Diagnostic Multidimensional Analysis Method, System, Device and Medium

Abstract

This application discloses a diagnostic multidimensional analysis method, system, device, and medium. By determining the mapping relationship between target medical information and evidence-based nodes, and evaluating the execution results of clinical pathways based on evidence-based evidence, this method helps reduce the influence of physicians' personal experience and subjective judgment on treatment plans, improving the scientific and rational nature of treatment decisions. For each diagnostic information in the target medical information, if a corresponding target clinical pathway exists, the execution status of the target clinical pathway for that diagnostic information can be evaluated based on the mapping relationship between the target medical information and evidence-based nodes, and the evidence-based evidence corresponding to at least one evidence-based node included in the target medical information and target clinical pathway. This helps to promptly identify and resolve potential problems during execution, optimize the execution process of clinical pathways, improve the quality and efficiency of medical services, and provide scientific, objective, and traceable evidence chain support for diagnostic selection.

Description

Technical Field

[0001] This application relates to the fields of medical diagnostic selection and deep learning technology, and in particular to a diagnostic multidimensional analysis method, system, device and medium. Background Technology

[0002] In the current healthcare environment, the writing of medical records and the selection of diagnoses have a crucial impact on patient treatment decisions and the allocation of medical resources. For hospital outpatient and inpatient medical records, different departments and doctors need to follow corresponding writing standards, including standardized fields for reporting, such as requirements for reporting the primary diagnosis and other diagnoses. These standards ensure the consistency and comparability of medical records, providing foundational data for subsequent diagnosis, treatment, and medical management.

[0003] However, in practice, due to the complexity and diversity of medical record content, the choice of diagnosis is often influenced by multiple factors. The diagnosis on the first page of the medical record typically focuses on reflecting the severity of the disease, while the medical insurance list pays more attention to the resolution of the patient's problems and the consumption of medical resources. This difference makes the selection of diagnosis codes and names more complex.

[0004] Traditional diagnostic selection methods often rely on doctors' experience and subjective judgment, lacking scientific and objective basis. Furthermore, with the deepening of healthcare payment reforms such as DRG (Diagnosis Related Groups) and DRP (Disease-Specific Reimbursement Rate), hospitals and doctors face increasing economic pressure. This pressure may lead clinical departments to favor high-weighted, high-scoring diagnoses to secure more medical insurance reimbursement, rather than truly reflecting the patient's actual condition and treatment needs.

[0005] To address this issue, existing technical methods typically rely on programming logic, OCR (Optical Character Recognition), and other technologies to retrieve medical record data from hospital information systems and extract the primary and other diagnoses according to predetermined rules. However, while this method can automatically extract diagnostic information from medical records, it still has shortcomings in terms of the scientific rigor and objectivity of diagnostic selection. In particular, when a doctor's subjective judgment or writing habits deviate from the rules, it can lead to biased diagnostic information, thereby affecting subsequent treatment decisions and the allocation of medical resources.

[0006] Therefore, there is an urgent need for a method that can scientifically and objectively analyze the rationality of a diagnosis, correlate the diagnosis with actual clinical treatment, and provide a traceable chain of evidence to support diagnostic choices. Summary of the Invention

[0007] This application provides a diagnostic multidimensional analysis method, system, device, and medium to address the shortcomings in the scientific rigor and objectivity of existing diagnostic options, and the fact that subsequent treatment decisions and medical resource allocation are easily influenced by physicians' subjective factors.

[0008] In a first aspect, this application provides a diagnostic multidimensional analysis method, the method comprising:

[0009] The acquired medical record data to be analyzed is preprocessed, and target medical information is extracted from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structure processing and terminology standardization, and the target medical information includes diagnostic information;

[0010] Determine the mapping relationship between the target medical information and the evidence-based nodes, and map the diagnostic information in the mapping relationship to the International Classification of Disease (ICD) codes;

[0011] For each diagnostic information in the target medical information, if a target clinical path corresponding to the diagnostic information is determined from the pre-configured clinical paths, the execution result of the target clinical path is determined based on the mapping relationship, the target medical information, and the evidence corresponding to at least one evidence-based node included in the target clinical path. The execution result includes one or more of the following: the type of clinical path used, whether the at least one evidence-based node was executed, the execution result corresponding to each of the at least one evidence-based node, the weight corresponding to each of the at least one evidence-based node, the data source path corresponding to each of the at least one evidence-based node, the node score corresponding to each of the at least one evidence-based node, and the final score of the target clinical path. If a target clinical path corresponding to the diagnostic information is determined not to exist from the pre-configured clinical paths, the clinical path for that diagnostic information is missing.

[0012] Based on the target medical information and the evidence-based results corresponding to each diagnostic information, evidence-based archive information is determined and saved; wherein, the evidence-based result of any diagnostic information is the execution result of that diagnostic information, or the clinical pathway of that diagnostic information is lost.

[0013] Secondly, this application also provides a diagnostic multidimensional analysis system, the system comprising:

[0014] The data extraction unit is used to preprocess the acquired medical record data to be analyzed and extract target medical information from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structure processing and terminology standardization, and the target medical information includes diagnostic information;

[0015] A determining unit is used to determine the mapping relationship between the target medical information and the evidence-based node, and to map the diagnostic information in the mapping relationship to the International Classification of Disease (ICD) code.

[0016] The processing unit is configured to, for each diagnostic information in the target medical information, if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information exists, then determine the execution result of the target clinical pathway based on the mapping relationship, the target medical information, and the evidence-based evidence corresponding to at least one evidence-based node included in the target clinical pathway; wherein, the execution result includes one or more of the following: the type of clinical pathway used, whether the at least one evidence-based node is executed, the execution result corresponding to the at least one evidence-based node, the weight corresponding to the at least one evidence-based node, the data source path corresponding to the at least one evidence-based node, the node score corresponding to the at least one evidence-based node, and the final score of the target clinical pathway; if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information does not exist, then determine that the clinical pathway for the diagnostic information is missing;

[0017] The storage unit is used to determine and save evidence-based archive information based on the target medical information and the evidence-based results corresponding to each diagnostic information; wherein, the evidence-based result of any diagnostic information is the execution result of the diagnostic information, or the clinical pathway of the diagnostic information is lost.

[0018] Thirdly, this application provides a computer device including a processor, which executes a computer program stored in a memory to implement the steps of the diagnostic multidimensional analysis method described above.

[0019] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the diagnostic multidimensional analysis method described above.

[0020] The beneficial effects of this application are as follows:

[0021] 1. By preprocessing the medical record data to be analyzed, including structuring and terminology standardization, this process ensures that medical record data from different sources and in different formats can be processed and analyzed in a uniform and standardized manner, avoiding the influence of doctors' writing habits on the results of the analysis of the medical record data.

[0022] 2. By determining the mapping relationship between target medical information and evidence-based nodes, and by evaluating the implementation results of clinical pathways based on evidence, this approach helps to reduce the influence of physicians' personal experience and subjective judgment on treatment plans, and improves the scientificity and rationality of treatment decisions.

[0023] 3. For each diagnostic information in the target medical information, if a corresponding target clinical pathway exists, the implementation status of the target clinical pathway for that diagnostic information can be evaluated based on the mapping relationship between the target medical information and the evidence-based nodes, as well as the evidence-based evidence corresponding to at least one evidence-based node contained in the target medical information and the target clinical pathway. This helps to identify and resolve potential problems in the implementation process in a timely manner, optimize the implementation process of the clinical pathway, improve the quality and efficiency of medical services, and provide scientific, objective, and traceable evidence chain support for diagnostic selection.

[0024] 4. Evidence-based clinical pathway management can more accurately assess the cost-effectiveness of different treatment options, thereby helping hospitals and policymakers allocate medical resources more rationally. At the same time, by reducing unnecessary medical interventions and waste, it can also reduce the medical burden on patients and improve the overall efficiency of healthcare services.

[0025] 5. By recording and preserving evidence-based archive information, including the execution results of clinical pathways, the weights and scores of evidence-based nodes, and the data source paths, this process enhances the transparency of medical decisions, making the medical process traceable and auditable. This not only helps improve the quality of medical care but also provides valuable data support for subsequent medical research and quality improvement. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of a diagnostic multidimensional analysis process provided in an embodiment of this application;

[0028] Figure 2 A schematic diagram of the structure of a functional module provided in an embodiment of this application;

[0029] Figure 3 This is a schematic diagram of the structure of a diagnostic multidimensional analysis system provided in an embodiment of this application;

[0030] Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0032] To improve the scientific rigor and objectivity of diagnostic selection and avoid the influence of doctors' subjective factors on subsequent treatment decisions and medical resource allocation, this application provides a method, system, equipment, and medium for multidimensional diagnostic analysis.

[0033] Example 1:

[0034] This application provides a diagnostic multidimensional analysis method. Figure 1 A schematic diagram of a diagnostic multidimensional analysis process provided in this application embodiment, the process including:

[0035] S101: Preprocess the acquired medical record data to be analyzed, and extract target medical information from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structured processing and terminology standardization, and the target medical information includes diagnostic information.

[0036] In this application, the diagnostic multidimensional analysis method is applied to computer equipment, which can be a smart terminal, such as a computer or a robot, or a server, such as an application server or a business server.

[0037] Medical record data in healthcare systems is typically stored across various data sources, such as hospital information systems and electronic medical record systems. To integrate medical record data from different data sources and improve the efficiency and accuracy of medical data analysis, thereby supporting better clinical decision-making, medical research, or public health policy formulation, this application describes a computer device used for diagnostic multidimensional analysis that can interface with different data sources to acquire the medical record data to be analyzed, enabling subsequent medical data analysis based on this data. For example, raw medical record data (denoted as the medical record data to be analyzed) can be obtained from hospital information systems, electronic medical record systems, or other medical data warehouses through a data extraction module. This medical record data to be analyzed may include the patient's personal information, medical history, examination results, medical orders, diagnoses, etc.

[0038] Generally, the content of medical record data to be analyzed is filled in based on the doctor's subjective judgment and writing habits, which may lead to problems such as inconsistent formatting and non-standard terminology, thus affecting the analysis of the medical record data. Therefore, this application proposes preprocessing the acquired medical record data. This preprocessing includes structuring and terminology standardization. For structuring, unstructured text medical record data can be converted into structured or semi-structured data formats. For example, key information fields in the medical record data to be analyzed, such as patient basic information, chief complaint, present illness, past medical history, physical examination, auxiliary examinations, diagnosis, and treatment, are identified and organized according to a predetermined data structure to facilitate more efficient subsequent data parsing and processing. For terminology standardization, medical terms in the medical record data to be analyzed are standardized to ensure that the same concept is expressed consistently across different medical record data, reducing misunderstandings caused by inconsistent terminology usage and improving the accuracy of data analysis. For example, this can be achieved by establishing a medical terminology database and using natural language processing technology for terminology recognition and mapping.

[0039] Optionally, the preprocessing may also include data cleaning, such as removing noisy data (e.g., incorrectly entered text, irrelevant blank lines, etc.) and handling missing values ​​(e.g., filling with the mean, median, or directly deleting records containing missing values).

[0040] For large-scale medical record data to be analyzed, directly analyzing the entire dataset can consume significant computational resources and time. Furthermore, the dataset may contain a large amount of redundant information, interfering with diagnostic information interpretation. Therefore, this application proposes extracting target medical information from preprocessed medical record data to significantly improve system processing speed and responsiveness by reducing the amount of data to be processed, thus optimizing user experience. For example, for free text, natural language processing techniques such as named entity recognition and semantic understanding models can be used, with the data extraction module identifying and extracting target medical information from the preprocessed medical record data. For structured data, the data extraction module can directly extract relevant information from the preprocessed medical record data through SQL queries or other programming methods. This target diagnostic information includes at least diagnostic information, which will be used for subsequent analysis.

[0041] S102: Determine the mapping relationship between the target medical information and the evidence-based nodes, and map the diagnostic information in the mapping relationship to the International Classification of Diseases (ICD) code.

[0042] To ensure the objectivity and scientific rigor of diagnostic selection, this application pre-configures at least one clinical pathway corresponding to each diagnostic piece of information, guiding diagnostic selection through this pathway. Each clinical pathway includes all evidence-based nodes associated with that diagnostic information, along with the corresponding evidence-based evidence for each node. For example, the evidence-based knowledge module of the computer device constructs a dynamic knowledge base by integrating and updating clinical guidelines, evidence-based medicine literature, and best practices. This knowledge base not only includes standardized clinical pathways but also records the evidence-based basis for each node, ensuring that all diagnostic and treatment choices have clear scientific grounds.

[0043] In one possible implementation, the clinical pathway includes a publicly available latest clinical pathway that integrates the latest clinical practice guidelines and evidence-based medicine research findings and records the corresponding versions, making it a publicly recognized latest clinical pathway.

[0044] To protect the security of publicly available clinical pathways, this application may prohibit manual modification and updating of data in publicly available clinical pathways.

[0045] In another possible implementation, the evidence-based knowledge module allows hospitals to personalize their clinical pathways, with evidence-based nodes within these pathways supporting import and modification. For example, it supports the import and modification of key evidence-based nodes such as initial diagnosis and screening (medical history taking, physical examination, preliminary diagnosis), diagnosis (laboratory tests, imaging examinations, diagnostic confirmation), treatment (drug therapy, surgery, rehabilitation therapy), and follow-up and management (regular check-ups, treatment plan adjustments, long-term management). In other words, the clinical pathways maintained in this evidence-based knowledge module also include custom-imported clinical pathways, which support co-creation and maintenance, with versions reviewed and maintained by professionals.

[0046] In one example, the at least one evidence-based node includes diagnostic criteria, rehabilitation assessment and plan, treatment plan, standard length of hospital stay, examination items during hospitalization, and resource nodes. The diagnostic criteria are the scientific basis for determining the patient's disease or health status, typically based on a comprehensive analysis of multiple methods such as clinical manifestations, laboratory tests, and imaging examinations, to ensure the accuracy and timeliness of the diagnosis, providing a solid foundation for subsequent treatment and avoiding misdiagnosis and mistreatment. The rehabilitation assessment and plan is a process of comprehensively evaluating the patient's functional status, quality of life, and psychological state at different stages after treatment or surgery, and developing personalized rehabilitation plans tailored to the patient's rehabilitation needs. These plans include rehabilitation training, lifestyle guidance, and psychological support, helping doctors and rehabilitation therapists understand the patient's rehabilitation progress and existing problems, develop or adjust personalized rehabilitation plans, allocate medical resources, and improve rehabilitation outcomes. The treatment plan is a treatment plan developed based on the diagnostic results and the patient's specific condition, including multiple methods such as drug therapy, surgical treatment, and physical therapy, to ensure the targeted and effective nature of treatment measures, reduce uncertainty and risks during treatment, improve treatment outcomes, and shorten the course of illness. This standard length of stay is a reasonable range of hospitalization time set for specific diseases or treatment procedures to optimize the allocation of medical resources, avoid over- or under-treatment, reduce patient hospitalization costs and time, and improve hospital bed turnover rate. During hospitalization, necessary examinations are conducted to monitor the patient's condition, assess treatment effectiveness, or detect potential complications, enabling timely identification and management of problems, adjustment of treatment plans, ensuring the safety and effectiveness of treatment, and improving patient satisfaction and quality of care. This resource node refers to the various resources involved in the medical service process, including human resources, material resources, and information resources, to avoid excessive examinations and over-standard consumption.

[0047] To facilitate the analysis of target medical information in conjunction with clinical pathways, this application, after obtaining the target medical information based on the above embodiments, can determine the mapping relationship between the target medical information and evidence-based nodes, so as to map the collected medical information to specific evidence-based nodes in the clinical pathway. This mapping relationship not only helps to quickly locate the patient's current stage of diagnosis and treatment, but also allows analysis of whether the diagnosis and treatment process conforms to the requirements of the clinical pathway, and whether there are any variations or deficiencies. For example, after obtaining the target medical information, the data extraction module can use a preset algorithm or rule engine to match and map this target medical information with evidence-based nodes in the clinical pathway.

[0048] In one possible implementation, when configuring the data required for the medical record data to be analyzed and the target medical information to be extracted, it can be determined according to the type of evidence-based node to ensure that the mapping relationship between the target medical information and the evidence-based node can be accurately determined subsequently.

[0049] For example, if the evidence-based nodes include diagnostic criteria, rehabilitation assessment and plan, treatment plan, standard length of stay, resource nodes, examination items during hospitalization, and resource nodes, then the medical record data to be analyzed includes medical history collection and initial diagnosis records, examination results, treatment records, rehabilitation records, inventory records, and discharge records; the target medical information includes medical history information in the medical history collection and initial diagnosis records, laboratory and imaging examination results in the examination results, drug treatment, surgical records, and other treatment measures in the treatment records, the time, content, and operation methods of each stage in the rehabilitation records, the three-category classification names and quantities in the inventory records, and the discharge time and diagnosis in the discharge records; wherein, the three-category classifications include drugs, diagnosis and treatment, and consumables. "Medical history information in medical history collection and initial diagnosis records" is used to match the evidence-based node "diagnostic basis"; "time, content, and operation methods of each stage in rehabilitation records" is used to match the evidence-based node "rehabilitation assessment and plan"; "drug treatment, surgical records, and other treatment measures in treatment records" is used to match "treatment plan"; "results of laboratory tests and imaging examinations in examination results" is used to match the evidence-based node "in-hospital examination items"; "three-item type summary name and quantity in checklist records" is used to match the evidence-based node "resource node"; "discharge time and diagnosis in discharge records" is used to match the evidence-based node "standard length of stay".

[0050] To facilitate subsequent matching, in this application, after obtaining the mapping relationship, the diagnostic information in the mapping relationship can be mapped to the International Classification of Diseases (ICD) code. Mapping diagnostic information to ICD codes is a universally accepted method for disease classification, providing a unified classification standard for diagnostic information. By mapping diagnostic information to ICD codes, comparability and interoperability of diagnostic information between different hospitals and regions can be achieved, while also facilitating analysis based on ICD codes.

[0051] S103: For each diagnostic information in the target medical information, if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information exists, then the execution result of the target clinical pathway is determined according to the mapping relationship, the target medical information, and the evidence corresponding to at least one evidence-based node included in the target clinical pathway; wherein, the execution result includes one or more of the following: the type of clinical pathway used, whether the at least one evidence-based node is executed, the execution result corresponding to the at least one evidence-based node, the weight corresponding to the at least one evidence-based node, the data source path corresponding to the at least one evidence-based node, the node score corresponding to the at least one evidence-based node, and the final score of the target clinical pathway; if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information does not exist, then it is determined that the clinical pathway for the diagnostic information is missing.

[0052] After obtaining the mapping relationship based on the above embodiments, the target medical information, mapping relationship, and pre-configured clinical pathways can be used to provide profiling support for the diagnostic conclusions of clinical departments, thereby standardizing hospital-side diagnosis and treatment behavior and medical resource consumption. For example, the target medical information of a patient, particularly the diagnostic information, is analyzed. This diagnostic information is then used to match various pre-configured clinical pathways. Then, for each diagnostic information in the target medical information, it is matched with the standard diagnostic information corresponding to each clinical pathway. If a matching standard diagnostic information is found, it indicates that a target clinical pathway corresponding to this diagnostic information exists among the pre-configured clinical pathways, and the clinical pathway corresponding to this standard diagnostic information is determined as the target clinical pathway; if no matching standard diagnostic information is found, it is determined that a target clinical pathway corresponding to this diagnostic information does not exist among the pre-configured clinical pathways.

[0053] If a target clinical pathway corresponding to the diagnostic information is determined from the pre-configured clinical pathways, then medical decisions are further evaluated and executed based on the evidence-based nodes included in the target clinical pathway. For example, for at least one evidence-based node in the target clinical pathway, according to the mapping relationship determined in the above embodiments, key data corresponding to the evidence-based node is determined from the target medical data, and the node execution result corresponding to the evidence-based node is determined based on the evidence-based evidence and key data. The node execution result includes whether the evidence-based node was executed, the execution result of the evidence-based node, the data source path for the evidence-based node, and the node score of the evidence-based node. Then, based on the node execution results corresponding to the at least one evidence-based node, the execution result of the target clinical pathway is determined. The execution result includes one or more of the following: the type of clinical pathway used, whether the at least one evidence-based node was executed, the execution results corresponding to the at least one evidence-based node, the weights corresponding to the at least one evidence-based node, the data source paths corresponding to the at least one evidence-based node, the node scores corresponding to the at least one evidence-based node, and the final score of the target clinical pathway. The clinical pathway types used include the latest public clinical pathways and custom imported clinical pathways. The weights corresponding to the at least one evidence-based node can be pre-configured according to the importance of each evidence-based node in the target clinical pathway.

[0054] If a target clinical pathway corresponding to the diagnostic information is found among the pre-configured clinical pathways, then the clinical pathway for the diagnostic information is determined to be missing.

[0055] In one possible implementation, the final score is determined based on the weights corresponding to the at least one evidence-based node, the node scores corresponding to the at least one evidence-based node, and the number of the at least one evidence-based node. For example, for at least one evidence-based node included in the target clinical pathway, the product of the node score and its corresponding weight for each evidence-based node is calculated. These products are then summed to obtain the total product of all evidence-based nodes. Finally, the final score is determined by dividing this sum by the total number of evidence-based nodes (i.e., the number of at least one evidence-based node).

[0056] For example, the formula for the final score can be expressed as follows:

[0057]

[0058] Where i is the i-th evidence-based node, Factor i Score is the weight of the i-th evidence-based node. iLet be the node score of the i-th evidence-based node, EvidenceNodeCount be the total number of evidence-based nodes in the target clinical pathway, and ClinicEvidenceChainScore be the final score.

[0059] In one possible implementation, if the pre-configured clinical pathways include both publicly available latest clinical pathways and custom-imported clinical pathways, the publicly available latest clinical pathways, representing best practices, will be given priority when determining the target clinical pathway; that is, the matching priority of publicly available latest clinical pathways is higher than that of custom-imported clinical pathways. This approach ensures that patients receive treatment plans based on the latest research and consensus.

[0060] In one example, after obtaining the target medical information, and before determining the target clinical pathway corresponding to each diagnostic information in the target medical information from the pre-configured clinical pathways, the method further includes:

[0061] The diagnostic information contained in the target medical information is merged and deduplicated.

[0062] To ensure the accuracy and efficiency of subsequent clinical pathway matching, after extracting the target medical information, further merging and deduplication processing can be performed on the diagnostic information contained within the target medical information. This merging process primarily targets and integrates diagnostic information that may have similar or repetitive expressions. Since medical information may originate from different systems or doctors, they may use slightly different terminology to describe the same diagnosis. Therefore, the purpose of merging is to unify these substantially identical but differently expressed diagnostic information into a standardized entry. For example, "acute bronchitis" and "bronchitis, acute" may differ in expression, but they essentially refer to the same disease. During the merging process, the system will identify these similar entries and merge them into a unified diagnostic information. For example, the merging process can be implemented based on a pre-set medical terminology database or algorithm that can identify semantic similarity or equivalence between terms. Deduplication is the process of deleting potentially duplicate diagnostic information from the target medical information. Due to various reasons (such as system errors, human error, etc.), the same diagnostic information may be recorded multiple times in the medical records of the same patient. Without deduplication, duplicate information not only increases the complexity and time cost of subsequent processing but may also mislead the selection of clinical pathways. For example, deduplication can be achieved by comparing the unique identifiers (such as ICD-10 codes) or the content itself of each diagnostic entry. Diagnostic entries with the same unique identifier or identical content are treated as duplicates and deleted, leaving only one entry.

[0063] S104: Based on the target medical information and the evidence-based results corresponding to each diagnostic information, determine and save the evidence-based archive information; wherein, the evidence-based result of any diagnostic information is the execution result of the diagnostic information, or the clinical pathway of the diagnostic information is lost.

[0064] Based on the above embodiments, evidence-based archive information can be determined and saved according to the evidence-based results corresponding to the target medical information and each diagnostic information, thereby improving the quality, efficiency, and safety of medical services. For example, the evidence-based results corresponding to the target medical information and each diagnostic information can be saved according to pre-configured format requirements. Another example is that, based on pre-configured archive data types, the basic data for each archive data type can be determined from the target medical information, and then the basic data and the evidence-based results corresponding to each diagnostic information can be saved according to pre-configured format requirements.

[0065] In one possible implementation, the method further includes:

[0066] For each diagnostic information in the target medical information, if the execution result corresponding to the diagnostic information is stored, and the final score corresponding to the diagnostic information is lower than the pre-configured score threshold, then it is determined that the diagnostic information has a low correlation with the target clinical pathway corresponding to the diagnostic information.

[0067] For diagnostic information that already has execution results, the correlation between the diagnostic information and the target clinical pathway can be evaluated, thereby helping healthcare professionals to better understand and optimize clinical decision-making processes.

[0068] Specifically, for each diagnostic piece of information within the target medical information, if the corresponding execution result is stored, the final score of that diagnostic piece of information can be further analyzed. This final score is then compared to a pre-configured score threshold. This score threshold, set based on clinical experience and / or statistical data, is used to determine the extent to which a diagnostic piece of information conforms to or deviates from the expected clinical pathway effect. If the final score of a diagnostic piece of information is lower than this score threshold, it can be determined that the diagnostic piece of information has a low correlation with the target clinical pathway. In this way, it can help the medical team identify diagnostic pieces of information that perform poorly in actual application, and then analyze the reasons, which may be problems with the diagnosis itself, deviations in the execution process, or individual patient differences. Based on such analysis, the medical team can adjust the clinical pathway, optimize diagnostic strategies, or develop personalized treatment plans for specific patients to improve overall medical quality and patient satisfaction.

[0069] In one example, the medical record data to be analyzed includes medical record data for which evidence-based archiving information has not been established, and medical record data for which evidence-based archiving information has been established and an appeal request has been received. In this application, for medical record data for which diagnostic multidimensional analysis has not been performed, i.e., medical record data for which evidence-based archiving information has not been established, the diagnostic multidimensional analysis method provided in this application can be used for processing. Of course, for medical record data for which diagnostic multidimensional analysis has been performed, i.e., medical record data for which evidence-based archiving information has been established, the diagnostic multidimensional analysis method provided in this application can be used again for processing upon appeal from the hospital. For example, if a hospital has an objection to a certain medical record data for which evidence-based archiving information has already been established, such as an input error in the content of a certain medical record data, it can input an appeal request through the appeal configuration interface provided by the computer device. This appeal request can carry the modified data. After receiving the appeal request through the appeal configuration interface, the diagnostic multidimensional analysis method provided in this application can be used again for processing.

[0070] The beneficial effects of this application are as follows:

[0071] 1. By preprocessing the medical record data to be analyzed, including structuring and terminology standardization, this process ensures that medical record data from different sources and in different formats can be processed and analyzed in a uniform and standardized manner, avoiding the influence of doctors' writing habits on the results of the analysis of the medical record data.

[0072] 2. By determining the mapping relationship between target medical information and evidence-based nodes, and by evaluating the implementation results of clinical pathways based on evidence, this approach helps to reduce the influence of physicians' personal experience and subjective judgment on treatment plans, and improves the scientificity and rationality of treatment decisions.

[0073] 3. For each diagnostic information in the target medical information, if a corresponding target clinical pathway exists, the implementation status of the target clinical pathway for that diagnostic information can be evaluated based on the mapping relationship between the target medical information and the evidence-based nodes, as well as the evidence-based evidence corresponding to at least one evidence-based node contained in the target medical information and the target clinical pathway. This helps to identify and resolve potential problems in the implementation process in a timely manner, optimize the implementation process of the clinical pathway, improve the quality and efficiency of medical services, and provide scientific, objective, and traceable evidence chain support for diagnostic selection.

[0074] 4. Evidence-based clinical pathway management can more accurately assess the cost-effectiveness of different treatment options, thereby helping hospitals and policymakers allocate medical resources more rationally. At the same time, by reducing unnecessary medical interventions and waste, it can also reduce the medical burden on patients and improve the overall efficiency of healthcare services.

[0075] 5. By recording and preserving evidence-based archive information, including the execution results of clinical pathways, the weights and scores of evidence-based nodes, and the data source paths, this process enhances the transparency of medical decisions, making the medical process traceable and auditable. This not only helps improve the quality of medical care but also provides valuable data support for subsequent medical research and quality improvement.

[0076] Example 2:

[0077] For ease of explanation, based on the functions included in this diagnostic multidimensional analysis method, the computer equipment is divided into an evidence-based knowledge module, a data extraction module, an evidence-based matching module, an evidence-based fusion module, and an evidence-based storage module. Figure 2 This is a schematic diagram of a functional module provided in an embodiment of this application. The diagnostic multidimensional analysis method provided in this application will be described below through specific embodiments and in conjunction with each functional module:

[0078] Assuming the evidence-based knowledge module currently only supports the clinical pathway for "limb fractures," this clinical pathway is represented in the table below:

[0079]

[0080]

[0081] The above clinical pathways are imported and the final computer-executable evidence-based nodes are as follows:

[0082]

[0083]

[0084]

[0085] The data extraction module obtains the medical record data of Patient A to be analyzed, such as: "Patient A (hospitalization record number PN0001) chief complaint: accidentally fell, suffered severe pain in the lower leg, and was unable to stand upright. The diagnosis on the first page of the medical record is 1. Limb fracture 2. Hemiplegia. The diagnosis on the checklist is 1. Limb fracture, and the patient is recovering well after surgery."

[0086] The data extraction module preprocesses the medical record data to be analyzed and extracts the target medical data from the preprocessed medical record data.

[0087] Then, the data extraction module uses a large natural language processing model to determine the mapping relationship between the target medical information and evidence-based nodes based on the input data containing the target medical data.

[0088] For example, the input data containing the target medical data is as follows:

[0089] Given the following medical record: Patient A, Case Summary: ******, List: ****. Please extract the diagnostic set mapping relationship based on the medical record and return it as an array in the following format:

[0090]

[0091] Then, by matching text similarity with ICD-10, the final mapping relationship is obtained:

[0092]

[0093] The evidence-based matching module can merge and deduplicate the diagnostic information contained in the target medical information. Then, for each diagnostic information, it determines whether a corresponding target clinical path exists from the pre-configured clinical paths according to the pre-configured clinical path matching priority. The matching priority of public latest clinical paths is higher than that of custom imported clinical paths. In this example, the target medical information includes two diagnostic information: "limb fracture" and "hemiplegia". For "limb fracture", it first performs matching with public latest clinical paths. If a matching target clinical path is found, further analysis is performed based on the data recorded in that target clinical path. For "hemiplegia", no matching clinical path was found, so the target clinical path is set to empty, as there is currently no evidence-based basis. It can be executed independently after a matching clinical path is imported later.

[0094] For cases involving multiple patients, the data is also assembled into the following array format.

[0095]

[0096] After receiving notification from the evidence-based matching module that a target clinical pathway corresponds to a "limb fracture," the evidence-based fusion module can retrieve data for that target clinical pathway from the evidence-based knowledge module and the target medical data from the data extraction module. Then, it performs the following operations:

[0097] a. Evidence-based node matching

[0098] Match the diagnostic criteria in the clinical pathway with the medical history collection and initial consultation record.

[0099] Match the "Rehabilitation Assessment and Plan" with the "Rehabilitation Record" in the clinical pathway.

[0100] Match the treatment plan with the treatment record in the clinical pathway.

[0101] Match the standard length of stay with the discharge record in the clinical pathway.

[0102] Match the "Inpatient Examination Items" with the "Examination Results" in the clinical pathway.

[0103] b. Difference Analysis

[0104] For each evidence-based node, based on the corresponding evidence and key data, determine the execution result of that node, and identify and record any discrepancies between the medical record content and the clinical pathway. These discrepancies may be due to individual patient differences, complications, or other special circumstances.

[0105] After the matching operation, the following result is produced:

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112] The evidence-based fusion module determines the execution result of the target clinical pathway based on the execution results of the nodes corresponding to at least one evidence-based node. Based on the target medical information and the evidence-based results corresponding to each diagnostic information, the module determines the evidence-based archive information and saves it in the evidence-based storage module.

[0113] Example 3:

[0114] Based on the same inventive concept, this application also provides a diagnostic multidimensional analysis system. Figure 3 This application provides a schematic diagram of the structure of a diagnostic multidimensional analysis system, the system comprising:

[0115] The data extraction unit 31 is used to preprocess the acquired medical record data to be analyzed and extract target medical information from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structured processing and terminology standardization, and the target medical information includes diagnostic information;

[0116] The determining unit 32 is used to determine the mapping relationship between the target medical information and the evidence-based node, and to map the diagnostic information in the mapping relationship to the International Classification of Diseases (ICD) code.

[0117] Processing unit 33 is configured to, for each diagnostic information in the target medical information, if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information exists, then determine the execution result of the target clinical pathway based on the mapping relationship, the target medical information, and the evidence-based evidence corresponding to at least one evidence-based node included in the target clinical pathway; wherein, the execution result includes one or more of the following: the type of clinical pathway used, whether the at least one evidence-based node is executed, the execution result corresponding to the at least one evidence-based node, the weight corresponding to the at least one evidence-based node, the data source path corresponding to the at least one evidence-based node, the node score corresponding to the at least one evidence-based node, and the final score of the target clinical pathway; if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information does not exist, then determine that the clinical pathway for the diagnostic information is missing;

[0118] Storage unit 34 is used to determine and save evidence-based archive information based on the target medical information and the evidence-based results corresponding to each diagnostic information; wherein, the evidence-based result of any diagnostic information is the execution result of the diagnostic information, or the clinical pathway of the diagnostic information is lost.

[0119] The diagnostic multidimensional analysis system in this embodiment is presented in the form of functional modules. Here, a module refers to an application-specific integrated circuit (ASIC), a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0120] Further functional descriptions of the above-mentioned units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0121] Example 4:

[0122] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of this application, such as... Figure 4As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output system (such as a display device coupled to the interface). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 10 as an example.

[0123] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0124] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0125] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0126] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0127] The computer device also includes an input system 30 and an output system 40. The processor 10, memory 20, input system 30, and output system 40 can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0128] The input system 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touchscreen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. The output system 40 may include a display device, an auxiliary lighting system (e.g., LEDs), and a haptic feedback system (e.g., a vibration motor). The aforementioned display devices include, but are not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0129] Example 5:

[0130] Based on the above embodiments, this application also provides a computer-readable storage medium storing a computer program executable by a processor. When the program runs on the processor, it causes the processor to perform the following steps:

[0131] The acquired medical record data to be analyzed is preprocessed, and target medical information is extracted from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structure processing and terminology standardization, and the target medical information includes diagnostic information;

[0132] Determine the mapping relationship between the target medical information and the evidence-based nodes, and map the diagnostic information in the mapping relationship to the International Classification of Disease (ICD) codes;

[0133] For each diagnostic information in the target medical information, if a target clinical path corresponding to the diagnostic information is determined from the pre-configured clinical paths, the execution result of the target clinical path is determined based on the mapping relationship, the target medical information, and the evidence corresponding to at least one evidence-based node included in the target clinical path. The execution result includes one or more of the following: the type of clinical path used, whether the at least one evidence-based node was executed, the execution result corresponding to each of the at least one evidence-based node, the weight corresponding to each of the at least one evidence-based node, the data source path corresponding to each of the at least one evidence-based node, the node score corresponding to each of the at least one evidence-based node, and the final score of the target clinical path. If a target clinical path corresponding to the diagnostic information is determined not to exist from the pre-configured clinical paths, the clinical path for that diagnostic information is missing.

[0134] Based on the target medical information and the evidence-based results corresponding to each diagnostic information, evidence-based archive information is determined and saved; wherein, the evidence-based result of any diagnostic information is the execution result of that diagnostic information, or the clinical pathway of that diagnostic information is lost.

[0135] Since the principle of the computer-readable storage medium in solving the problem is similar to that of the diagnostic multidimensional analysis method, the implementation of the computer-readable storage medium can be found in Examples 1-2 of the method, and the repeated parts will not be described again.

[0136] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A diagnostic multidimensional analysis method, characterized in that, The method includes: The acquired medical record data to be analyzed is preprocessed, and target medical information is extracted from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structure processing and terminology standardization, and the target medical information includes diagnostic information; Determine the mapping relationship between the target medical information and the evidence-based nodes, and map the diagnostic information in the mapping relationship to the International Classification of Disease (ICD) codes; For each diagnostic information in the target medical information, if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information exists, then the execution result of the target clinical pathway is determined based on the evidence-based evidence corresponding to the mapping relationship, the evidence-based evidence corresponding to the target medical information, and the evidence-based evidence of at least one evidence-based node included in the target clinical pathway. The execution result includes: the type of clinical pathway used, whether the at least one evidence-based node was executed, the execution result corresponding to the at least one evidence-based node, the weights corresponding to the at least one evidence-based node, the data source paths corresponding to the at least one evidence-based node, the node scores corresponding to the at least one evidence-based node, and the final score of the target clinical pathway. If it is determined from the pre-configured clinical pathways that no target clinical pathway corresponding to the diagnostic information exists, then the clinical pathway for that diagnostic information is missing. The final score is determined based on the weights corresponding to the at least one evidence-based node, the node scores corresponding to the at least one evidence-based node, and the number of the at least one evidence-based node. Based on the target medical information and the evidence-based results corresponding to each diagnostic information, evidence-based archive information is determined and saved; wherein, the evidence-based result of any diagnostic information is the execution result of that diagnostic information, or the clinical pathway of that diagnostic information is lost.

2. The method as described in claim 1, characterized in that, The at least one evidence-based node includes diagnostic criteria, rehabilitation assessment and plan, treatment plan, standard length of stay, examination items during hospitalization, and resource nodes.

3. The method as described in claim 2, characterized in that, The medical record data to be analyzed includes medical history collection and initial diagnosis records, examination results, treatment records, rehabilitation records, inventory records, and discharge records; The target medical information includes medical history information in the medical history collection and initial diagnosis record, laboratory test and imaging test results in the examination results, drug treatment, surgical records and other treatment measures in the treatment record, time, content and operation methods of each stage in the rehabilitation record, the three-category classification name and quantity in the list record, and discharge time and diagnosis in the discharge record; wherein, the three-category classification includes drugs, diagnosis and treatment and consumables.

4. The method as described in claim 1, characterized in that, After obtaining each diagnostic information from the target medical information, and before determining the target clinical pathway corresponding to each diagnostic information from the pre-configured clinical pathways, the method further includes: The diagnostic information contained in the target medical information is merged and deduplicated.

5. The method as described in claim 1, characterized in that, The pre-configured clinical pathways include public latest clinical pathways and custom imported clinical pathways, with the public latest clinical pathways having a higher matching priority than the custom imported clinical pathways.

6. The method as described in claim 1, characterized in that, The method further includes: For each diagnostic information in the target medical information, if the execution result corresponding to the diagnostic information is stored, and the final score corresponding to the diagnostic information is lower than the pre-configured score threshold, then it is determined that the diagnostic information has a low correlation with the target clinical pathway corresponding to the diagnostic information.

7. The method as described in claim 1, characterized in that, The medical record data to be analyzed includes medical record data for which evidence-based archiving information has not been established, as well as medical record data for which evidence-based archiving information has been established and for which appeal requests have been received.

8. A diagnostic multidimensional analysis system, characterized in that, The system includes: The data extraction unit is used to preprocess the acquired medical record data to be analyzed and extract target medical information from the preprocessed medical record data to be analyzed; wherein, the preprocessing includes structure processing and terminology standardization, and the target medical information includes diagnostic information; A determining unit is used to determine the mapping relationship between the target medical information and the evidence-based node, and to map the diagnostic information in the mapping relationship to the International Classification of Disease (ICD) code. The processing unit is configured to, for each diagnostic information in the target medical information, if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information exists, then determine the execution result of the target clinical pathway based on the evidence-based evidence corresponding to the mapping relationship, the evidence-based evidence corresponding to the target medical information, and the evidence-based evidence of at least one evidence-based node included in the target clinical pathway; wherein, the execution result includes: the type of clinical pathway used, whether the at least one evidence-based node was executed, the execution result corresponding to the at least one evidence-based node, the weight corresponding to each of the at least one evidence-based node, the data source path corresponding to each of the at least one evidence-based node, the node score corresponding to each of the at least one evidence-based node, and the final score of the target clinical pathway; if it is determined from the pre-configured clinical pathways that a target clinical pathway corresponding to the diagnostic information does not exist, then determine that the clinical pathway for the diagnostic information is missing; The final score is determined based on the weights corresponding to the at least one evidence-based node, the node scores corresponding to the at least one evidence-based node, and the number of the at least one evidence-based node. The storage unit is used to determine and save evidence-based archive information based on the target medical information and the evidence-based results corresponding to each diagnostic information; wherein, the evidence-based result of any diagnostic information is the execution result of the diagnostic information, or the clinical pathway of the diagnostic information is lost.

9. A computer device, characterized in that, The computer device includes a processor that executes a computer program stored in a memory to implement the steps of the diagnostic multidimensional analysis method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program executable by a computer device, which, when run on the computer device, causes the computer device to perform the steps of the diagnostic multidimensional analysis method as described in any one of claims 1-7.

Images