if not, it will be marked as a suspected error or non-compliant project.
[0063] Exception handling: For unmatched project codes, the system needs to initiate an exception handling process, including but not limited to: querying similar projects, prompting manual review intervention, or directly rejecting the approval of the settlement order information.
[0064] Reporting and feedback: For any mismatches or anomalies found during the verification process, the system generates a detailed verification report, including error code, project name, difference between expected and actual values, etc., and provides correction suggestions or guidance for quick correction.
[0065] Continuous optimization: Based on verification results and manual review feedback, the medical project coding system database is continuously optimized and updated to ensure its timeliness and accuracy, and to improve the accuracy of future settlement form matching verification.
[0066] Through the above method, the project matching verification rules can effectively ensure the accuracy of medical project information in the settlement form, reduce human errors, and improve the efficiency and quality of medical settlement.
[0067] Further details on the implementation of the amount verification process, including basic total consistency verification and detection of abnormally high fees or unreasonable fee combinations, can be summarized as follows:
[0068] Basic amount consistency check: First, the system reads the total amount listed on the medical settlement form through the intelligent verification module.
[0069] Next, the system parses the settlement statement item by item and adds up each medical service item and its corresponding fee.
[0070] The accumulated total of project expenses is accurately compared with the total amount on the settlement form to ensure that the two are consistent. If they are inconsistent, it will be recorded as an amount verification failure and enter the exception handling process.
[0071] Abnormally high cost detection: Establish a benchmark database for medical service costs, covering the normal price range of common medical services. For each medical service cost, the system compares it with the standard cost or historical average cost in the database. If a cost is significantly higher than its benchmark range, the system marks it as abnormally high cost, and further manual review is required to determine whether there is a billing error or special reason.
[0072] Analysis of unreasonable cost combinations: Develop algorithms to identify possible unreasonable cost combination patterns, such as unusual service item combinations, coexistence of mutually exclusive service items, etc. The system analyzes the service item combinations on the settlement form, matches patterns with conventional combinations in historical data, and identifies cost combinations that do not conform to medical logic or clinical practice. For the identified unreasonable combinations, the system will generate warning messages to indicate possible billing anomalies or entry errors.
[0073] Early warning and feedback mechanism: For abnormally high fees or unreasonable fee combinations detected, the system sends early warning notifications to responsible personnel, including specific abnormal item details and recommended review measures. Provides a user interface to display these anomalies, allowing operators to mark, comment or directly correct suspected problems directly in the system.
[0074] Continuous learning and optimization: Through machine learning algorithms, the system continuously optimizes the anomaly detection model based on historical verification results and manual review feedback to improve the ability to identify new or complex billing errors. Regularly update the cost benchmark database and unreasonable cost combination model to ensure synchronization with the latest medical billing policies and practices.
[0075] Through the above steps, this method not only realizes the basic amount verification, but also deeply detects potential billing errors through intelligent means, thereby improving the accuracy and rationality of medical settlement at the source.
[0076] Furthermore, the specific implementation steps for patient identity verification are achieved through integration with the hospital information system (HIS) to ensure the legitimacy of the patient's identity and the consistency of the settlement form information. The complete method is outlined as follows:
[0077] Data interface establishment: First, this system establishes a data interface with the hospital information system (HIS) to achieve real-time information interaction, including but not limited to the exchange protocol of patient basic information, medical records, expense details and other data.
[0078] Information collection and preprocessing: During the medical settlement form information entry stage, the system uses OCR technology or manual entry to collect key identity information such as the patient's name and ID number. The collected information is preliminarily cleaned and formatted to ensure standardization of the information so that it can be accurately matched with the data in the HIS system.
[0079] Real-time comparison and verification: The key patient information (such as name and ID number) on the settlement form is transmitted to the HIS system through the data interface in real time. After receiving the information, the HIS system will immediately compare it with the patient registration information stored in the system, including but not limited to the complete match of the name, the validity and uniqueness of the ID number.
[0080] Identity consistency confirmation: If all key information matches completely, the system confirms that the patient's identity is legitimate and the settlement form information is consistent with the HIS record, allowing the settlement process to continue. If a mismatch is found, such as a typo in the name, an incorrect ID number, etc., the system will suspend the settlement process and generate an exception report, indicating that manual intervention is required for verification and correction.
[0081] Exception handling and feedback: In the case of verification failure, the system provides detailed error information, such as whether the name does not match or the ID number is wrong, and may provide correction suggestions. At the same time, the system can trigger a notification mechanism to immediately send warning information to relevant personnel for quick response and processing.
[0082] Security and privacy protection: During the entire identity verification process, an encrypted communication protocol is adopted to ensure the security of patient personal information during transmission, in compliance with medical data privacy protection regulations.
[0083] Continuous integration and optimization: The system design needs to support continuous integration, and regularly synchronize patient information updates with the HIS system to ensure the timeliness and accuracy of verification. According to the problems and feedback encountered in actual applications, the identity verification algorithm and process are continuously optimized to improve verification efficiency and user experience. Through the above method, the real-time and accurate verification of patient identity between the medical settlement form information processing system and the hospital information system is realized, which effectively prevents identity fraud and information errors, and enhances the security and reliability of medical settlement.
[0084] Furthermore, the complete method for detailed classification labels of error reports and generating correction suggestions based on historical correction examples can be summarized as follows:
[0085] Intelligent verification and error classification: In the intelligent verification module, the system first conducts a comprehensive check on the medical settlement form information according to the preset multiple verification rules (such as item matching, amount verification, patient information verification, etc.). When an error is detected, the system classifies the error into a preset error type label, such as data omission (missing necessary information items), format error (incorrect date and amount format), logical conflict (treatment items do not match diagnosis), etc., to ensure that the error information is clearly marked and easy to identify.
[0086] Error report and suggestion output: For medical settlement forms that fail verification, the system not only provides a detailed error report that lists all error items and their classification labels, but also comes with correction suggestions generated based on historical cases. The error report is presented in an easy-to-read format, such as HTML or PDF, to ensure that the back-end manual reviewers can quickly understand the problem and how to correct it.
[0087] Interactive Correction Interface: Provides an interactive interface that allows manual reviewers to adopt or modify the correction suggestions generated by the system directly in the report to quickly correct errors. The interface design is intuitive and supports one-click application of suggestions, manual adjustments or adding notes, which optimizes the work efficiency of reviewers.
[0088] Feedback loop and continuous optimization: The system tracks the adoption of correction suggestions by manual reviewers and the final correction results, and uses this feedback as new training data to continuously optimize the machine learning model to improve the accuracy and practicality of future correction suggestions.
[0089] Through this series of steps, this solution aims to accelerate the back-end manual review process through intelligent means, reduce manual review time, and improve the accuracy of corrections, thereby improving the efficiency and quality of the entire medical settlement form processing.
[0090] Furthermore, it involves the structured sorting step, with particular emphasis on the application of natural language processing (NLP) technology to parse unstructured text and extract key information from medical billing statements. The specific implementation steps can be summarized as follows:
[0091] 1. Data preprocessing
[0092] Document import and format conversion: First, the system supports the import of medical settlement form files in various formats (such as PDF, pictures, Word documents, etc.), and uses OCR technology to convert pictures or non-text formats into processable text data.
[0093] Text cleaning: De-noise the converted text and remove irrelevant symbols, blank lines, advertising slogans and other non-critical information to ensure the accuracy of subsequent processing.
[0094] 2. Natural Language Processing (NLP) Parsing
[0095] Keyword recognition: Use the named entity recognition (NER) function in NLP technology to identify and mark key entities in medical settlement forms, such as "medical project", "department name", "date", "patient name", etc.
[0096] Semantic understanding: Use semantic analysis technology to understand the contextual relationships in the text, distinguish the different meanings of departments or projects with the same name, and ensure the accuracy of extracted information.
[0097] Information extraction: Based on contextual understanding and preset rules, key information can be accurately extracted from the text, such as identifying and extracting specific drug names, inspection items, execution dates, etc.
[0098] 3. Structured information construction
[0099] Template matching and data mapping: The extracted key information is structured according to the preset template. The template defines the various information fields required for the medical settlement form, such as "drug expenses", "examination fees", "treatment fees", etc., as well as their arrangement order and data type.
[0100] Filling: Fill the extracted information into the corresponding template fields to form a structured data record.
[0101] 4. Output and verification
[0102] Formatted output: Output structured data in a unified format, such as XML, CSV, or directly into a structured table in a database to facilitate subsequent processing, storage, and query.
[0103] Data verification: Perform consistency check on structured data to ensure that all required items are complete and the data type complies with the specifications, such as the date format is correct and the amount value is reasonable.
[0104] Error handling and feedback: If any problems are found that fail the verification, the system will provide a detailed error report and, if possible, make correction suggestions so that manual intervention can be made to correct them.
[0105] Through the above steps, this solution realizes the efficient and accurate structured processing of the unstructured text of medical settlement forms, laying a solid foundation for subsequent intelligent verification, processing, storage and management.
[0106] Furthermore, with regard to the specific implementation steps of the backup mechanism, distributed storage technology is used to ensure the security and recoverability of medical settlement form information. The complete method is as follows:
[0107] 1. Data standardization and encryption
[0108] Data standardization: First, the information processing module is used to structure the verified medical settlement form information to form a standardized data format for easy storage and retrieval.
[0109] Data encryption: Advanced encryption is performed on standardized data, using industry-standard encryption algorithms (such as AES-256) to ensure the confidentiality of data during transmission and storage.
[0110] 2. Distributed storage architecture design
[0111] Multi-node deployment: Create storage nodes on multiple remote servers or cloud storage service providers, each of which stores a copy of standardized data. This disperses storage risks, and even if a node fails, it will not affect the integrity of the data.
[0112] Data segmentation and replication: Split the standardized data set into multiple data blocks, use redundant storage strategies (such as RAID technology or data replication), and save multiple copies on different storage nodes to improve data availability.
[0113] 3. Real-time data synchronization
[0114] Synchronization mechanism: Implement a real-time data synchronization strategy to ensure that whenever the data in the primary database changes, these changes are immediately pushed to all backup nodes to maintain data consistency.
[0115] Version control: A version control mechanism is added during the synchronization process to record the historical version of each data change, making it easier to select the correct version to roll back when restoring data.
[0116] 4. Regular integrity checks
[0117] Integrity check: Set periodic data integrity check tasks to verify data integrity by calculating the checksum of the data block (such as MD5 or SHA-256 hash value) and comparing the checksum of the backup data with the primary database data.
[0118] Health monitoring: Real-time monitoring of the operating status of storage nodes, including disk usage, network connection status, system stability, etc., to ensure the health of the data storage environment.
[0119] 5. Access Control and Auditing
[0120] Permission management: Implement strict access control mechanisms to ensure that only authorized users and systems can access stored data, avoiding unauthorized access or tampering.
[0121] Audit log: A log that records all data access and operations, including backup and recovery operations, to facilitate traceability and auditing and enhance security.
[0122] Through this series of measures, the backup mechanism ensures the high availability, integrity and security of medical settlement information in a distributed storage environment, providing strong protection for the stable operation of the medical information system.
[0123] On the other hand, the present invention provides a method for verifying and processing medical settlement form information, such as Figure 2 As shown, including:
[0124] The image acquisition and OCR conversion module is used to obtain the image information of the paper medical settlement form through the image acquisition device, and use the OCR technology to convert the text information in the image into editable electronic data, and convert the electronic data into a medical settlement form file;
[0125] The information completion and verification module is used to identify and supplement the missing information in the medical settlement form based on the intelligent model of deep learning, and verify it through multiple dimensions such as item matching, amount verification, and patient identity verification;
[0126] The error analysis and correction suggestion module is used to analyze the error type and provide error reports and correction suggestions for settlement orders that fail verification, and guide the back-end manual review and intervention correction;
[0127] The data structuring and analysis display module is used to structure and organize the verified settlement form information to form standardized data, classify and summarize the standardized data according to different dimensions of medical projects, departments, and time, and use charts and graphs to display the distribution characteristics and time series trends of medical settlement data;
[0128] The data backup and management module is used to regularly back up standardized data to a remote server or cloud storage, and supports the query, export and printing of medical settlement form information.
[0129] In summary, the present invention realizes the rapid conversion from paper documents to electronic data, greatly shortens the information processing cycle, greatly improves data accuracy through deep learning models and multiple verification rules, reduces disputes caused by errors, and enhances the credibility of the settlement process. The introduced error analysis and correction suggestion mechanism provides precise guidance for manual review and accelerates the processing speed of problematic documents. The structured processing and multi-dimensional analysis of settlement order information not only facilitates data management and retrieval, but also provides medical institutions with powerful data analysis capabilities, which helps in resource optimization and cost control. The data visualization function enables management to intuitively understand the distribution and trends of settlement data, and provides data support for policy formulation and process improvement. The above technical improvements significantly reduce the operating costs of medical institutions, improve patient satisfaction, and promote the overall improvement of medical service quality.
[0130] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for verifying and processing medical settlement form information, characterized in that: The following steps are involved: The image information of the paper medical settlement form is obtained through the image acquisition device, and the text information in the image is converted into editable electronic data using the OCR technology, and the electronic data is converted into a medical settlement form file; Based on the intelligent model of deep learning, it can identify and supplement the missing information in the medical settlement documents, and verify them through multiple dimensions such as item matching, amount verification and patient identity verification; For settlement orders that fail verification, analyze the error type and provide error reports and correction suggestions, and guide the backend manual review and intervention for correction; Structuralize and organize the verified settlement form information to form standardized data, classify and summarize the standardized data according to different dimensions of medical projects, departments, and time, and use charts and graphs to show the distribution characteristics and time series trends of medical settlement data; Standardized data is backed up regularly to remote servers or cloud storage, and support query, export and printing of medical settlement information.
2. A method for verifying and processing medical settlement form information according to claim 1, characterized in that: The use of the image acquisition device and the information processing process include: Pre-process the images obtained by the scanner to remove noise and shadows in the images, use OCR software to convert the pre-processed high-definition images into electronic text, and match the data recognized by OCR with the preset medical settlement form template.
3. A method for verifying and processing medical settlement form information according to claim 1, characterized in that: The verification rules for the project matching include: Establish a medical project code database; use OCR technology and intelligent information recognition modules to extract all project codes and related information from medical settlement forms, and then standardize the project codes: standardize the extracted project codes; compare the standardized project codes with the medical project coding system database in real time, and then obtain the verification results.
4. A method for verifying and processing medical settlement form information according to claim 3, characterized in that: During the real-time comparison process, determine whether each project code on the settlement form has a corresponding record in the database; if there is a match, then confirm whether the project name, execution department, and cost information are consistent; If there is any inconsistency, it will be marked as a non-compliant item.
5. A method for verifying and processing medical settlement form information according to claim 1, characterized in that: The verification rules for the amount verification include: The total amount listed on the medical settlement form is read through the intelligent verification module; the settlement form is parsed item by item, and each medical service item and its corresponding fee are added up; the total of the accumulated item fees is compared with the total amount on the settlement form.
6. A method for verifying and processing medical settlement form information according to claim 1, characterized in that: The patient identity verification rules include: Establish a data interface with the hospital information system, collect patient information during the medical settlement form information entry stage, and transmit the patient information on the settlement form to the hospital information system through the data interface in real time. After the HIS system receives the information, it will immediately compare it with the patient registration information.
7. A method for verifying and processing medical settlement form information according to claim 5, characterized in that: The analysis error type and provision of error reports and correction suggestions include: In the intelligent verification module, the medical settlement form information is checked according to multiple preset verification rules. When an error is detected, the error is classified under a preset error type label. For medical settlement forms that fail the verification, error reports and correction suggestions are provided.
8. A method for verifying and processing medical settlement form information according to claim 1, characterized in that: The structured arrangement of the verified settlement order information includes: Convert the image or non-text format of the medical settlement form into processable text data, then extract the key information, structure the extracted key information according to the preset template, and output the structured data in a unified format.
9. A method for verifying and processing medical settlement form information according to claim 1, characterized in that: The standardized data is backed up regularly to a remote server or cloud storage, including: The medical settlement form information that has passed the verification is structured to form a standardized data format, and the standardized data is encrypted at an advanced level; storage nodes are created on multiple remote servers or cloud storage service providers, with each node storing a copy of the standardized data; the standardized data set is divided into multiple data blocks, and multiple copies are saved on different storage nodes.
10. A method for verifying and processing medical settlement form information, characterized in that: include: The image acquisition and OCR conversion module is used to obtain the image information of the paper medical settlement form through the image acquisition device, and use the OCR technology to convert the text information in the image into editable electronic data, and convert the electronic data into a medical settlement form file; The information completion and verification module is used to identify and supplement the missing information in the medical settlement form based on the intelligent model of deep learning, and verify it through multiple dimensions such as item matching, amount verification, and patient identity verification; The error analysis and correction suggestion module is used to analyze the error type and provide error reports and correction suggestions for settlement orders that fail verification, and guide the back-end manual review and intervention correction; The data structuring and analysis display module is used to structure and organize the verified settlement form information to form standardized data, classify and summarize the standardized data according to different dimensions of medical projects, departments, and time, and use charts and graphs to display the distribution characteristics and time series trends of medical settlement data; The data backup and management module is used to regularly back up standardized data to a remote server or cloud storage, and supports the query, export and printing of medical settlement form information.
Citation Information
Cited By
Intelligent file scanning and storing integrated cabinet
CN119324959A
Environmental data and quality information management system based on artificial intelligence
CN119848037A
Artificial intelligence-based anesthesia record list and charging list error detection system and method
CN120220990A
Artificial intelligence-based anesthesia record and charge sheet error detection system and method
CN120220990B
Cloud management platform charging strategy determination method
CN120410654A
