Clinical information management method and related equipment

Abstract

The invention relates to the technical field of information management, in particular to a clinical information management method, which comprises the following steps: constructing an alliance chain network, calling an unstructured clinical document, extracting corresponding key information from the unstructured clinical document, converting the key information into structured data, and storing the structured data in a database; associating the structured data to generate a corresponding clinical information base; performing multi-dimensional analysis on the structured data, generating a corresponding clinical decision support report, recording the clinical decision support report in a clinical information base, executing a data sharing protocol on the clinical decision support report, and determining a corresponding authority mechanism; and determining a user role of the visitor, determining a corresponding target access permission based on the user role, judging whether access can be performed based on the target access permission and a permission mechanism, if so, calling corresponding target information in the clinical information base, and providing the target information to the visitor in a graphical interface. According to the invention, effective management of clinical information can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of information management, and in particular to a clinical information management method and related equipment. Background Art

[0002] With the rapid development of information technology, clinical information management systems have become an indispensable part of modern hospitals; however, existing clinical information management systems have some significant problems that limit their potential in improving the quality and efficiency of medical services. First, the so-called "information island" phenomenon has been caused by the lack of communication between the information systems of various medical institutions, which has seriously affected the cross-institutional patient information sharing and collaborative treatment.

[0003] Secondly, insufficient patient privacy protection measures can easily lead to sensitive information leakage, bringing unnecessary risks to patients. In addition, traditional clinical information management systems have limited capabilities in data analysis and processing, and cannot fully utilize big data and artificial intelligence technologies to assist doctors in making more accurate diagnosis and treatment decisions.

[0004] From the above, we can see that how to achieve effective management of clinical information remains to be solved. Summary of the invention

[0005] In order to achieve effective management of clinical information, the present application provides a clinical information management method and related equipment.

[0006] In the first aspect, the present application provides a clinical information management method, which adopts the following technical solution: A clinical information management method comprises: constructing a consortium chain network, retrieving unstructured clinical documents, extracting corresponding key information from the unstructured clinical documents, converting the key information into structured data, and using knowledge graph technology to associate the structured data to generate a corresponding clinical information base, wherein each node in the consortium chain network represents a certified medical institution; performing multi-dimensional analysis on the structured data based on a federated learning algorithm, generating a corresponding clinical decision support report based on the corresponding analysis result data, recording the clinical decision support report in the clinical information base, executing a data sharing agreement on the clinical decision support report, and determining a permission mechanism corresponding to the clinical decision support report, wherein the clinical decision support report includes disease diagnosis, treatment recommendations and prognosis evaluation; when accessing information, determining the user role of the visitor, and determining the corresponding target access rights based on the user role, judging whether access is possible based on the target access rights and the permission mechanism, and if so, retrieving the corresponding target information in the clinical information base, and providing the target information to the visitor in a graphical interface.

[0007] By adopting the above technical solutions, building a consortium chain network and using knowledge graph technology, we can extract and structure key information in unstructured clinical documents, combine federated learning algorithms for multi-dimensional analysis, generate clinical decision support reports, and ensure the safe sharing and effective management of clinical information through fine-grained permission control and dynamic access mechanisms, thereby improving the scientificity and accuracy of medical decisions.

[0008] Optionally, the method also includes: when the medical institution corresponding to the node in the alliance chain network generates new clinical data, immediately notifying the clinical information library through an event trigger mechanism to trigger data updating; determining the corresponding target medical institution, obtaining the latest clinical data corresponding to the target medical institution, and recording the version information of the latest clinical data, wherein the version information includes update time, update content and updater information; during the data update process, if multiple nodes submit updates to the same data at the same time, updating is performed based on a pre-set conflict resolution mechanism.

[0009] By adopting the above technical solutions, the clinical information base is updated in real time through the event trigger mechanism to ensure the timeliness and accuracy of the data. When new clinical data is generated, the system will immediately notify the clinical information base and record detailed version information, including update time, content and updater information. When multiple nodes submit updates at the same time, the system ensures the consistency and integrity of the data through a pre-set conflict resolution mechanism, thereby achieving efficient management and real-time synchronization of clinical data.

[0010] Optionally, in the process of generating a clinical decision support report, the method also includes: retrieving the medical history data of the corresponding patient, wherein the medical history data includes disease diagnosis, treatment plan, medication records, laboratory test results, and imaging data; retrieving a pre-trained prediction model, inputting the structured data and the medical history data into the prediction model, and obtaining corresponding prediction results, wherein the prediction results include the predicted probability of disease recurrence, the risk of complications, and the expected treatment effect; and recording the prediction results in the clinical decision support report.

[0011] By adopting the above technical solutions, by retrieving the patient's medical history data and pre-trained prediction models, the patient's disease recurrence probability, complication risk and treatment effect can be accurately predicted, and these prediction results can be included in the report to provide doctors with scientific and personalized clinical decision support, thereby improving the accuracy and effectiveness of medical decisions.

[0012] Optionally, after generating the clinical decision support report, the method further includes: sending the clinical decision support report to a doctor or other medical professional, obtaining modification feedback data of the clinical decision support report from the doctor or other medical professional; judging whether it is necessary to modify the clinical decision support report based on the modification feedback data, wherein the modification content is the diagnosis conclusion of the disease, the specific treatment recommendation, and the detailed description of the prognosis assessment in the clinical decision support report; if necessary, modifying the clinical decision support report, recording the corresponding modification data, and saving the versions of the clinical decision support report before and after the modification, wherein the modification data includes the modification time, modification content and modifier information.

[0013] By adopting the above technical solution, by collecting modification feedback from doctors or other medical professionals, necessary adjustments and optimizations are made to the report, and detailed modification data, including modification time, content and modifier information, are recorded to ensure the accuracy and practicality of the report. At the same time, the versions before and after the modification are retained for easy auditing and traceability.

[0014] Optionally, after each information access, the method also includes: collecting the visitor's operation log, wherein the operation log includes login time, access frequency, access data type, and operation type; performing analysis based on the operation log to obtain the visitor's corresponding operation habits and access patterns; adjusting the visitor's authority level based on the operation habits and access patterns, and sending the adjusted authority level to the administrator for review.

[0015] By adopting the above technical solution, by collecting and analyzing the visitor's operation logs, identifying their operation habits and access patterns, and dynamically adjusting the visitor's permission level accordingly, the adjusted permission level is sent to the administrator for review, thereby ensuring the security of the system and preventing permission abuse and potential security risks.

[0016] Optionally, the method also includes: retrieving the visitor's operation log and history log, and detecting whether the visitor has abnormal behavior based on the operation log and the history log; if so, determining the corresponding abnormal log, and determining the corresponding abnormal level based on the abnormal log; and judging whether to reduce the visitor's authority level or stop the authority based on the abnormal level.

[0017] By adopting the above technical solution, by retrieving the visitor's operation log and historical log, it is detected whether there is abnormal behavior, and the abnormal level is determined based on the abnormal log, and then it is decided whether to reduce the visitor's authority level or suspend his authority, thereby enhancing the security of the system and timely preventing and responding to potential security threats.

[0018] Optionally, the method also includes: regularly checking the operating status of each node in the alliance chain network, the operating status including online status, resource utilization, network connection status, node offline, response timeout, and resource exhaustion; when a node failure is detected, obtaining the current load status of each node in the alliance chain network, and determining the corresponding target node based on all current load conditions; transferring the corresponding tasks of the failed node to the target node.

[0019] By adopting the above technical solution, the operating status of each node in the alliance chain network is regularly checked to timely detect and handle node failures. When a node failure is detected, the system will automatically transfer the tasks of the failed node to the appropriate target node based on the current load of each node, ensuring the high availability of the system and the continuity of data.

[0020] In a second aspect, the present application provides a clinical information management method, which adopts the following technical solution: A clinical information management related device, comprising: A clinical information base generation module builds a consortium chain network, retrieves unstructured clinical documents, extracts corresponding key information from the unstructured clinical documents, converts the key information into structured data, and uses knowledge graph technology to associate the structured data to generate a corresponding clinical information base, wherein each node in the consortium chain network represents a certified medical institution; A clinical decision support report generation module, which performs multi-dimensional analysis on structured data based on a federated learning algorithm, generates a corresponding clinical decision support report based on the corresponding analysis result data, records the clinical decision support report in the clinical information database, executes a data sharing agreement on the clinical decision support report, and determines the authority mechanism corresponding to the clinical decision support report, wherein the clinical decision support report includes disease diagnosis, treatment recommendations and prognosis assessment; The access judgment module determines the user role of the visitor when accessing information, and determines the corresponding target access rights based on the user role. The target access rights and the permission mechanism are used to determine whether access is possible. If so, the corresponding target information in the clinical information library is retrieved and provided to the visitor in a graphical interface.

[0021] In a third aspect, the present application provides a clinical information management method, which adopts the following technical solution: A clinical information management method comprises a processor, wherein the program of the clinical information management method described above is run in the processor.

[0022] In a fourth aspect, the present application provides a storage medium, which adopts the following technical solution: A storage medium stores a program of the clinical information management method described above.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: By building a consortium chain network and using knowledge graph technology, we can extract and structure key information from unstructured clinical documents, combine federated learning algorithms for multi-dimensional analysis, generate clinical decision support reports, and ensure the safe sharing and effective management of clinical information through fine-grained permission control and dynamic access mechanisms. In addition, the clinical information database is updated in real time through an event trigger mechanism to ensure the timeliness and accuracy of the data, while the consistency and integrity of the data are ensured through a pre-set conflict resolution mechanism, thereby achieving efficient management and real-time synchronization of clinical data.

[0024] In addition, by retrieving the patient's medical history data and pre-trained prediction models, the patient's disease recurrence probability, complication risk and treatment effect are accurately predicted, and these prediction results are included in the report to provide doctors with scientific and personalized clinical decision support and improve the accuracy and effectiveness of medical decisions. At the same time, by collecting and analyzing the visitor's operation logs, identifying their operation habits and access patterns, and dynamically adjusting the visitor's permission level, the system security is ensured, permission abuse and potential security risks are prevented, and the reliability and security of clinical information management are further enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention is a flowchart of a clinical information management method according to an exemplary embodiment.

[0026] Figure 2 It is a structural block diagram of a clinical information management related device according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0028] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0029] The present application embodiment discloses a clinical information management method, referring to Figure 1 ,include: S100 builds a consortium chain network to retrieve unstructured clinical documents, extract corresponding key information from unstructured clinical documents, convert key information into structured data, and use knowledge graph technology to associate structured data to generate a corresponding clinical information database.

[0030] Each participating medical institution must pass a strict certification process to ensure its legitimacy and credibility. The certification process may include qualification review, security audit, etc. The alliance chain network is built using blockchain technology, and each node represents a certified medical institution. The nodes use consensus mechanisms (such as PBFT, Raft, etc.) to ensure data consistency and immutability. At the same time, a clear data sharing agreement needs to be formulated to specify data access rights, sharing scope and usage rules to ensure data security and compliance.

[0031] Then, unstructured clinical documents can be retrieved from the electronic medical record systems of various medical institutions through API interfaces, file uploads, etc., and the retrieved unstructured clinical documents are stored in the distributed storage system of the alliance chain network to ensure data security and backup. Use NLP technologies, such as named entity recognition (NER) and text classification, to extract key information from unstructured documents, such as basic patient information, disease diagnosis, treatment plans, medication records, etc. The extracted key information is cleaned and standardized to remove noise and missing values ​​to ensure data quality and consistency; the cleaned key information is converted into structured data formats, such as JSON, XML, etc.

[0032] Finally, using knowledge graph technology, the entities and relationships in structured data are modeled to form a knowledge graph containing multiple entities and relationships. Through entity linking and relationship reasoning, data from different sources can be associated to form a comprehensive clinical information database, and the generated clinical information database needs to be stored in the distributed database of the alliance chain network. S110, perform multi-dimensional analysis on structured data based on the federated learning algorithm, generate a corresponding clinical decision support report based on the corresponding analysis result data, record the clinical decision support report in the clinical information database, execute the data sharing agreement on the clinical decision support report, and determine the authority mechanism corresponding to the clinical decision support report.

[0033] Among them, the pre-built federated learning model is called up, each node trains the data locally, and the model parameters are aggregated through a secure aggregation algorithm to generate a global model. Based on the federated learning model, structured data can be analyzed in multiple dimensions, such as disease diagnosis, treatment recommendations, and prognosis evaluation; at the same time, a clinical decision support report is generated based on the analysis results, and the report content includes disease diagnosis, treatment recommendations, and prognosis evaluation.

[0034] In addition, according to the data sharing agreement, clinical decision support reports are stored in the alliance chain network to ensure data transparency and traceability. At the same time, a fine-grained permission control mechanism is set up to determine access rights based on user roles to ensure that only authorized users can access sensitive data in the report.

[0035] S120, when accessing information, determine the user role of the visitor, and determine the corresponding target access rights based on the user role, and determine whether access is possible based on the target access rights and the permission mechanism. If so, retrieve the corresponding target information in the clinical information library and provide the target information to the visitor in a graphical interface.

[0036] Among them, the visitor first needs to pass identity authentication, such as username and password, multi-factor authentication (MFA), etc., to ensure that the visitor is a legitimate user. Then the system identifies the user role based on the visitor's login information. The user role may include but is not limited to doctors, nurses, researchers, administrators, etc., and binds the visitor's identity to his or her user role and stores it in the system for subsequent permission control.

[0037] In addition, the system has preset permission configuration tables for different user roles, which define the data types and operation permissions that each role can access. For example, doctors can access all medical records of patients, but can only modify the parts they are responsible for; nurses can view the basic information and nursing records of patients, but cannot modify the diagnosis information. The system reads the corresponding target access rights from the permission configuration table according to the user role of the visitor; in addition, in order to improve access efficiency, the system can cache the visitor's permission information in memory to reduce the permission query time for each visit.

[0038] After receiving an access request, the system first verifies whether the visitor's permissions are consistent with the requested resources. For example, if a visitor requests to view a patient's detailed medical records, the system will check whether the visitor has the permission to view the patient's medical records. The system further determines whether the visitor's permissions meet the requirements of the access request based on the preset permission mechanism. For example, the system may check whether the visitor accesses within the allowed time period, whether the visitor accesses from a permitted terminal device, etc. Regardless of whether the access request is successful, the system will record the access log, including visitor information, access time, access resources, access results, etc., for subsequent auditing and tracing.

[0039] The system queries the corresponding target information from the clinical information database based on the parameters in the access request. For example, if a visitor requests to view a patient's medical record, the system will query the patient's medical record data. The system will filter the loaded data based on the visitor's permissions to ensure that the visitor can only see the part he or she has permission to access. For example, if the visitor is a nurse, the system will only display the patient's basic information and nursing records, and will not display detailed diagnostic information.

[0040] Finally, the loaded target information is rendered on the graphical interface to ensure the correctness and completeness of the data. For example, the patient's medical records are displayed in a table and the examination results are displayed in a chart.

[0041] Based on executing steps S100 to S120, by building a consortium chain network and using knowledge graph technology, the extraction and structuring of key information in unstructured clinical documents are realized, and multi-dimensional analysis is performed in combination with federated learning algorithms to generate clinical decision support reports. Through fine-grained permission control and dynamic access mechanisms, the secure sharing and effective management of clinical information are ensured.

[0042] For medical institutions corresponding to the alliance chain network, new data is generated every day. In order to ensure that the data in the clinical information database can be updated in a timely manner, the method also includes: S101: When new clinical data is generated by the medical institution corresponding to the node in the alliance chain network, the clinical information database is immediately notified through the event trigger mechanism to trigger data update Among them, an event detection mechanism is set up in each node (i.e., medical institution) to monitor the generation and update of clinical data; for example, whether new clinical data is generated can be detected through database triggers, file system monitoring, etc. When new clinical data is detected, an event is immediately triggered. This event can be a message in a message queue, an HTTP request, a WebSocket message, etc. After the event is triggered, the event information is sent to the clinical information library through a pre-configured notification mechanism. For example, the event message can be pushed to the clinical information library through a message queue (such as Kafka, RabbitMQ), or the clinical information library can be directly notified through an API call.

[0043] S102, determining the corresponding target medical institution, obtaining the latest clinical data corresponding to the target medical institution, and recording the version information of the latest clinical data.

[0044] After receiving the latest clinical data, the clinical information database generates version information, which includes update time, update content, and updater information. For example, the update time can be the timestamp of data reception, the update content can be the specific clinical data changes, and the updater information can be the medical institution or individual who submitted the data. The generated version information is stored in the clinical information database together with the latest clinical data.

[0045] S103, during the data update process, if multiple nodes submit updates for the same data at the same time, the update is performed based on a pre-set conflict resolution mechanism.

[0046] When receiving new clinical data, the clinical information database checks whether other nodes have submitted the same update at the same time, and can perform conflict detection through the unique identifier of the data (such as the primary key).

[0047] If a conflict is detected, it will be handled according to the pre-set conflict resolution mechanism. Common conflict resolution mechanisms include: Timestamp priority: select the data with a newer update timestamp as the final update. Version number priority: select the data with a higher version number as the final update. Node priority: select the data submitted by the node with a higher priority as the final update based on the node priority (such as hospital grade, data source reliability). Manual review: Submit the conflicting data to the administrator for manual review, and the administrator will decide the final updated data.

[0048] Finally, after the final updated data is determined according to the conflict resolution mechanism, the data is updated to the clinical information database and a log of the update operation is recorded.

[0049] Through the above detailed steps from S101 to S103, real-time update and management of newly generated clinical data are realized in the alliance chain network. The specific functions include: real-time update to ensure the timeliness and accuracy of data, data traceability to ensure data traceability and version management, conflict resolution to ensure data consistency and integrity, and data security to ensure data security and privacy protection during transmission and storage. In summary, this method not only improves the real-time and accuracy of clinical data, but also ensures the security and consistency of data, providing strong support for the information construction and data management of the medical industry.

[0050] In generating the clinical decision support report, the method further comprises: S111, retrieve the medical history data of the corresponding patient.

[0051] The patient's medical history data is queried from the clinical information database, and the queried medical history data is integrated to form a complete data set to ensure the integrity and consistency of the data. Medical history data includes but is not limited to the following: Disease diagnosis: The patient’s previous disease diagnosis records.

[0052] Treatment plan: The treatment plan and treatment process the patient has received.

[0053] Medication records: type of medication used by the patient, dosage, medication time, etc.

[0054] Laboratory test results: The patient's blood test, urine test, biochemical indicators and other laboratory test results.

[0055] Imaging data: X-rays, CT scans, MRI and other imaging data of the patient.

[0056] S112, retrieve the pre-trained prediction model, input the structured data and medical history data into the prediction model, and obtain the corresponding prediction result.

[0057] Among them, the patient's structured data and medical history data are preprocessed to ensure that the data format meets the input requirements of the prediction model. Preprocessing may include data cleaning, feature extraction, data standardization, etc. The preprocessed data is input into the prediction model, and the input data can be a multidimensional array, where each dimension represents a feature; the prediction model calculates based on the input data and generates prediction results. The prediction results may include: Probability of disease recurrence: The likelihood that a patient's disease will recur in the future.

[0058] Risk of complications: The risk of complications that a patient may experience in the future.

[0059] Treatment effect expectation: The expected effect of the patient's current treatment plan.

[0060] S113, record the prediction results in a clinical decision support report.

[0061] Among them, a clinical decision support report is generated based on the prediction results, and the prediction results are recorded in detail in the report to ensure the completeness and accuracy of the information. The content of the record includes the specific values, explanations and suggestions of the prediction results. The report content includes but is not limited to: Disease Diagnosis: The patient's current disease diagnosis.

[0062] Treatment recommendations: Treatment recommendations based on predicted results, such as adjusting drug dosage, adding adjuvant therapy, etc.

[0063] Prognosis assessment: The patient's prognosis assessment, including the probability of disease recurrence, the risk of complications, and the expected treatment effect.

[0064] Through the above detailed steps from S111 to S113, when generating a clinical decision support report, the patient's medical history data is retrieved and analyzed using the pre-trained prediction model to generate accurate prediction results and record them in the report. This process not only provides personalized decision support and improves the accuracy and reliability of predictions, but also ensures the integrity and traceability of information, thereby significantly improving the scientificity and effectiveness of medical decisions and improving the quality of medical care.

[0065] After generating the clinical decision support report, the method further includes: S114, sending the clinical decision support report to the doctor or other medical professional, and obtaining the doctor or other medical professional's feedback data on the modification of the clinical decision support report.

[0066] Among them, the generated clinical decision support report will be prepared in electronic form (such as PDF, HTML, etc.), ensuring that the report format is standardized and the content is complete, and the report will be sent to doctors or other medical professionals through secure communication channels. Common sending methods include e-mail, internal messaging system, secure file transfer protocol (such as SFTP), etc.; ensure that the report is sent successfully, and record the sending time and receipt confirmation information for subsequent tracking and management.

[0067] Then, provide multiple feedback channels, such as online forms, emails, internal systems, etc., to facilitate doctors or other medical professionals to submit feedback, and the requirements for feedback content need to be clear, including the specific content of the modification suggestions, the reasons for the modification, etc. For example, the doctor may suggest modifying the diagnosis conclusion of the disease, the specific treatment recommendations, the detailed description of the prognosis assessment, etc.; and the system automatically collects feedback data, records the feedback time, feedback content and feedback information, to ensure the integrity and traceability of the feedback data.

[0068] S115, judging whether the clinical decision support report needs to be modified based on the modified feedback data.

[0069] Among them, the system automatically or manually reviews the collected feedback data, evaluates the rationality and necessity of the feedback, and decides whether the report needs to be modified based on the preset judgment criteria. For example, if the feedback content involves important diagnostic conclusions or treatment recommendations, and there is sufficient medical basis, the system will determine that modification is required; if modification is required, the system will notify the relevant person in charge to confirm the modification request and record the confirmation information. It should be pointed out here that the modification content is the diagnostic conclusion of the disease, the specific treatment recommendation plan, and the detailed description of the prognosis assessment in the clinical decision support report; S116: If necessary, modify the clinical decision support report, record the corresponding modification data, and save the versions of the clinical decision support report before and after the modification.

[0070] Among them, the clinical decision support report is modified according to the feedback content, and the modification data is recorded, including the modification time, modification content and modifier information. The modified data can be stored in a log file or database to ensure the integrity and traceability of the data. At the same time, the versions of the clinical decision support report before and after the modification are saved to ensure that each version has a clear identification and record for subsequent audit and traceability.

[0071] Through the above detailed steps S114 to S116, after generating the clinical decision support report, the report is sent to doctors or other medical professionals, their modification feedback is collected and evaluated, and it is decided whether to modify it, and the modification data and version information are recorded and saved. Not only an effective feedback mechanism is established to improve the accuracy and practicality of the report, but also the integrity and traceability of the data are ensured through version management and modification records, so as to continuously improve the quality of clinical decision support reports, enhance the scientificity and effectiveness of medical decision-making, and ultimately improve medical quality and patient satisfaction.

[0072] After each information access, the method further includes: S121, collecting the visitor's operation log.

[0073] Among them, the system automatically generates an operation log after each information access; the generated operation log is stored in the log management system to ensure the integrity and security of the log. The log management system can use a database, log file or a dedicated log management platform, and regularly back up the operation log to prevent data loss and ensure the long-term availability of the log.

[0074] It should be noted here that the log content includes but is not limited to: Login Time: The time when the visitor logged into the system.

[0075] Visit frequency: the number of visits by a visitor within a certain period of time.

[0076] Type of data accessed: The type of data accessed by the visitor, such as medical records, test results, imaging materials, etc.

[0077] Operation type: The type of operation performed by the visitor, such as view, modify, delete, etc.

[0078] S122, analyzing based on the operation log to obtain the operation habits and access patterns corresponding to the visitor.

[0079] Among them, the operation log data is extracted from the log management system, and the operation log is analyzed using data analysis tools and technologies (such as data mining and machine learning). Based on the analysis results, the behavior patterns and operation habits of visitors are identified, for example, identifying that some visitors often access sensitive data at night, or that some visitors frequently modify patient information. It should be pointed out here that the analysis content includes but is not limited to: Visit frequency analysis: Count the visitor's visit frequency in different time periods and identify visit peaks and troughs.

[0080] Data type analysis: Analyze the data types accessed by visitors and identify the data areas that visitors are interested in.

[0081] Operation type analysis: Analyze the types of operations performed by visitors and identify the main operation behaviors of visitors.

[0082] S123, adjusting the visitor's permission level based on the operation habits and access mode, and sending the adjusted permission level to the administrator for review.

[0083] Among them, based on the identified operating habits and access patterns, the visitor's permission requirements and potential risks are evaluated. For example, if the visitor frequently accesses sensitive data, it may be necessary to lower his permission level; then based on the evaluation results, the visitor's permission level is dynamically adjusted. Permission adjustments may include increasing or decreasing access rights, restricting specific operations, etc.; at the same time, the operation log of permission adjustments is recorded, including adjustment time, adjustment reason, changes in permissions before and after adjustment, etc., to ensure the traceability of the adjustment process.

[0084] In addition, the adjusted permission level is sent to the system administrator to generate an audit request, which includes the reason for the adjustment, the change in permissions before and after the adjustment, and other information; the administrator reviews the permission adjustment request through the audit system. The audit process can include preliminary review, re-review and other links to ensure the rigor and fairness of the audit; and the administrator decides whether to approve the permission adjustment based on the audit results. If approved, the system will implement the permission adjustment; if not approved, the system will retain the original permission level and record the audit results. Finally, the audit results will be notified to visitors and relevant managers to ensure the transparency and timeliness of information.

[0085] Through the above detailed steps S121 to S123, after each information access, the visitor's operation log is collected and analyzed, their operation habits and access patterns are identified, the visitor's permission level is dynamically adjusted, and the adjustment results are sent to the administrator for review. This process not only improves the security and data protection level of the system through behavior monitoring and dynamic permission adjustment, but also ensures the rationality and necessity of permission adjustment through the review mechanism, effectively preventing misoperation and abuse of permissions.

[0086] In addition, the method includes: S124, retrieve the operation log and history log of the visitor, and detect whether the visitor has abnormal behavior based on the operation log and history log.

[0087] Among them, the visitor's current operation log and historical operation log are retrieved from the log management system, the current operation log and the historical operation log are integrated to form a complete time series data set, and the integrated log data is stored in the log management system to ensure the continuity and integrity of the data.

[0088] It should be pointed out here that the operation log content includes but is not limited to: Login Time: The time when the visitor logged into the system.

[0089] Visit frequency: the number of visits by a visitor within a certain period of time.

[0090] Type of data accessed: The type of data accessed by the visitor, such as medical records, test results, imaging materials, etc.

[0091] Operation type: The type of operation performed by the visitor, such as view, modify, delete, etc.

[0092] S125: If it exists, determine the corresponding abnormality log, and determine the corresponding abnormality level based on the abnormality log.

[0093] Among them, based on historical log data, a visitor behavior model is established. The behavior model can include normal behavior patterns and abnormal behavior patterns, which are modeled through machine learning algorithms (such as clustering and classification); the current operation log is compared with the behavior model to detect whether there is abnormal behavior, and the detected abnormal behavior is recorded, including abnormal time, abnormal type, abnormal operation, etc., to ensure the traceability of abnormal behavior. It should be pointed out here that abnormal behavior may include but is not limited to: Frequent access to sensitive data: Sensitive data is frequently accessed within a short period of time.

[0094] Non-normal time access: Accessing the system during non-working hours or non-normal time periods.

[0095] Abnormal operation type: Uncommon or high-risk operations are performed, such as deleting important data, modifying key information, etc.

[0096] S126, judging whether to lower the visitor's authority level or suspend the authority based on the abnormality level.

[0097] Among them, according to the abnormal level of abnormal behavior, the visitor's permission requirements and potential risks are evaluated. For example, if the abnormal level is level 1 (high risk), the visitor's permissions may need to be immediately suspended. Based on the evaluation results, it is decided whether to reduce the visitor's permission level or suspend his permissions. Permission adjustments may include: Reduce privilege levels: Reduce the access rights of visitors to restrict their access to sensitive data or high-risk operations.

[0098] Stop permissions: Completely stop the visitor's access rights to prevent potential security threats.

[0099] Based on executing the above steps S124 to S126, not only can potential security threats be discovered and handled in a timely manner through anomaly detection and risk assessment, but the security of the system and the protection of data can also be ensured through permission adjustment, providing comprehensive security protection for medical information management.

[0100] Finally, the method also includes: S131, regularly check the operating status of each node in the alliance chain network.

[0101] Among them, set the period of regular inspection, such as every hour, every minute, etc., to ensure the timeliness and frequency of the inspection. The specific inspection content is to check the operating status of each node, which includes but is not limited to: Online status: whether the node is online and whether it can respond to requests normally.

[0102] Resource utilization: the usage of the node's CPU, memory, disk and other resources.

[0103] Network connection status: The network connection status of the node, including bandwidth, latency, etc.

[0104] Node offline: Whether the node is offline and cannot communicate normally.

[0105] Response timeout: Whether the response time of the node exceeds the preset threshold.

[0106] Resource exhaustion: Whether the node's resources have been exhausted, such as memory overflow, insufficient disk space, etc.

[0107] S132: When a node failure is detected, the current load status of each node in the alliance chain network is obtained, and the corresponding target node is determined based on all current load conditions.

[0108] Among them, according to the inspection results, it is determined whether there is a node failure. If a node failure is detected, the fault handling process is immediately started, and the current load of all nodes in the alliance chain network is obtained through monitoring tools or API interfaces. By recording the current load of each node, the integrity and accuracy of the data are ensured.

[0109] It should be noted here that load conditions include but are not limited to: CPU Utilization: CPU usage of the node.

[0110] Memory Utilization: Memory usage of the node.

[0111] Disk Utilization: Disk usage of the node.

[0112] Network bandwidth: The network bandwidth usage of the node.

[0113] S133, transferring the corresponding tasks of the failed node to the target node.

[0114] Among them, according to the acquired load conditions, the load capacity of each node is evaluated, and based on the evaluation results, the node with the lowest load and sufficient resources is selected as the target node, and the selected target node information, including node ID, load conditions, etc., is recorded to ensure the traceability of the selection process. Multiple candidate nodes can also be selected. In the embodiment of the present application, the evaluation criteria may include: resource surplus, the remaining resources of the node (such as CPU, memory, disk space); network bandwidth, whether the network bandwidth of the node is sufficient; historical performance, the historical performance of the node, such as the speed and stability of processing tasks.

[0115] In addition, all tasks on the failed node, including ongoing tasks and pending tasks, are identified, and the identified tasks are migrated to the target node.

[0116] It should be noted here that task migration can include: Data synchronization: Synchronize task-related data to the target node to ensure data consistency and integrity.

[0117] State transfer: transfer the task status information (such as task progress, task results) to the target node to ensure the continuity of the task.

[0118] Task start: Start the migrated task on the target node to ensure the normal execution of the task.

[0119] Status update: Update the status information of the task, record the process and results of task migration, and ensure the traceability of task status.

[0120] Based on executing the above steps S131 to S133, not only can node failures be discovered and handled in a timely manner through regular inspection and fault detection, but also through load balancing and task migration, high availability of the system and continuity of data can be ensured, providing reliable technical support for medical information management.

[0121] The present application embodiment discloses a clinical information management related device, referring to Figure 2 ,include: The clinical information base generation module 200 builds a consortium chain network, retrieves unstructured clinical documents, extracts corresponding key information from the unstructured clinical documents, converts the key information into structured data, and uses knowledge graph technology to associate the structured data to generate a corresponding clinical information base, wherein each node in the consortium chain network represents a certified medical institution; A clinical decision support report generation module 210 performs multi-dimensional analysis on structured data based on a federated learning algorithm, generates a corresponding clinical decision support report based on the corresponding analysis result data, records the clinical decision support report in a clinical information database, executes a data sharing agreement on the clinical decision support report, and determines the authority mechanism corresponding to the clinical decision support report, wherein the clinical decision support report includes disease diagnosis, treatment recommendations, and prognosis assessment; The access judgment module 220 determines the user role of the visitor when accessing information, and determines the corresponding target access rights based on the user role. The target access rights and the permission mechanism are used to determine whether access is possible. If so, the corresponding target information in the clinical information library is retrieved and the target information is provided to the visitor in a graphical interface.

[0122] Furthermore, related equipment includes but is not limited to: The trigger module, when the medical institution corresponding to the node in the alliance chain network generates new clinical data, immediately notifies the clinical information database through the event trigger mechanism to trigger data update; The latest clinical data acquisition module determines the corresponding target medical institution, is used to obtain the latest clinical data corresponding to the target medical institution, and records the version information of the latest clinical data, wherein the version information includes update time, update content and updater information; The update module, during the data update process, if multiple nodes submit updates to the same data at the same time, uses a pre-set conflict resolution mechanism to perform the update.

[0123] Furthermore, related equipment includes but is not limited to: The medical history data retrieval module is used to retrieve the medical history data of the corresponding patient, where the medical history data includes disease diagnosis, treatment plan, medication records, laboratory test results, and imaging data; The prediction result acquisition module calls the pre-trained prediction model, inputs the structured data and medical history data into the prediction model, and is used to obtain the corresponding prediction results, wherein the prediction results include the probability of disease recurrence, the risk of complications, and the expected treatment effect; The prediction result recording module is used to record the prediction results in the clinical decision support report.

[0124] Furthermore, related equipment includes but is not limited to: A modification feedback data acquisition module sends the clinical decision support report to doctors or other medical professionals to obtain modification feedback data of the clinical decision support report from doctors or other medical professionals; A modification module is used to determine whether the clinical decision support report needs to be modified based on the modification feedback data, wherein the modification content is the diagnosis conclusion of the disease, the specific treatment recommendation, and the detailed description of the prognosis assessment in the clinical decision support report; The modification data recording module modifies the clinical decision support report if necessary, and is used to record the corresponding modification data and save the versions of the clinical decision support report before and after the modification, wherein the modification data includes the modification time, modification content and modifier information.

[0125] Furthermore, related equipment includes but is not limited to: An operation log collection module is used to collect the operation logs of visitors, wherein the operation logs include login time, access frequency, access data type, and operation type; The access pattern acquisition module is used to analyze the operation logs to obtain the corresponding operation habits and access patterns of visitors; The permission level adjustment module is used to adjust the visitor's permission level based on operating habits and access patterns, and send the adjusted permission level to the administrator for review.

[0126] Furthermore, related equipment includes but is not limited to: The historical log retrieval module is used to retrieve the visitor's operation log and historical log, and detect whether the visitor has abnormal behavior based on the operation log and historical log; An exception level determination module, if any, determines the corresponding exception log, and determines the corresponding exception level based on the exception log; The judgment module is used to judge whether to reduce the visitor's permission level or stop the permission based on the abnormal level.

[0127] Furthermore, related equipment includes but is not limited to: The operation status check module is used to regularly check the operation status of each node in the alliance chain network. The operation status includes online status, resource utilization, network connection status, node offline, response timeout, and resource exhaustion; The target node determination module, when a node failure is detected, obtains the current load status of each node in the alliance chain network, and determines the corresponding target node based on all current load conditions; The task transfer module is used to transfer the corresponding tasks of the failed node to the target node.

[0128] An embodiment of the present application further discloses a clinical information management method, including a processor, in which a program of the clinical information management method described above is run.

[0129] The embodiment of the present application also discloses a storage medium storing a program of the clinical information management method described above.

[0130] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A clinical information management method, characterized in that: include: Build a consortium chain network, retrieve unstructured clinical documents, extract corresponding key information from the unstructured clinical documents, convert the key information into structured data, and use knowledge graph technology to associate the structured data to generate a corresponding clinical information library, where each node in the consortium chain network represents a certified medical institution; Perform multi-dimensional analysis on structured data based on a federated learning algorithm, generate a corresponding clinical decision support report based on the corresponding analysis result data, record the clinical decision support report in the clinical information database, execute a data sharing agreement on the clinical decision support report, and determine the authority mechanism corresponding to the clinical decision support report, wherein the clinical decision support report includes disease diagnosis, treatment recommendations and prognosis assessment; When accessing information, the user role of the visitor is determined, and the corresponding target access rights are determined based on the user role. Based on the target access rights and the permission mechanism, it is determined whether access is possible. If so, the corresponding target information in the clinical information library is retrieved and provided to the visitor in a graphical interface.

2. The clinical information management method according to claim 1, characterized in that: The method also includes: When new clinical data is generated by the medical institution corresponding to the node in the alliance chain network, the clinical information database is immediately notified through the event trigger mechanism to trigger data update; Determine the corresponding target medical institution, obtain the latest clinical data corresponding to the target medical institution, and record the version information of the latest clinical data, wherein the version information includes update time, update content and updater information; During the data update process, if multiple nodes submit updates to the same data at the same time, the update is performed based on a pre-set conflict resolution mechanism.

3. The clinical information management method according to claim 1, characterized in that: In generating the clinical decision support report, the method further comprises: Retrieve the medical history data of the corresponding patient, including disease diagnosis, treatment plan, medication records, laboratory test results, and imaging data; Retrieving a pre-trained prediction model, inputting the structured data and the medical history data into the prediction model, and obtaining corresponding prediction results, wherein the prediction results include the probability of disease recurrence, the risk of complications, and the expected treatment effect; The prediction result is recorded in the clinical decision support report.

4. The clinical information management method according to claim 3, characterized in that: After generating the clinical decision support report, the method further includes: Sending the clinical decision support report to a doctor or other medical professional, and obtaining feedback data on modifications to the clinical decision support report from the doctor or other medical professional; Determining whether the modified feedback data needs to be modified in the clinical decision support report, wherein the modified content is the diagnosis conclusion of the disease, the specific treatment recommendation, and the detailed description of the prognosis assessment in the clinical decision support report; If necessary, the clinical decision support report is modified, and the corresponding modification data is recorded, and the versions of the clinical decision support report before and after the modification are saved, wherein the modification data includes the modification time, modification content and modifier information.

5. The clinical information management method according to claim 1, characterized in that: After each information access, the method further includes: Collect the visitor's operation log, where the operation log includes login time, access frequency, access data type, and operation type; Analyze the operation log to obtain the operation habits and access patterns of the visitor; The permission level of the visitor is adjusted based on the operating habits and the access mode, and the adjusted permission level is sent to the administrator for review.

6. The clinical information management method according to claim 5, characterized in that: The method also includes: Retrieving the operation log and history log of the visitor, and detecting whether the visitor has abnormal behavior based on the operation log and the history log; If so, determine the corresponding exception log, and determine the corresponding exception level based on the exception log; Determine whether to lower the visitor's permission level or suspend the permission based on the abnormality level.

7. The clinical information management method according to claim 1, characterized in that: The method also includes: Regularly check the operating status of each node in the alliance chain network, including online status, resource utilization, network connection, node offline, response timeout, and resource exhaustion; When a node failure is detected, the current load status of each node in the alliance chain network is obtained, and the corresponding target node is determined based on all current load conditions; The corresponding tasks of the failed node are transferred to the target node.

8. A clinical information management related device, characterized in that: include: A clinical information base generation module builds a consortium chain network, retrieves unstructured clinical documents, extracts corresponding key information from the unstructured clinical documents, converts the key information into structured data, and uses knowledge graph technology to associate the structured data to generate a corresponding clinical information base, wherein each node in the consortium chain network represents a certified medical institution; A clinical decision support report generation module, which performs multi-dimensional analysis on structured data based on a federated learning algorithm, generates a corresponding clinical decision support report based on the corresponding analysis result data, records the clinical decision support report in the clinical information database, executes a data sharing agreement on the clinical decision support report, and determines the authority mechanism corresponding to the clinical decision support report, wherein the clinical decision support report includes disease diagnosis, treatment recommendations and prognosis assessment; The access judgment module determines the user role of the visitor when accessing information, and determines the corresponding target access rights based on the user role. The target access rights and the permission mechanism are used to determine whether access is possible. If so, the corresponding target information in the clinical information library is retrieved and provided to the visitor in a graphical interface.

9. A clinical information management method, characterized in that: It comprises a processor, in which a program of the clinical information management method according to any one of claims 1 to 7 is running.

10. A storage medium, characterized in that: A program storing the clinical information management method according to any one of claims 1 to 7.

Images