Medical information joint management method and system based on DRG and DIP

Abstract

The invention discloses a medical information joint management method and system based on DRG and DIP, and relates to the technical field of medical information management systems.The method comprises the steps that diagnosis and treatment information of a patient is acquired, and based on the information, a first planning path and a corresponding patient score index are generated through a path planning model; diagnosis and treatment information of a patient is identified through a DRG system, disease type groups and disease type complexity are determined, and the information is input into a DIP system for disease type score prediction. And comparing the patient score with the predicted score, and outputting a score difference index. The difference indicator is fed back to the joint information management system to optimize the path planning model. The technical problem that the system efficiency and optimization decision are affected due to insufficient data integration and real-time optimization capability in the existing medical information management system is solved, and the technical effects of improving the system operation efficiency and decision precision and optimizing medical resource configuration are achieved by improving the data integration and real-time optimization capability of the medical information management system.

Description

Technical Field

[0001] The present application relates to the technical field of medical information management systems, and in particular to a medical information joint management method and system based on DRG and DIP. Background Art

[0002] With the continuous development of medical informatization, traditional medical information management systems face problems such as lagging data integration, poor information flow, and lack of real-time optimization decision support when processing large amounts of patient data and medical resource allocation. Especially in the aspects of disease diagnosis, treatment path planning, and resource allocation, information is not updated in a timely manner, and the decision-making process relies on static rules and manual intervention, resulting in low treatment efficiency and waste of resources. Traditional systems are usually difficult to efficiently integrate data from different sources and lack flexibility and automated optimization mechanisms. Therefore, there is an urgent need for a system that can integrate, process, and optimize medical data in real time to improve medical service efficiency, reduce resource waste, and ensure the quality of patient medical services, thereby promoting the intelligence and efficiency of medical management.

[0003] At the current stage, relevant technologies suffer from insufficient data integration and real-time optimization capabilities in medical information management systems, leading to technical problems that affect system efficiency and optimization decisions. Summary of the invention

[0004] This application solves the technical problems of insufficient data integration and real-time optimization capabilities in existing medical information management systems, which affect system efficiency and optimization decisions, by providing a medical information joint management method and system based on DRG and DIP.

[0005] This application provides a joint management method for medical information based on DRG and DIP, including: Construct a joint information management system, which includes a DRG information system and a DIP information system, and is connected to a path planning model; obtain patient diagnosis and treatment information, and the path planning model obtains a first planned path and a patient score index of the first planned path according to the patient diagnosis and treatment information; identify the patient diagnosis and treatment information according to the DRG information system, and determine DRG group information, and the DRG group information includes disease group and disease complexity; the DIP information system predicts the disease score of the DRG group information, obtains a predicted score index, compares the patient score index with the predicted score index, and outputs a score difference index; records the score difference index, and feeds the score difference index back to the joint information management system to optimize the path planning model.

[0006] This application also provides a joint medical information management system based on DRG and DIP, including: A joint information management system construction module, the joint information management system construction module is used to construct a joint information management system, the joint information management system includes a DRG information system and a DIP information system, and the joint information management system is connected to a path planning model; a first planning path acquisition module, the first planning path acquisition module is used to acquire patient diagnosis and treatment information, the path planning model acquires a first planning path and a patient score index of the first planning path according to the patient diagnosis and treatment information; a diagnosis and treatment information identification module, the diagnosis and treatment information identification module is used to identify the patient diagnosis and treatment information according to the DRG information system, and determine the DRG group information, the DRG group information includes disease group and disease complexity; a prediction score index acquisition module, the prediction score index acquisition module is used for the DIP information system to predict the disease score of the DRG group information, obtain the prediction score index, compare the patient score index with the prediction score index, and output the score difference index; a path planning model optimization module, the path planning model optimization module is used to record the score difference index, and feed the score difference index back to the joint information management system to optimize the path planning model.

[0007] The medical information joint management method and system based on DRG and DIP proposed in this application is to first obtain patient diagnosis and treatment information, and based on this information, generate a first planning path and a corresponding patient score index through a path planning model. The DRG system identifies the patient's diagnosis and treatment information, determines the disease group and disease complexity, and inputs the information into the DIP system for disease score prediction. The patient score is compared with the predicted score, and the score difference index is output. The difference index is fed back to the joint information management system to optimize the path planning model. By improving the data integration and real-time optimization capabilities of the medical information management system, the technical effect of improving system operation efficiency and decision-making accuracy and optimizing the allocation of medical resources is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0009] Figure 1 A flow chart of the joint management method of medical information based on DRG and DIP provided in an embodiment of the present application.

[0010] Figure 2A structural diagram of the medical information joint management system based on DRG and DIP provided in an embodiment of the present application.

[0011] Explanation of the reference numerals: joint information management system construction module 10, first planning path acquisition module 20, diagnosis and treatment information identification module 30, prediction score index acquisition module 40, path planning model optimization module 50. DETAILED DESCRIPTION

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0013] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0014] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0015] The present application embodiment provides a joint management method of medical information based on DRG and DIP, such as Figure 1 As shown, the method includes: Step S100, construct a joint information management system, the joint information management system includes a DRG information system and a DIP information system, and the joint information management system is connected to the path planning model. Specifically, when constructing a joint information management system, it is necessary to equip high-performance servers, large-capacity storage devices and stable network architecture to meet the needs of data processing, storage and transmission. The DRG information system relies on the International Classification of Diseases (ICD) coding system, uses program codes to extract and standardize information such as patient disease diagnosis, and based on the detailed grouping rules pre-stored in the database, with the help of logical judgment algorithms and data matching algorithms, accurately determines the DRG group to which the patient belongs and calculates the complexity of the disease to generate structured data. The DIP information system uses data collection tools to collect the payment standard data of the disease score of the medical insurance department and the historical diagnosis and treatment data of the hospital. After data cleaning, it uses statistical methods such as linear regression to construct a disease score prediction model to achieve accurate estimation of the disease score and output relevant data. The path planning model uses linear programming and integer programming algorithms in operations research, combined with hospital resource allocation information and patients' DRG and DIP data, to construct objective functions and constraint equations, which are solved through optimization algorithms. At the same time, the decision tree algorithm is used to learn and train historical data, thereby planning the sequence and time arrangements of each link from the patient's entry into the medical process, including the sequence of examination items, estimated length of hospital stay, etc., to ultimately achieve efficient optimization and precise support for the entire medical process by the joint information management system, improve the efficiency of medical resource utilization and scientific management, ensure close collaboration and stable operation of all parts of the system, and realize intelligent and refined medical information management.

[0016] In one possible implementation, a joint information management system is constructed, the joint information management system includes a DRG information system and a DIP information system, the joint information management system is connected to a path planning model, and step S100 further includes step S110, initializing the path planning model, including defining paths and path nodes, wherein each path represents all processes from initial diagnosis and treatment to the end of diagnosis and treatment, and the connection between nodes represents the diagnosis and treatment steps. Specifically, when initializing the path planning model, a graph data structure is used to describe the diagnosis and treatment process, wherein vertices serve as path nodes, covering registration, examination, diagnosis, treatment and other diagnosis and treatment stages, and attribute values ​​such as time consumption, required medical resource types and quantities are assigned based on historical medical data statistical analysis and expert experience; the connecting edges between nodes represent the sequential relationship of the diagnosis and treatment steps, and the direction and weight are set according to the actual situation to clarify the sequence and closeness. At the same time, through in-depth mining and analysis of a large number of historical medical records, we have sorted out common medical treatment process patterns for different disease types and severity, such as colds and heart disease surgery, and abstracted them into path templates. After being encoded and stored by algorithms, we can quickly match and call the initial path for new patients, providing a basic framework for guiding the subsequent medical treatment process, improving efficiency and quality, and rationally allocating resources.

[0017] Step S120, collect patient diagnosis and treatment sample information within a preset time period. Specifically, when collecting patient diagnosis and treatment sample information within a preset time period, a comprehensive analysis is required to determine the appropriate time period range. For example, a comprehensive hospital may select the past year to ensure that the data is both timely and representative. Then, with the help of data interface technology, data is extracted from multiple information systems such as the hospital's HIS, EMR, LIS, PACS, etc., covering basic patient information, diagnosis and treatment records, inspection and test numerical results, and imaging data, etc., to ensure that the data is complete and accurate. In this process, data cleaning technology is used to remove invalid and duplicate data by relying on data deduplication algorithms and quality assessment rules to ensure reliable data quality. Finally, according to the research purpose and path planning model requirements, the data is classified and labeled. For example, for specific diseases, relevant patient data is screened out and key information such as disease type, severity of the disease, and treatment results are labeled to lay a solid data foundation for model training optimization, improve path planning effects, and assist in medical decision-making and patient treatment.

[0018] Step S130, construct an objective function, wherein the indicators of the objective function include diagnosis and treatment time, resource utilization and medical cost. Specifically, the construction of the objective function requires a comprehensive analysis of the three key indicators of diagnosis and treatment time, resource utilization and medical cost. For diagnosis and treatment time, from the time the patient is admitted to the hospital, the hospital information system is used to record the time of each link, collect a large amount of historical data, and use statistical analysis methods to calculate the average, median, standard deviation time and correlation of each link under different diseases and conditions, and construct the function In terms of resource utilization, for medical equipment, the equipment management system is used to obtain the boot, idle, and usage time, calculate the utilization rate and analyze the idle situation; for ward beds, the bed management system is used to calculate the occupancy rate, vacancy time, turnover number, etc.; for medical staff, the working time and saturation are calculated in combination with the shift schedule and work records, and the rationality of the allocation is analyzed, so as to build a In terms of medical cost indicators, drug procurement costs are calculated based on procurement and dispensing records; inspection and testing costs are calculated based on charging standards and the number of times performed; surgical costs include consumables, anesthesia, and labor costs; ward costs are calculated based on ward grade charges and hospital stay days; and medical staff labor costs are apportioned based on the salary system and working hours. Finally, the weight coefficient is determined by using the hierarchical analysis method. , , , construct the objective function , and the goal in the subsequent path planning model is to minimize it, optimize the diagnosis and treatment process, improve hospital efficiency and quality, control costs and ensure effective use of resources.

[0019] Step S140, according to the objective function, the initialization path planning model is trained with the patient diagnosis and treatment sample information as input, with minimization of the objective function as the convergence condition, the converged path planning model is output, and the converged path planning model is connected to the joint information management system. Specifically, the association between the objective function and the patient diagnosis and treatment sample information is first clarified. The former is constructed based on indicators such as diagnosis and treatment time, resource utilization, and medical cost, and is used to measure the efficiency and benefit of diagnosis and treatment. The latter is a data set that reflects the actual diagnosis and treatment situation. When training the initialization path planning model, the patient diagnosis and treatment sample information is input in sequence, and the model simulates a variety of diagnosis and treatment paths and resource allocation schemes based on this, and evaluates it with the objective function. The training takes minimization of the objective function as the convergence condition, uses the optimization algorithm, and adjusts the path and resource allocation according to the derivative information of the objective function to reduce the objective function value. When the convergence threshold is met, the converged path planning model is output. Finally, by establishing a data interface and communication protocol, the model is connected to the joint information management system, so that the model can obtain the diagnosis and treatment information in the system and can also feed back the optimization plan to the system, thereby improving the intelligence and efficiency of the hospital's diagnosis and treatment process and providing patients with better medical services.

[0020] Step S200, obtain the patient's diagnosis and treatment information, and the path planning model obtains the first planned path and the patient score index of the first planned path according to the patient's diagnosis and treatment information. Specifically, first collect the patient's basic information, symptoms, diagnosis results, examination and test results, treatment methods and other data from various information systems of the hospital, and integrate them through the integrated platform to provide a complete and accurate data basis for the follow-up. Then, the path planning model uses operations research and machine learning technology based on the preset rule algorithm, refers to the historical diagnosis and treatment paths of similar patients, the current resource status of the hospital and the diagnosis and treatment guidelines, and generates the first planned path for the patient, covering subsequent examinations, treatment adjustments, hospitalization time and the sequence of diagnosis and treatment links. At the same time, the model analyzes factors such as expected rehabilitation effects, possible medical expenses, resource consumption and diagnosis and treatment time, and uses mathematical formulas based on different weights to calculate the patient score index, so as to quantitatively measure the pros and cons of the first planned path and provide a basis for subsequent optimization.

[0021] In a possible implementation, the patient's diagnosis and treatment information is obtained, and the path planning model obtains a first planning path and a patient score index of the first planning path according to the patient's diagnosis and treatment information. Step S200 further includes step S210, extracting the key planning path of the first planning path. Specifically, to extract the key planning path of the first planning path, first parse the generated first planning path, which covers many diagnosis and treatment links and steps, and each link has information such as corresponding time arrangements, resource requirements and sequence relationships. Then, the key path is determined by combining data analysis with professional knowledge: from the time dimension, the links that have a great impact on the total diagnosis and treatment time, strict order and long time consumption, such as specific examination items or waiting for expert diagnosis, are included; from the perspective of resource dependence, the diagnosis and treatment steps that rely on scarce medical resources (such as high-end equipment, professional medical care) are taken into account; from the perspective of logical relationship, the link sequence with strong correlation and sequential constraint relationship, where the output of the previous link is a necessary condition for the next link, is regarded as the key part. After comprehensive analysis and screening of the above-mentioned factors, we extracted the links and steps that play a key role and have a significant impact on the diagnosis and treatment process, forming a key planning path so that we can focus on monitoring and management in the future to ensure efficient and smooth diagnosis and treatment.

[0022] Step S220, real-time monitoring of each node on the key planning path is performed, and a real-time monitoring data set is output. Specifically, the specific diagnosis and treatment links corresponding to each node in the key planning path must first be determined, such as registration, examination, diagnosis, treatment, drug distribution, hospitalization status, etc. Then, corresponding monitoring means are implemented for each node, such as recording relevant information and monitoring the process through the hospital information system at the registration place; the inspection items are connected to the equipment management system to obtain the inspection time, results and equipment operation status; the diagnosis link uses the electronic medical record system and monitors the doctor-patient communication records in accordance with regulations; the treatment implementation connects the equipment interface to collect treatment parameters and patient reactions; the drug distribution is associated with the pharmacy system to record the distribution details; the hospitalization status uses the ward management system to monitor various types of hospitalization information. Finally, the data continuously collected by each node is sorted and summarized according to a unified format standard to form a real-time monitoring data set containing rich key diagnosis and treatment information, which provides data support for subsequent diagnosis and treatment delay risk analysis and process optimization, ensures efficient and safe diagnosis and treatment, and improves the quality of medical services and patient satisfaction.

[0023] Step S230, perform diagnosis and treatment delay risk analysis according to the real-time monitoring data set to obtain a delay risk index. Specifically, first obtain a real-time monitoring data set containing information such as the start, expected and actual completion time of each node of the key planning path, resource allocation, data transmission status, etc. from the database, and input it into the diagnosis and treatment delay risk analysis model. The model uses a time series analysis algorithm to process time data to identify potential delay nodes, and uses an association rule mining algorithm to analyze resource dependencies and data interaction patterns to find out delay risk factors. For each factor, a quantitative evaluation is performed according to preset rules and weights. For example, if a certain inspection node is delayed for a certain period of time, the contribution value to the overall delay risk is calculated in combination with its impact weight. The model accumulates the contribution values ​​of all factors and analyzes complex situations such as delay compensation and chain reactions. Finally, a delay risk index with a value of 0 to 1 is calculated and output. The closer the value is to 1, the more intervention is needed to ensure that diagnosis and treatment are carried out efficiently according to the plan, provide support for hospital decision-making, and improve the efficiency and quality of medical services.

[0024] Step S240, when the delay risk index is greater than the preset risk threshold, the first planning path is optimized and the second planning path is output. Specifically, the delay risk index is continuously compared with the preset risk threshold, which is set in advance based on past diagnosis and treatment data, expert experience and hospital requirements. Once the delay risk index exceeds the threshold, the optimization process of the first planning path is started. The path optimization module first disassembles it into a computer-processable data structure, such as presenting the diagnosis and treatment process in a graph structure, with nodes as links and directed edge tables with order and dependencies. Then, an intelligent search algorithm (such as simulated annealing algorithm) is used to search in the solution space, try to adjust the link sequence and reallocate resources, and consider early or parallel operations in the case of high-delay inspection links. At the same time, a constraint satisfaction algorithm is used to ensure that medical condition constraints are met, and the real-time monitoring data and the diagnosis and treatment knowledge base are used to evaluate the impact of the adjustment plan on the delay risk. After multiple iterative comparisons, a plan that makes the delay risk index reach an acceptable range is selected and integrated into the second planning path. Continuous monitoring and evaluation are performed in the future to optimize again, ensure efficient and orderly diagnosis and treatment, and provide support for hospital management.

[0025] Step S300, the patient's diagnosis and treatment information is identified according to the DRG information system, and the DRG group information is determined, and the DRG group information includes the disease group and the complexity of the disease. Specifically, the DRG information system first obtains the patient's diagnosis and treatment information from various databases and information systems of the hospital, including basic data, medical records, examinations and tests, treatments, costs, and length of hospital stay, and transmits it to the processing center through a standardized interface. It is then cleaned and standardized, the text information is converted into unified medical terms using natural language processing technology, the numerical data is verified and standardized, and the treatment measures are coded according to the classification coding system. Then the disease group is determined, the main diagnostic codes are accurately matched with the help of the ICD coding knowledge base, the accompanying diseases, complications and treatment operations are analyzed, and the rule base built based on clinical pathways and big data analysis is combined with the decision tree algorithm for division. When calculating the complexity of a disease, the complexity of the disease is evaluated from the dimensions of age, comorbidities, and severity of the disease. Each factor is quantified and accumulated, and the index value is obtained by normalization from 0 to 1. The higher the value, the more complex the disease. The complexity of resource consumption analyzes the use of various resources, and the inspections, surgeries, drugs, and hospital days are calculated and summarized separately, and the index value is obtained by normalization. Finally, the linear weighted method is used to determine the comprehensive value of the complexity of the disease according to the pre-set weight coefficient (such as 0.6 for the complexity of the disease and 0.4 for the complexity of resource consumption), and the DRG group information is generated by integrating the disease group and complexity to provide support for medical management decision-making and medical insurance fee settlement, so as to achieve reasonable allocation and efficient use of resources and improve hospital operation management and service quality.

[0026] In a possible implementation, the patient's diagnosis and treatment information is identified according to the DRG information system to determine the DRG group information, which includes the disease group and the complexity of the disease. Step S300 further includes step S310, in which the patient's diagnosis and treatment information is processed in ICD coding format according to the DRG information system to obtain the patient's ICD code. Specifically, the DRG information system first collects the patient's basic information, clinical data, examination and test results, treatment measures and other diagnosis and treatment information from various information sources in the hospital, and transmits them to the data receiving module through the data transmission channel to ensure that the data is complete and accurate, providing a basis for subsequent coding. Then, the text parsing and keyword extraction program is started, and the text information such as diagnosis description, symptom manifestation and treatment methods are analyzed using natural language processing technology, and the key information is extracted and organized into a standardized data structure. Then the key information is matched with the built-in ICD coding library that follows international standards and is optimized and expanded in combination with local medical practices. The coding is found by combining matching and semantic similarity algorithms, and the final coding is determined by semantic analysis for complex diagnoses. During this process, the system performs coding verification and quality control based on medical logic rules and statistical data, checks the rationality of coding, and compares historical data. If any abnormality is found, an alarm is triggered and feedback is given to professionals for review and correction to ensure the high quality and reliability of the patient's ICD coding, provide accurate data support for subsequent work, and improve hospital management level and resource allocation efficiency.

[0027] Step S320, match according to the patient's ICD code to determine the disease group to which it belongs. Specifically, the DRG information system stores a DRG grouping rule library based on a large amount of medical data, expert experience and relevant standards, which clarifies the mapping relationship between different ICD codes and disease groups, covering the main diagnosis, concomitant diagnosis, treatment methods and patient age, gender and other factors. After obtaining the patient's ICD code, the system starts the intelligent matching program, first parses the code to extract key disease diagnoses and additional information, and then uses a multi-dimensional matching algorithm to compare it with the rules of the rule library one by one, accurately matches the main diagnosis code and comprehensively analyzes the concomitant diagnosis and other factors. If the patient's main diagnosis is "coronary atherosclerotic heart disease" and is accompanied by "hypertension", the system will look for the corresponding rules, and if it meets the requirements, the disease group will be determined. For special cases, the system has fuzzy matching and intelligent reasoning capabilities, finds close matching results based on similar cases and clinical logical reasoning, and records key data and decision-making basis for retrospective review. After the disease group is determined, the information will be used in subsequent medical processes to help hospitals with refined management and scientific decision-making, ensure efficient and fair medical services, and improve the operating efficiency and service quality of the medical system.

[0028] Step S330, output the complexity of the disease and the complexity of resource consumption according to the patient's diagnosis and treatment information. Specifically, the DRG information system first collects patient diagnosis and treatment information from various information systems of the hospital, covering basic information, symptoms, diagnosis results, examination and test data, treatment process information and details of hospitalization expenses, etc., and transmits them to the core analysis module through standardized interfaces and security protocols. Then, natural language processing and data mining algorithms are used to preprocess and extract features of text data, quantify key information of symptoms and diagnosis texts, and judge the degree of abnormality and assign values ​​to numerical data according to medical standards. When calculating the complexity of the disease, the model based on machine learning algorithm is used to analyze and quantify the scores based on factors such as age, underlying diseases, symptoms, diagnosis, complications and examination results. The weights of each factor are determined by historical case training to obtain the complexity of the disease. When calculating the complexity of resource consumption, the detailed classification statistics of hospitalization expenses are used, the weight of the proportion of expenses is considered, and the evaluation model is used to calculate weighted calculations in combination with the use of medical resources and the number of days in hospital. Finally, these two key indicators are output to provide a basis for subsequent work, help hospitals to carry out refined management and reasonable allocation of resources, improve the efficiency and quality of medical services, and protect the rights and interests of patients.

[0029] Step S340, calculate according to the complexity of the disease and the complexity of resource consumption to determine the complexity of the disease. Specifically, the quantified values ​​of the complexity of the disease and the complexity of resource consumption obtained from the database represent the disease and resource consumption respectively and have a standardized quantitative form. Then call the optimized disease complexity calculation module based on historical cases, statistical laws and expert experience. This module determines the weight coefficient and calculation method of the complexity of the disease and resource consumption according to the disease type. For example, the weight of the complexity of acute and severe diseases is high, and the weight of the complexity of resource consumption of chronic diseases is relatively prominent. Then substitute the two complexity values ​​into the module and calculate accurately according to the established rules, such as multiplying the weights and adding them to obtain the complexity value of the disease. After that, the system automatically verifies and reviews the value from multiple aspects, compares and analyzes with the historical data of the same type of disease, and verifies according to the preset medical logic rules. If the value is abnormal, it will be traced back to troubleshoot the problem. Only after confirming that it is reasonable and accurate, it will be determined as the complexity of the disease and stored for association integration, which will be used for medical cost accounting, resource allocation, program optimization and quality evaluation, etc., to ensure the scientificity, rationality and efficiency of medical services, and provide support and basis for patients and medical institutions.

[0030] Step S350, generate DRG group information according to the disease group and disease complexity. Specifically, after obtaining the disease group and disease complexity, access the DRG group information template library formulated based on national and local medical conditions, which contains DRG group information templates corresponding to various disease groups and corresponding disease complexity ranges, with detailed structure and content format. Then, the disease group is located by an accurate matching algorithm, and the disease complexity value is compared with the template complexity interval to select a suitable template. Subsequently, the system extracts and organizes the disease group, disease complexity and other related feature identification, statistical coding and other information according to the template requirements, and fills and combines them into complete DRG group information according to the established format. In this process, the system strictly verifies the format and logic, checks the data type, required items and logical relationships, and automatically alarms and prompts error messages if there are any problems. After the information passes the verification, it is stored in the database for subsequent work such as medical expense settlement, quality assessment, and performance management, which helps to reasonably allocate and efficiently utilize medical resources, improve the operational efficiency and service quality of the medical system, and protect the rights and interests of patients and the scientificity and fairness of medical services.

[0031] In a possible implementation, according to the patient's diagnosis and treatment information, the complexity of the disease and the complexity of resource consumption are output, and step S330 further includes step S331, wherein the complexity of the disease is obtained by extracting the number of complications and the severity index of complications, as well as the number of concomitant diseases and the severity index of concomitant diseases from the patient's diagnosis and treatment information, and calculating the number of complications and the severity index of complications, as well as the number of concomitant diseases and the severity index of concomitant diseases. Specifically, firstly, the hospital's various information databases are connected to obtain detailed diagnosis and treatment information of the patient including admission, course of disease, examination and testing, doctor's orders and surgical records, and then natural language processing and medical semantic recognition algorithms are used for in-depth analysis to accurately extract the number of complications and the number of concomitant diseases. When extracting the severity index of complications, the abnormal degree of physiological indicators of the examination and testing, the image lesion situation and the severity of symptoms are comprehensively considered to determine the quantitative value, and the severity index of concomitant diseases is extracted and scored according to the impact on daily living ability and treatment and nursing needs. Then, according to the statistical analysis of a large number of case data and the experience of medical experts, corresponding weights are set for each indicator for different disease categories, and when calculating, the numerical value of each indicator is multiplied by the weight and then accumulated and summed to obtain the quantified value of the complexity of the disease. During the process, we control the accuracy and completeness of the data, cross-check the data and perform logical verification, compare the results with historical data, and if the values ​​are abnormal, conduct a retrospective review and issue an alarm to notify personnel to review and correct them, to ensure the reliability of the results, provide support for medical decision-making, resource allocation and treatment plan optimization, improve the quality and efficiency of medical services, and protect the rights and interests of patients.

[0032] Step S332, the resource consumption complexity is calculated by recording the treatment time and medical equipment usage information of each node from the patient's diagnosis and treatment information. Specifically, first integrate with the hospital information system to obtain the patient's complete diagnosis and treatment information, then use text parsing and data extraction technology to identify each diagnosis and treatment node, accurately record its treatment time (accurate to minutes and seconds) and the name, model, start and end time and frequency of use of the medical equipment used. Then, according to the preset rules, the data is quantified and processed, and the treatment time is converted into a score according to the average resource consumption of different departments and projects, such as the ordinary outpatient diagnosis time score is low, and the complex surgery time score is high; medical equipment sets a resource consumption coefficient based on purchase, maintenance, energy consumption, clinical importance and scarcity, such as expensive large-scale imaging equipment with a high coefficient and ordinary thermometers with a low coefficient, and the resource consumption quantification value is the number of times or duration of equipment use multiplied by the coefficient. Finally, the quantitative score of treatment time and the quantitative value of equipment resource consumption are added together to obtain the resource consumption complexity value. During the process, the system verifies the data integrity and rationality of the results multiple times. If any abnormality is found, it will be back-checked and corrected, and log records will be generated to ensure that the calculation is accurate and reliable, providing support for hospital cost control, resource allocation, service pricing, etc., improving operational management efficiency and service quality, and protecting rights and interests.

[0033] Step S400, the DIP information system predicts the disease score of the DRG group information, obtains the predicted score index, compares the patient score index with the predicted score index, and outputs the score difference index. Specifically, the DIP information system first integrates the massive historical medical data from multiple sources such as the information systems of various departments of the hospital and the medical insurance database, cleans and unifies the format and coding standards, and then uses feature engineering to extract key features. Then, a machine learning algorithm such as GBDT or DNN is selected to divide the data into a training set, a validation set, and a test set in proportion, and the case feature vector is used as input, and the actual disease score is used as the output label training model. After multiple rounds of iterative training, the parameters are saved after the model meets the performance index on the test set. When the DRG group information is received, the system parses and extracts the feature vector input model to obtain the predicted score index, and at the same time calculates the patient score index according to the established rules based on the actual diagnosis and treatment data of the patient, and compares the two to obtain the score difference index, which is output to hospital managers, medical insurance auditors, clinicians, etc. in the form of reports or charts. Hospital managers optimize resource allocation based on this, medical insurance auditors review the rationality of expenses, and clinical doctors reflect on treatment plans to protect the rights and interests of patients and the safe and sustainable operation of medical insurance funds, and achieve rational use of medical resources and effective cost control.

[0034] In a possible implementation, the DIP information system predicts the disease score of the DRG group information, obtains the predicted score index, compares the patient score index with the predicted score index, and outputs the score difference index. Step S400 further includes step S410, wherein the DIP information system includes a disease score prediction model, and the disease score prediction model is connected to the path planning model to receive the historical planning path samples output by the path planning model, and the path scores corresponding to each historical planning path sample. Specifically, the DIP information system is first connected to the information systems of various departments of the hospital and the medical insurance database through an interface to collect massive historical medical data covering comprehensive medical information such as diagnosis, treatment, examination and testing, length of hospital stay, and cost details. Then, a data cleaning algorithm is used to correct erroneous data and fill in missing values, and the data format and coding standards are unified according to the medical term mapping table. Then, a feature vector set containing key features such as diagnosis, treatment, and basic patient information is constructed using disease classification coding and quantitative means. After that, the path planning model loads the processed data, uses cluster analysis and other techniques to generate historical planning path samples containing information such as diagnosis and treatment processes and resource consumption, and assigns path scores according to comprehensive rules. At the same time, algorithms such as GBDT or DNN are used to build a disease score prediction model, divide the training, validation, and test sets, use the feature vector as input and the actual disease score as the output label to train the model, use the validation set to tune the hyperparameters, and save the parameters when the test set performance indicators meet the standards. When receiving the DRG group information, the system parses and extracts the feature vector input into the disease score prediction model to obtain the predicted score index, and then calculates the patient score index based on the patient's actual diagnosis and treatment data. The two are compared to obtain the score difference index, and finally presented in the form of reports or charts to hospital managers, medical insurance auditors, and clinicians for analysis and evaluation, so that managers can optimize resource allocation, auditors can supervise costs, and doctors can optimize diagnosis and treatment plans to protect the rights and interests of patients and the safe and sustainable operation of medical insurance funds.

[0035] Step S420, the DIP information system obtains similar planning path samples of the DRG group information according to the DRG group information. Specifically, the DIP information system first parses and extracts features of the input DRG group information, identifies key data such as disease diagnosis codes, surgical operation codes, and patient age, and converts them into structured feature vectors. Then, the internal historical medical database is connected to extract feature data of all cases, and the feature data has the same structural dimension as the DRG group feature vector. Then, a similarity calculation technology based on distance (such as Euclidean distance, cosine similarity) or model (such as decision tree, neural network) is used to determine the similarity between historical cases and current DRG group information, and historical cases with higher scores are screened out according to the set similarity threshold, and the corresponding diagnosis and treatment paths are similar planning path samples. Finally, the diagnosis and treatment details of these samples are deeply excavated, including information such as examinations and tests, treatment methods, drug use, hospitalization time, and resource consumption, and are organized into data sets for use in disease score prediction, resource allocation optimization, diagnosis and treatment process improvement, etc., to improve the quality and efficiency of medical services, and to protect the rights and interests of patients and the stable development of the medical system.

[0036] Step S430, collecting the path score samples corresponding to the similar planning path samples. Specifically, clarify the constituent elements of the path score samples, set the score calculation model based on the professional evaluation indicators of medical resource consumption (including medicines, examinations, surgeries, hospitalization, etc.), disease complexity (measured by the number of complications, severity indicators), treatment effect (combined with indicators such as cure rate and improvement rate), etc. The score calculation model can be a weighted sum linear model, such as linear regression, to accurately determine the score of each similar planning path sample. Using database association technology, the diagnosis and treatment link data of the similar planning path samples are associated with the data required for each score calculation (obtained from hospital charges, examinations, surgery management, electronic medical records, etc.), and the specific data values ​​are extracted according to the rules through data extraction tools and algorithms, and substituted into the score calculation model to calculate the path score samples. After the calculation is completed, the accuracy and rationality are verified by comparing and analyzing the path score data of historical cases and inviting medical experts to manually review. If there is a deviation, it is retrospectively checked and optimized to ensure the quality of the path score samples, which provides strong support for disease score prediction and medical decision-making.

[0037] Step S440, predict the disease score according to the path score sample and obtain the predicted score index. Specifically, the collected path score sample data is sorted, the integrity is checked and the missing values ​​are filled by the mean, median or interpolation method, and then the data is standardized by means of logarithmic transformation, normalization and encoding conversion to unify the dimension and numerical range. Then, key features such as diagnosis and treatment process, disease characteristics, and basic information of patients are extracted from the sample to construct a high-dimensional feature space, and redundant features are removed by methods such as Pearson correlation coefficient or PCA. Subsequently, a machine learning algorithm such as linear regression and decision tree is selected according to the data characteristics and task requirements, and the data is divided into training set, validation set and test set in proportion. The model is trained with feature vector as input and path score as output label, and the hyperparameters are tuned with the validation set. The parameters are saved when the performance indicators of the test set meet the standards. When predicting the score of a new disease, first preprocess its information and perform feature engineering, input it into the trained model to obtain the predicted score indicator, and then compare it with the historical data of the same type of disease. If there is any abnormality, retrospective inspection and optimization are performed to ensure that the indicator is accurate and reliable, providing a basis for medical decision-making.

[0038] Step S500, record the score difference index, and feed the score difference index back to the joint information management system to optimize the path planning model. Specifically, create a data storage structure, such as a database table or an independent file, to accurately and completely record the score difference index and related metadata such as patient basic information and DRG group information, and verify the integrity and accuracy of the data when entering. Then design an interface program that follows the agreed communication protocol (such as HTTP, etc.), encrypt the score difference index data before transmission (using algorithms such as AES) and add a check code, and transmit it to the joint information management system safely and stably. After receiving it, the joint information management system stores it in a buffer area or table, and the data analysis module reads it regularly, finds out the problems of the path planning model through statistical analysis and mining, and determines the optimization direction. For rule-based models, directly modify the rule parameters, such as adjusting the disease hospitalization duration estimation rule; for machine learning models, collect more data for retraining, and use the score difference index as a feedback signal to adjust the weight and structure, so that the model is more adaptable to reality, provide better diagnosis and treatment path suggestions for medical services, and improve medical quality and efficiency.

[0039] In a possible implementation, the score difference index is recorded, and the score difference index is fed back to the joint information management system to optimize the path planning model. Step S500 further includes step S510, defining a difference threshold. If the score difference index is greater than or equal to the difference threshold, the model parameters of the path planning model are optimized by the first feedback optimization path. Specifically, based on the statistical analysis of historical data and the experience of medical experts, factors such as the disease type, the complexity of diagnosis and treatment, and the range of resource fluctuations are comprehensively analyzed to accurately set the difference threshold, which will be used as a key indicator for determining whether to optimize the model. Then, the system reads the score difference index from the stored data, and performs rigorous numerical comparison and logical judgment with the set threshold, and records the relevant case information in detail for traceability. When the score difference index is greater than or equal to the threshold, the first feedback optimization path is opened, and the range of model parameters to be adjusted related to medical resource allocation, diagnosis and treatment process, etc. is first determined. For example, for the situation where resource consumption is high due to too many examination items, the inspection resource allocation parameters are selected. Then, optimization algorithms such as gradient descent or genetic algorithms are used to calculate the parameter gradient based on the score differences and update the parameter values ​​according to the learning rate. During the process, model performance indicators are continuously monitored, and historical and verification data are used to verify the model's generalization ability and accuracy to ensure that the optimized model can better fit the actual medical scenario, improve the rationality and accuracy of path planning, and ensure the quality of medical services, patient rights and interests, and the rational allocation of resources.

[0040] Step S520, if the score difference index is less than the difference threshold, the model parameters of the path planning model are optimized by the second feedback optimization path. Specifically, the score difference index value is first extracted from the special storage area, and the conditional judgment statement (such as "if-else" statement) in the programming is accurately compared with the preset difference threshold. When it is determined that the score difference index is less than the threshold, the second feedback optimization path is selected according to the preset optimization strategy library. This path focuses on model fine-tuning for the situation where the score difference is small but there is room for optimization. Then identify the subset of model parameters related to the path, such as parameters that affect the details of the diagnosis and treatment path, the secondary part of the medical resource allocation weight, and the secondary dimension of the disease characteristics. Then, a specific optimization algorithm such as a simulated annealing algorithm or a small step random gradient descent variant algorithm is used to adjust the selected parameters, and the model performance indicators (such as changes in the score difference index, the accuracy and recall rate of the verification set, etc.) are continuously monitored during the optimization process to ensure that the optimization direction is correct and does not affect the existing performance, so that the path planning model can more accurately plan the diagnosis and treatment path, improve the efficiency of medical resource utilization, service quality and effect, and protect the rights and interests of patients and the stable operation of the medical system.

[0041] Step S530, wherein the feedback adjustment weight of the first feedback optimization path is greater than the feedback adjustment weight of the second feedback optimization path. Specifically, when the system is initialized, feedback adjustment weights are set for the first and second feedback optimization paths respectively based on medical expertise and historical data analysis. Since the first feedback optimization path is for situations with large score differences, its weight is relatively large to quickly correct large deviations of the model, involving key resource allocation and major diagnosis and treatment decision parameters; while the second feedback optimization path is for situations with small score differences, its weight is relatively small, acting on secondary parameters. The system regularly reads the score difference index and compares it with the threshold. If it is greater than or equal to the threshold, the first path is selected, and if it is less than the threshold, the second path is selected, and the conditional judgment statement is used to ensure accurate selection. After selecting the first path, the key parameter set is determined, and an efficient algorithm (such as a gradient algorithm) is used to make large adjustments based on large weights, and the performance is monitored in real time; when the second path is selected, a subset of secondary parameters is identified, and a mild algorithm (such as a simulated annealing variant) is used to fine-tune based on small weights, and performance changes are continuously observed, so that the model can be reasonably optimized under different circumstances, the path planning capability is improved, and the quality of medical services and resource utilization are guaranteed.

[0042] The embodiment of the present application adopts the method of obtaining patient diagnosis and treatment information, and based on this information, generates a first planned path and a corresponding patient score index through a path planning model. The patient diagnosis and treatment information is identified through the DRG system, the disease group and the complexity of the disease are determined, and the information is input into the DIP system for disease score prediction. The patient score is compared with the predicted score, and the score difference index is output. The difference index is fed back to the joint information management system to optimize the path planning model. By improving the data integration and real-time optimization capabilities of the medical information management system, the technical effect of improving the system operation efficiency and decision-making accuracy and optimizing the allocation of medical resources is achieved.

[0043] In the above, refer to Figure 1 The medical information joint management method based on DRG and DIP according to an embodiment of the present invention is described in detail. Figure 2 A medical information joint management system based on DRG and DIP according to an embodiment of the present invention is described.

[0044] The medical information joint management system based on DRG and DIP according to the embodiment of the present invention is used to solve the technical problem that the data integration and real-time optimization capabilities of the existing medical information management system are insufficient, which affects the system efficiency and optimization decision-making. By improving the data integration and real-time optimization capabilities of the medical information management system, the technical effect of improving the system operation efficiency and decision-making accuracy and optimizing the allocation of medical resources is achieved. The medical information joint management system based on DRG and DIP includes: a joint information management system construction module 10, a first planning path acquisition module 20, a diagnosis and treatment information identification module 30, a prediction score index acquisition module 40, and a path planning model optimization module 50.

[0045] The joint information management system construction module 10 is used to construct a joint information management system, which includes a DRG information system and a DIP information system, and the joint information management system is connected to a path planning model.

[0046] The first planned path acquisition module 20 is used to acquire patient diagnosis and treatment information. The path planning model acquires the first planned path and the patient score index of the first planned path according to the patient diagnosis and treatment information.

[0047] The diagnosis and treatment information identification module 30 is used to identify the patient's diagnosis and treatment information according to the DRG information system and determine the DRG group information. The DRG group information includes the disease group and the disease complexity.

[0048] The prediction score index acquisition module 40 is used for the DIP information system to predict the disease score of the DRG group information, obtain the prediction score index, compare the patient score index with the prediction score index, and output the score difference index.

[0049] The path planning model optimization module 50 is used to record the score difference index and feed the score difference index back to the joint information management system to optimize the path planning model.

[0050] The specific configuration of the joint information management system construction module 10 will be described in detail below. As described above, a joint information management system is constructed, the joint information management system includes a DRG information system and a DIP information system, the joint information management system is connected to a path planning model, and the joint information management system construction module 10 further includes: a path planning model initialization unit, the path planning model initialization unit is used to initialize the path planning model, including defining paths and path nodes, wherein each path represents all processes from initial diagnosis and treatment to the end of diagnosis and treatment, and the connection between nodes represents the diagnosis and treatment steps; a sample information collection unit, the sample information collection unit is used to collect patient diagnosis and treatment sample information within a preset time period; an objective function construction unit, the objective function construction unit is used to construct an objective function, wherein the indicators of the objective function include diagnosis and treatment time, resource utilization and medical cost; a path planning model output unit, the path planning model output unit is used to train the initialized path planning model according to the objective function with the patient diagnosis and treatment sample information as input, to minimize the objective function as a convergence condition, output a converged path planning model, and connect the converged path planning model to the joint information management system.

[0051] Below, the specific configuration of the first planning path acquisition module 20 will be described in detail. As described above, the patient diagnosis and treatment information is obtained, and the path planning model obtains the first planning path and the patient score index of the first planning path according to the patient diagnosis and treatment information. The first planning path acquisition module 20 further includes: a key planning path extraction unit, the key planning path extraction unit is used to extract the key planning path of the first planning path; a real-time monitoring data set output unit, the real-time monitoring data set output unit is used to perform real-time monitoring of each node on the key planning path and output a real-time monitoring data set; a diagnosis and treatment delay risk analysis unit, the diagnosis and treatment delay risk analysis unit is used to perform diagnosis and treatment delay risk analysis according to the real-time monitoring data set and obtain a delay risk index; a second planning path output unit, the second planning path output unit is used to optimize the first planning path when the delay risk index is greater than a preset risk threshold, and output a second planning path.

[0052] The specific configuration of the diagnosis and treatment information identification module 30 will be described in detail below. As described above, the patient diagnosis and treatment information is identified according to the DRG information system to determine the DRG group information, and the DRG group information includes the disease group and the disease complexity. The diagnosis and treatment information identification module 30 further includes: a coding format processing unit, the coding format processing unit is used to perform ICD coding format processing on the patient diagnosis and treatment information according to the DRG information system to obtain the patient's ICD code; a disease group determination unit, the disease group determination unit is used to match according to the patient's ICD code to determine the disease group to which it belongs; a complexity output unit, the complexity output unit is used to output the complexity of the disease and the complexity of resource consumption according to the patient's diagnosis and treatment information; a disease complexity determination unit, the disease complexity determination unit is used to calculate according to the complexity of the disease and the complexity of resource consumption to determine the complexity of the disease; a group information generation unit, the group information generation unit is used to generate DRG group information according to the disease group and the complexity of the disease.

[0053] Among them, according to the patient's diagnosis and treatment information, the complexity of the disease and the complexity of resource consumption are output, and the complexity output unit further includes: an indicator calculation and acquisition subunit, the indicator calculation and acquisition subunit is used for the complexity of the disease by extracting the number of complications and complication severity indicators, as well as the number of concomitant diseases and concomitant disease severity indicators from the patient's diagnosis and treatment information, and calculating and acquiring the number of complications and complication severity indicators, as well as the number of concomitant diseases and concomitant disease severity indicators; a usage information acquisition subunit, the usage information acquisition subunit is used for the resource consumption complexity by recording the treatment time and medical equipment usage information of each node from the patient's diagnosis and treatment information, and calculating and acquiring the treatment time and medical equipment usage information.

[0054] The specific configuration of the prediction score index acquisition module 40 will be described in detail below. As described above, the DIP information system predicts the disease score of the DRG group information, obtains the prediction score index, compares the patient score index with the prediction score index, and outputs the score difference index. The prediction score index acquisition module 40 further includes: a path score receiving unit, the path score receiving unit is used in which the DIP information system includes a disease score prediction model, the disease score prediction model is connected to the path planning model, and is used to receive the historical planning path samples output by the path planning model, and the path scores corresponding to each historical planning path sample; a similar planning path sample acquisition unit, the similar planning path sample acquisition unit is used for the DIP information system to obtain similar planning path samples of the DRG group information according to the DRG group information; a path score sample acquisition unit, the path score sample acquisition unit is used to collect the path score samples corresponding to the similar planning path samples; a prediction score index acquisition unit, the prediction score index acquisition unit is used to predict the disease score according to the path score samples and obtain the prediction score index.

[0055] The specific configuration of the path planning model optimization module 50 will be described in detail below. As described above, the score difference index is recorded, and the score difference index is fed back to the joint information management system to optimize the path planning model. The path planning model optimization module 50 further includes: a difference threshold definition unit, the difference threshold definition unit is used to define a difference threshold, if the score difference index is greater than or equal to the difference threshold, the model parameters of the path planning model are optimized with a first feedback optimization path; a model parameter optimization unit, the model parameter optimization unit is used to optimize the model parameters of the path planning model with a second feedback optimization path if the score difference index is less than the difference threshold; a weight feedback adjustment unit, the weight feedback adjustment unit is used in which the feedback adjustment weight of the first feedback optimization path is greater than the feedback adjustment weight of the second feedback optimization path.

[0056] The medical information joint management system based on DRG and DIP provided in the embodiment of the present invention can execute the medical information joint management method based on DRG and DIP provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0057] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0058] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A joint management method of medical information based on DRG and DIP, characterized in that: The method comprises: Constructing a joint information management system, the joint information management system includes a DRG information system and a DIP information system, and the joint information management system is connected to a path planning model; Acquiring patient diagnosis and treatment information, wherein the pathway planning model acquires a first planned pathway and a patient score index of the first planned pathway according to the patient diagnosis and treatment information; Identify the patient's diagnosis and treatment information according to the DRG information system and determine DRG group information, wherein the DRG group information includes disease group and disease complexity; The DIP information system predicts the disease score of the DRG group information, obtains the predicted score index, compares the patient score index with the predicted score index, and outputs the score difference index; The score difference index is recorded, and the score difference index is fed back to the joint information management system to optimize the path planning model.

2. The method according to claim 1, characterized in that The method further comprises: Initialize the path planning model, including defining paths and path nodes, where each path represents all processes from the initial diagnosis and treatment to the end of the diagnosis and treatment, and the connections between nodes represent the diagnosis and treatment steps; Collect patient diagnosis and treatment sample information within a preset time period; Constructing an objective function, wherein the indicators of the objective function include diagnosis and treatment time, resource utilization and medical cost; The initialized path planning model is trained according to the objective function with the patient treatment sample information as input, with minimizing the objective function as a convergence condition, and a converged path planning model is output, and the converged path planning model is connected to the joint information management system.

3. The method according to claim 1, characterized in that Identifying the patient's diagnosis and treatment information according to the DRG information system and determining DRG group information, the method includes: Process the patient's diagnosis and treatment information in ICD coding format according to the DRG information system to obtain the patient's ICD code; Matching is performed according to the patient's ICD code to determine the disease group to which the patient belongs; Outputting the complexity of the disease and the complexity of resource consumption according to the patient's diagnosis and treatment information; Calculate the complexity of the disease based on the complexity of the disease and the complexity of resource consumption to determine the complexity of the disease; DRG group information is generated based on the disease groups and disease complexity.

4. The method according to claim 3, characterized in that The complexity of the disease is obtained by extracting the number of complications and the severity index of complications, as well as the number of concomitant diseases and the severity index of concomitant diseases from the patient's diagnosis and treatment information, and calculating the number of complications and the severity index of complications, as well as the number of concomitant diseases and the severity index of concomitant diseases; The resource consumption complexity is obtained by calculating the treatment time and medical equipment usage information of each node recorded from the patient diagnosis and treatment information.

5. The method according to claim 1, characterized in that The DIP information system predicts the disease score of the DRG group information and obtains the prediction score index. include: Wherein, the DIP information system includes a disease score prediction model, which is connected to the path planning model and is used to receive the historical planning path samples output by the path planning model, and the path scores corresponding to each historical planning path sample; The DIP information system obtains similar planned path samples of the DRG group information according to the DRG group information; Collecting path score samples corresponding to the similar planned path samples; The disease score is predicted based on the path score samples to obtain a predicted score index.

6. The method according to claim 1, characterized in that Feeding back the score difference index to the joint information management system to optimize the path planning model, the method includes: defining a difference threshold, and if the score difference index is greater than or equal to the difference threshold, optimizing the model parameters of the path planning model using the first feedback optimization path; If the score difference index is less than the difference threshold, optimizing the model parameters of the path planning model using a second feedback optimization path; The feedback adjustment weight of the first feedback optimization path is greater than the feedback adjustment weight of the second feedback optimization path.

7. The method according to claim 1, characterized in that After the path planning model acquires the first planned path according to the patient diagnosis and treatment information, the method further includes: Extracting a key planning path of the first planning path; Performing real-time monitoring on each node on the key planning path and outputting a real-time monitoring data set; Performing diagnosis and treatment delay risk analysis according to the real-time monitoring data set to obtain a delay risk indicator; When the delay risk indicator is greater than a preset risk threshold, the first planned path is optimized and a second planned path is output.

8. The medical information joint management system based on DRG and DIP is characterized by: The system is used to implement the medical information joint management method based on DRG and DIP according to any one of claims 1 to 7, and the system includes: A joint information management system construction module, wherein the joint information management system construction module is used to construct a joint information management system, wherein the joint information management system includes a DRG information system and a DIP information system, and wherein the joint information management system is connected to a path planning model; A first planned path acquisition module, the first planned path acquisition module is used to acquire patient diagnosis and treatment information, the path planning model acquires a first planned path and a patient score index of the first planned path according to the patient diagnosis and treatment information; A diagnosis and treatment information identification module, the diagnosis and treatment information identification module is used to identify the patient's diagnosis and treatment information according to the DRG information system, and determine DRG group information, the DRG group information includes disease group and disease complexity; A prediction score index acquisition module, which is used by the DIP information system to predict the disease score of the DRG group information, obtain the prediction score index, compare the patient score index with the prediction score index, and output the score difference index; A path planning model optimization module, wherein the path planning model optimization module is used to record the score difference index and feed the score difference index back to the joint information management system to optimize the path planning model.

Images