DRG-based analysis control method for medical insurance comprehensive management and medical quality supervision

The method addresses inconsistent medical costs by using data-driven models to predict overspending and suggest treatment plans that align with DRG standards, optimizing medical expense management.

CN119092064AInactive Publication Date: 2024-12-06JIANGSU MR ZHI INFORMATION TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410982293.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The current medical insurance system faces challenges in controlling medical costs due to varying treatment plans for the same disease, leading to inconsistent medical expenses that can exceed DRG medical insurance standards, causing inefficiencies and irrational spending.

Method used

A method involving data preprocessing, disease association analysis, and symptom judgment models to predict medical overspending and provide treatment plans that meet DRG medical insurance standards, using disease and medical data to optimize treatment choices.

Benefits of technology

This approach improves the rationality of treatment plans by providing options that adhere to DRG standards, reducing unnecessary expenses and enhancing cost-effectiveness through data-driven decision-making.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119092064A_ABST
    Figure CN119092064A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a DRG-based medical insurance comprehensive management and medical quality supervision analysis control method, and belongs to the technical field of medical management, and the method comprises the steps: collecting, obtaining and sorting data, building a symptom correlation analysis model, calculating a DRG medical insurance standard value, predicting a medical service hyperbranched condition, obtaining a hyperbranched balance condition, and generating a medical hyperbranched model. The method comprises the following steps: performing comprehensive physical examination on a patient, establishing a symptom result judgment model, obtaining a disease to which the current physical symptom data of the patient belongs, obtaining a disease associated with the current disease by using a symptom association analysis model, providing a plurality of treatment schemes, inputting each treatment scheme into a medical hyperbranched model, and judging whether the treatment scheme meets a DRG medical insurance standard value or not. According to the embodiment of the invention, the treatment scheme conforming to the DRG medical insurance standard value is referenced and selected by the doctor and the patient, the treatment scheme can conform to the DRG medical insurance standard value, various treatment schemes for the patient and the doctor to select are given by combining associated diseases, and the reasonability of the treatment scheme is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the field of medical management technology, and specifically, relates to an analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG. Background Art

[0002] At present, the medical insurance system is an important livelihood security system in my country's social security system and an important guarantee for the people to enjoy reasonable medical and health services.

[0003] Sometimes for the same disease, different doctors may give different treatment plans. Even if these treatment plans can treat the disease, different treatment plans will bring different treatment costs, which will also bring different medical burdens to patients. At the same time, with the current reform trend of DRG / DIP payment methods, different treatment plans bring different treatment costs with different paths that are very different. For example, two doctors in the same department in the same hospital use two different medical plans when treating the same type of disease. The first doctor's medical plan may meet the DRG medical insurance standard value and will not cause medical insurance overspending, while the other doctor's medical plan will exceed the DRG medical insurance standard value, resulting in medical insurance overspending, which will cause a gap in the performance of the two doctors. Therefore, medical cost control has become a pain point for most medical institutions and medical workers. What are the standards for treatment plans, which treatment plans will cause medical insurance overspending, which are unreasonable expenses, and which drugs should not be prescribed are often troubled by clinicians and hospital managers. Summary of the invention

[0004] In order to solve the above problems and technical defects, the embodiment of the present application adopts the following technical solution, which is an analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG, including the following steps:

[0005] Step 1: Collect, obtain and organize various disease data, medical records, business data and medical insurance data, and establish a disease association analysis model based on the disease data and medical records;

[0006] Step 2: Calculate the DRG medical insurance standard value based on the disease association analysis model and medical insurance data, predict the medical service overspending situation based on the DRG medical insurance standard value, obtain the overspending surplus situation, and generate a medical overspending model based on the overspending surplus situation;

[0007] Step 3: Conduct a comprehensive physical examination on the patient, obtain the patient's current physical condition data and the patient's vital signs and physical behavior before and after treatment, obtain the patient's historical medical record data and record and save the patient's current physical condition data;

[0008] Step 4: Establish a symptom result judgment model, input the patient's current physical condition data into the symptom result judgment model, and obtain the disease to which the patient's current physical condition data belongs based on the symptom analysis and judgment results;

[0009] Step 5: using the disease association analysis model to obtain the disease associated with the current disease according to the disease to which the patient's current physical disease data belongs;

[0010] Step 6. Provide multiple treatment plans based on the current symptoms and related symptoms, input each treatment plan into the medical overspending model, determine whether the treatment plan meets the DRG medical insurance standard value, and allow doctors and patients to refer to and select treatment plans that meet the DRG medical insurance standard value.

[0011] Preferably, when collecting, acquiring and organizing various disease data, medical records, business data and medical insurance data, these data need to be preprocessed, and the preprocessing steps are as follows:

[0012] S101, clean the data and complete the processing of missing values ​​and outliers;

[0013] S102, extracting features by processing text and time series data of disease entity;

[0014] S103, performing standardization and normalization processing to eliminate the influence of different feature dimensions and scale the extracted features to the same scale;

[0015] S104. Encode and classify the data. When the categorical variable has no order, each category is converted into a new binary column. When the categorical variable has a clear order, the category is assigned an integer that reflects the order relationship between the categories.

[0016] S105 , dividing the encoded and classified data set according to a preset ratio, and dividing the data set into a training set and a test set.

[0017] Furthermore, after the preprocessing is completed, it is necessary to use visualization methods and descriptive statistics to evaluate the preprocessed data, obtain the quality of the preprocessed data and the effectiveness of the features, and adjust the preprocessing strategy according to the evaluation results.

[0018] Preferably, the steps of establishing the disease association analysis model are as follows:

[0019] S201, obtaining symptom information of each disease, performing data mining and data analysis on the symptom information, and obtaining similarities and correlations between the symptom information;

[0020] S202, acquiring other symptoms associated with the target symptom according to the similarity and association of the symptom information of the target symptom, and using the other symptoms as associated symptoms of the target symptom;

[0021] S203, setting a symptom node for each symptom, filling the symptom information of each symptom into the symptom node, establishing a node link between the symptom node of the target symptom and the symptom node of the associated symptom, and the symptom information of each symptom corresponds to a symptom node one by one.

[0022] Furthermore, the node links are divided into direct links, indirect links and potential links. When a certain symptom is causally related to a target symptom, the node link of the symptom nodes between the symptom and the target symptom is a direct link. When there is no causal relationship between a certain symptom and the target symptom, but there are two or more direct links between the symptom node of the symptom and the symptom node of the target symptom, the node link of the symptom nodes between the symptom and the target symptom is an indirect link. When there is neither a causal relationship nor an indirect link between a certain symptom and the target, medical experts shall set a potential link between the symptom and the target symptom based on actual conditions.

[0023] Preferably, the historical medical record data is the patient's medical record data collected through multiple channels, including the patient's personal information, diagnosis results, medical history, treatment process, symptoms and medication records, and is finally confirmed by the patient to supplement and modify the medical record data.

[0024] Preferably, the judgment process of the symptom result judgment model is as follows:

[0025] S301, obtaining the symptom characteristics of each disease;

[0026] S302, calculating the symptom characteristics of the patient's current physical symptom data and the symptom characteristics of the symptom information of each disease, to obtain a feature similarity vector between the current target symptom and each disease;

[0027] S303, taking the maximum value of the feature similarity vectors between the target symptom and each disease, and obtaining the disease to which the patient's current physical symptom data belongs according to the disease to which the maximum value belongs.

[0028] Preferably, providing multiple treatment plans based on the current symptoms and related symptoms is to propose multiple treatment plans for the current symptoms based on medical records combined with the patient's physical data, and each treatment plan includes corresponding medical time and cost requirements based on the medical expenses and medical insurance data in the symptom data. At the same time, based on the incidence of related symptoms, prevention plans for related symptoms and corresponding medical time and cost requirements are provided.

[0029] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG as described in any one of the above items are implemented.

[0030] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG as described in any of the above items.

[0031] Compared with the prior art, the beneficial effects of the embodiments of the present application are:

[0032] This application collects, acquires and organizes various disease data, medical records, business data and medical insurance data, establishes a disease association analysis model based on the disease data and medical records, calculates the DRG medical insurance standard value based on the disease association analysis model and medical insurance data, and thus obtains a medical overspending model. After obtaining the symptoms to which the patient's current physical disease data belongs, the disease association analysis model is used to obtain the symptoms associated with the current symptoms. Not only does it provide treatment plans for the patient's current physical symptoms, but it also uses the medical overspending model to judge the treatment plans so that the treatment plans meet the DRG medical insurance standard values. By combining related symptoms, a variety of treatment plans are given for patients and doctors to choose from, thereby improving the rationality of treatment plans. Treatment plans can be given without the doctor's manual diagnosis. Treatment plans that meet the DRG medical insurance standard values ​​can also correct unreasonable medical expenses and improve the cost-effectiveness of medical plan costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In the attached picture:

[0034] Figure 1 Schematic diagram of the method steps of an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0036] Example 1

[0037] like Figure 1 As shown, the analysis and control method of comprehensive medical insurance management and medical quality supervision based on DRG includes the following steps:

[0038] Collect, obtain and organize various disease data, medical records, business data and medical insurance data, and establish disease association analysis models based on disease data and medical records;

[0039] When collecting, acquiring and organizing various disease data, medical records, business data and medical insurance data, these data need to be preprocessed. The preprocessing steps are as follows:

[0040] S101, clean the data and complete the processing of missing values ​​and outliers;

[0041] S102, extracting features by processing text and time series data of disease entity;

[0042] S103, performing standardization and normalization processing to eliminate the influence of different feature dimensions and scale the extracted features to the same scale;

[0043] S104. Encode and classify the data. When the categorical variable has no order, each category is converted into a new binary column. When the categorical variable has a clear order, the category is assigned an integer that reflects the order relationship between the categories.

[0044] S105 , dividing the encoded and classified data set according to a preset ratio, and dividing the data set into a training set and a test set.

[0045] After the preprocessing is completed, it is necessary to use visualization methods and descriptive statistics to evaluate the preprocessed data to obtain the quality of the preprocessed data and the effectiveness of the features, and adjust the preprocessing strategy based on the evaluation results.

[0046] The steps for establishing the disease association analysis model are as follows:

[0047] S201, obtaining symptom information of each disease, performing data mining and data analysis on the symptom information, and obtaining similarities and correlations between the symptom information;

[0048] S202, acquiring other symptoms associated with the target symptom according to the similarity and association of the symptom information of the target symptom, and using the other symptoms as associated symptoms of the target symptom;

[0049] S203, setting a symptom node for each symptom, filling the symptom information of each symptom into the symptom node, establishing a node link between the symptom node of the target symptom and the symptom node of the associated symptom, and the symptom information of each symptom corresponds to a symptom node one by one.

[0050] Node links are divided into direct links, indirect links and potential links. When a certain symptom is causally related to a target symptom, the node link of the symptom nodes between the symptom and the target symptom is a direct link. When there is no causal relationship between a certain symptom and the target symptom, but there are two or more direct links between the symptom node of the symptom and the symptom node of the target symptom, the node link of the symptom nodes between the symptom and the target symptom is an indirect link. When there is neither a causal relationship nor an indirect link between a certain symptom and the target, medical experts will set a potential link between the symptom and the target symptom based on the actual situation.

[0051] The DRG medical insurance standard value is calculated based on the disease association analysis model and medical insurance data, the medical business overspending situation is predicted based on the DRG medical insurance standard value, the overspending surplus situation is obtained, and the medical overspending model is generated based on the overspending surplus situation.

[0052] Conduct a comprehensive physical examination on the patient, obtain the patient's current physical condition data and the patient's vital signs and physical behavior before and after treatment, obtain the patient's historical medical record data and record and preserve the patient's current physical condition data.

[0053] Historical medical record data is the medical record data of patients collected through multiple channels, including the patient's personal information, diagnosis results, medical history, treatment process, symptoms and medication records, and finally confirmed by the patient to supplement and modify the medical record data.

[0054] A symptom-result judgment model is established, and the patient's current physical condition data is input into the symptom-result judgment model. The disease to which the patient's current physical condition data belongs is obtained based on the symptom analysis and judgment results.

[0055] The judgment process of the symptom-result judgment model is as follows:

[0056] S301, obtaining the symptom characteristics of each disease;

[0057] S302, calculating the symptom characteristics of the patient's current physical symptom data and the symptom characteristics of the symptom information of each disease, to obtain a feature similarity vector between the current target symptom and each disease;

[0058] S303, taking the maximum value of the feature similarity vectors between the target symptom and each disease, and obtaining the disease to which the patient's current physical symptom data belongs according to the disease to which the maximum value belongs.

[0059] According to the disease to which the patient's current physical disease data belongs, a disease association analysis model is used to obtain the disease associated with the current disease.

[0060] Provide multiple treatment options based on current symptoms and related symptoms, input each treatment option into the medical overspending model to determine whether the treatment option meets the DRG medical insurance standard value, and doctors and patients will reference and select treatment options that meet the DRG medical insurance standard value.

[0061] Providing multiple treatment plans based on current symptoms and related symptoms means proposing multiple treatment plans for the current symptoms based on medical records and the patient's physical data. According to the medical expenses and medical insurance data in the disease data, each treatment plan includes corresponding medical time and cost requirements. At the same time, based on the incidence of related symptoms, prevention plans for related symptoms and corresponding medical time and cost requirements are provided.

[0062] Example 2

[0063] Acquire the symptom information of each disease, perform data mining and data analysis on the symptom information, and obtain the similarity and correlation between the symptom information; acquire other diseases associated with the target disease according to the similarity and correlation of the symptom information of the target disease, and use the other diseases as the associated diseases of the target disease; set up a disease node for each disease, fill the symptom information of each disease into the disease node, establish a node link between the disease node of the target disease and the disease node of the associated disease, and the symptom information of each disease corresponds to a disease node one by one.

[0064] Node links are divided into direct links, indirect links and potential links.

[0065] When a certain symptom is causally related to a target symptom, the node link of the symptom node between the symptom and the target symptom is a direct link.

[0066] For example, the relationship between viral influenza and pneumonia. Assuming that viral influenza is symptom A and pneumonia is symptom B, symptom A may cause a high fever and induce symptom B, or it may just be a low fever and not cause symptom B. Similarly, symptom B is not necessarily caused by symptom A, but may be caused by other symptoms. However, symptom A does have the possibility of causing symptom B, so at this time, symptom A and symptom B are causally related, and there is a direct link between symptom node A and symptom node B.

[0067] When a patient has symptom A, the doctor will focus on the examination of symptom B based on these data to avoid only treating symptom A while ignoring the occurrence of symptom B.

[0068] When a patient suffers from symptom B, the doctor will check whether symptom A causes symptom B, so as to accurately find out the cause of the symptom and achieve targeted treatment effects.

[0069] When there is no causal relationship between a certain symptom and a target symptom, but there are two or more direct links between the symptom node of the symptom and the symptom node of the target symptom, the node link between the symptom node of the symptom and the target symptom is an indirect link.

[0070] Similarly, when a patient suffers from viral influenza, that is, symptom A, it is possible that the autoimmune system will be disrupted, that is, symptom C, and symptom C will add new allergies, which is symptom D. For example, some people are not allergic to cat hair and dog hair, but after having a high fever and symptom C caused by symptom A, they will become allergic to cat hair and dog hair. At this time, there is a direct link between the A symptom node and the C symptom node, a direct link between the C symptom node and the D symptom node, and an indirect link between the A symptom node and the D symptom node.

[0071] When a patient suffers from symptom A, the doctor can inform the patient based on the above data before symptom D occurs, so that the patient can understand and take preventive measures.

[0072] If the patient did not originally have symptom D but now has symptom D, the doctor can obtain the cause of symptom D based on the patient's historical medical data and can also check for other allergens.

[0073] When there is neither a causal relationship nor an indirect link between a certain disease and the target, medical experts will set up a potential link between the disease and the target disease based on the actual situation.

[0074] There are fewer samples between some diseases. Doctors speculate that there is a certain relationship between the two diseases based on experience. However, the medical field has not yet studied a definite relationship between the two diseases. Therefore, the causal relationship between the two diseases is not yet determined. Only potential links are used, which has a certain reference value for doctors and patients.

[0075] Conduct a comprehensive physical examination on the patient, obtain the patient's current physical condition data, obtain the patient's historical medical record data and record and preserve the patient's current physical condition data;

[0076] Establish a symptom result judgment model, input the patient's current physical condition data into the symptom result judgment model, and obtain the disease to which the patient's current physical condition data belongs based on the symptom analysis and judgment results;

[0077] According to the disease to which the patient's current physical disease data belongs, a disease association analysis model is used to obtain the disease associated with the current disease;

[0078] Historical medical record data is the medical record data of patients collected through multiple channels, including the patient's personal information, diagnosis results, medical history, treatment process, symptoms, medication records, etc., and finally confirmed by the patient to supplement and modify the medical record data. For example, the patient had symptom A before, but the symptom reaction was not obvious, and the patient did not realize that he had symptom A, so he did not come to the hospital for treatment, so the hospital had no treatment data. However, when the patient came to the hospital for treatment because of symptom D, the doctor could infer the cause of symptom D based on the patient's oral supplement, the patient's physical examination data and the disease association analysis model, and find out the cause of the disease and prescribe the right medicine.

[0079] Provide a variety of treatment options based on the current symptoms and related symptoms, input each treatment option into the medical overspending model to determine whether the treatment option meets the DRG medical insurance standard value. The treatment option that meets the DRG medical insurance standard value is selected by doctors and patients for reference. Providing a variety of treatment options is to propose a variety of treatment options for the current symptoms based on medical records combined with the patient's physical data. Each treatment option includes corresponding medical time and cost requirements based on the medical expenses in the symptom data and the medical insurance data. At the same time, based on the incidence of related symptoms, provide prevention options for related symptoms and corresponding medical time and cost requirements. For example, when a patient is detected with symptom D, provide a variety of options, such as choosing to get an anti-allergy injection, or choose to stay away from allergens, that is, reduce contact with cats and dogs, or choose to feed anti-allergic pet food to cats and dogs to reduce allergens secreted by pets. Patients can choose one or more treatment options according to their needs.

[0080] The patient wears the medical device, which monitors and collects data on the patient's vital signs, physical behavior and device status before and after treatment in real time, determines the changes in various body data before and after treatment, and then compares the data changes with the symptom data of related diseases to determine whether the patient has induced related diseases.

[0081] Example 3

[0082] Medical data cleaning begins with missing value processing. First, through data exploration, we identify which features contain missing values ​​and the proportion of missing values. For continuous variables, we use the mean, median, or predicted values ​​from regression models based on other variables to fill in missing values. For categorical variables, we can use the mode or prediction model. If the number of missing values ​​for a feature exceeds the preset proportion, we can delete the feature. If some data rows are seriously missing, we can also delete these rows.

[0083] Then perform outlier processing, identify outliers through descriptive statistical maximum and minimum values ​​and visualization methods, and decide whether to delete outliers, replace them with other values, or mark them for subsequent analysis based on business understanding.

[0084] Feature extraction is to first process text data, extract keywords or phrases from the text as features, and identify specific entities in the text, such as disease names or drug names, and use these entities as important features.

[0085] At the same time, data processing is performed based on the time series, converting the timestamp into a more analytically valuable form, such as year, month, day, day of the week, time of the day, etc., and calculating the intervals between events in the time series as an important feature for predicting future events.

[0086] Data normalization is done by eliminating the effects of different dimensions so that each feature contributes equal importance, calculating the mean and standard deviation of each feature, and then using these parameters to transform the data into a distribution with zero mean and unit variance.

[0087] Data normalization is the process of scaling features to the same scale. For each feature, subtract the minimum value from each value and then divide by the difference between the maximum and minimum values.

[0088] And encode categorical data. When the categorical variable has no order, each category is converted into a new binary column, where 1 indicates that the category appears and 0 indicates that it does not appear. When the categorical variable has a clear order, the category is assigned an integer that reflects the sequential relationship between the categories.

[0089] Divide the data set into a training set and a test set to evaluate the generalization ability of the model. Randomly divide the data set according to a certain preset ratio to ensure that the training set and the test set are similar in feature distribution to avoid overfitting or underfitting of the model.

[0090] During the implementation of data preprocessing, it is necessary to adjust and verify the effectiveness of the preprocessing strategy multiple times, perform iterations, pay attention to the relationship between data quality and analysis objectives, and ensure that the preprocessing steps can effectively support subsequent data analysis or model training.

[0091] Iterative optimization is to evaluate the quality of data and the effectiveness of features through visualization methods and descriptive statistics after each preprocessing, and adjust the preprocessing strategy according to the evaluation results; during the model training phase, if the model performance is found to be poor, it is necessary to return to the data preprocessing phase to adjust feature selection, data cleaning or other preprocessing steps.

[0092] You need to ensure that all data processing steps are consistent across the entire dataset, especially when dealing with large-scale data. Even after initial data cleaning, you should regularly check whether new outliers or errors have appeared in the data.

[0093] Example 4

[0094] From the hardware level, the present application provides an embodiment of an electronic device that includes all or part of the contents of the analysis and control method of DRG-based comprehensive medical insurance management and medical quality supervision, and the electronic device specifically includes the following contents:

[0095] Step 1: Collect, obtain and organize various disease data, medical records, business data and medical insurance data, and establish a disease association analysis model based on the disease data and medical records;

[0096] Step 2: Calculate the DRG medical insurance standard value based on the disease association analysis model and medical insurance data, predict the medical service overspending situation based on the DRG medical insurance standard value, obtain the overspending surplus situation, and generate a medical overspending model based on the overspending surplus situation;

[0097] Step 3: Conduct a comprehensive physical examination on the patient, obtain the patient's current physical condition data and the patient's vital signs and physical behavior before and after treatment, obtain the patient's historical medical record data and record and save the patient's current physical condition data;

[0098] Step 4: Establish a symptom result judgment model, input the patient's current physical condition data into the symptom result judgment model, and obtain the disease to which the patient's current physical condition data belongs based on the symptom analysis and judgment results;

[0099] Step 5: using the disease association analysis model to obtain the disease associated with the current disease according to the disease to which the patient's current physical disease data belongs;

[0100] Step 6. Provide multiple treatment plans based on the current symptoms and related symptoms, input each treatment plan into the medical overspending model, determine whether the treatment plan meets the DRG medical insurance standard value, and allow doctors and patients to refer to and select treatment plans that meet the DRG medical insurance standard value.

[0101] Example 5

[0102] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG in the above-mentioned embodiments, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG in the above-mentioned embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0103] Step 1: Collect, obtain and organize various disease data, medical records, business data and medical insurance data, and establish a disease association analysis model based on the disease data and medical records;

[0104] Step 2: Calculate the DRG medical insurance standard value based on the disease association analysis model and medical insurance data, predict the medical service overspending situation based on the DRG medical insurance standard value, obtain the overspending surplus situation, and generate a medical overspending model based on the overspending surplus situation;

[0105] Step 3: Conduct a comprehensive physical examination on the patient, obtain the patient's current physical condition data and the patient's vital signs and physical behavior before and after treatment, obtain the patient's historical medical record data and record and save the patient's current physical condition data;

[0106] Step 4: Establish a symptom result judgment model, input the patient's current physical condition data into the symptom result judgment model, and obtain the disease to which the patient's current physical condition data belongs based on the symptom analysis and judgment results;

[0107] Step 5: using the disease association analysis model to obtain the disease associated with the current disease according to the disease to which the patient's current physical disease data belongs;

[0108] Step 6. Provide multiple treatment plans based on the current symptoms and related symptoms, input each treatment plan into the medical overspending model, determine whether the treatment plan meets the DRG medical insurance standard value, and allow doctors and patients to refer to and select treatment plans that meet the DRG medical insurance standard value.

[0109] The above-mentioned embodiments only express the preferred implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for ordinary technicians in this field, several modifications, improvements and substitutions can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. The analysis and control method of comprehensive medical insurance management and medical quality supervision based on DRG is characterized by: The following steps are involved: Step 1: Collect, obtain and organize various disease data, medical records, business data and medical insurance data, and establish a disease association analysis model based on the disease data and medical records; Step 2: Calculate the DRG medical insurance standard value based on the disease association analysis model and medical insurance data, predict the medical service overspending situation based on the DRG medical insurance standard value, obtain the overspending surplus situation, and generate a medical overspending model based on the overspending surplus situation; Step 3: Conduct a comprehensive physical examination on the patient, obtain the patient's current physical condition data and the patient's vital signs and physical behavior before and after treatment, obtain the patient's historical medical record data and record and save the patient's current physical condition data; Step 4: Establish a symptom result judgment model, input the patient's current physical condition data into the symptom result judgment model, and obtain the disease to which the patient's current physical condition data belongs based on the symptom analysis and judgment results; Step 5: using the disease association analysis model to obtain the disease associated with the current disease according to the disease to which the patient's current physical disease data belongs; Step 6. Provide multiple treatment plans based on the current symptoms and related symptoms, input each treatment plan into the medical overspending model, determine whether the treatment plan meets the DRG medical insurance standard value, and allow doctors and patients to refer to and select treatment plans that meet the DRG medical insurance standard value.

2. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 1 is characterized in that: When collecting, acquiring and organizing various disease data, medical records, business data and medical insurance data, these data need to be preprocessed. The preprocessing steps are as follows: S101, clean the data and complete the processing of missing values ​​and outliers; S102, extracting features by processing text and time series data of disease entity; S103, performing standardization and normalization processing to eliminate the influence of different feature dimensions and scale the extracted features to the same scale; S104. Encode and classify the data. When the categorical variable has no order, each category is converted into a new binary column. When the categorical variable has a clear order, the category is assigned an integer that reflects the order relationship between the categories. S105 , dividing the encoded and classified data set according to a preset ratio, and dividing the data set into a training set and a test set.

3. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 2 is characterized in that: After the preprocessing is completed, it is necessary to use visualization methods and descriptive statistics to evaluate the preprocessed data, obtain the quality of the preprocessed data and the effectiveness of the features, and adjust the preprocessing strategy according to the evaluation results.

4. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 1 is characterized in that: The steps for establishing the disease association analysis model are as follows: S201, obtaining symptom information of each disease, performing data mining and data analysis on the symptom information, and obtaining similarities and correlations between the symptom information; S202, acquiring other symptoms associated with the target symptom according to the similarity and association of the symptom information of the target symptom, and using the other symptoms as associated symptoms of the target symptom; S203, setting a symptom node for each symptom, filling the symptom information of each symptom into the symptom node, establishing a node link between the symptom node of the target symptom and the symptom node of the associated symptom, and the symptom information of each symptom corresponds to a symptom node one by one.

5. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 4 is characterized in that: The node links are divided into direct links, indirect links and potential links. When a certain symptom is causally related to a target symptom, the node link of the symptom nodes between the symptom and the target symptom is a direct link. When there is no causal relationship between a certain symptom and the target symptom, but there are two or more direct links between the symptom node of the symptom and the symptom node of the target symptom, the node link of the symptom nodes between the symptom and the target symptom is an indirect link. When there is neither a causal relationship nor an indirect link between a certain symptom and the target, medical experts shall set a potential link between the symptom and the target symptom based on actual conditions.

6. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 1 is characterized in that: The historical medical record data is the medical record data of the patient collected through multiple channels, including the patient's personal information, diagnosis results, medical history, treatment process, symptoms and medication records, and is ultimately confirmed by the patient to supplement and modify the medical record data.

7. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 1 is characterized in that: The judgment process of the symptom result judgment model is as follows: S301, obtaining the symptom characteristics of each disease; S302, calculating the symptom characteristics of the patient's current physical symptom data and the symptom characteristics of the symptom information of each disease, to obtain a feature similarity vector between the current target symptom and each disease; S303, taking the maximum value of the feature similarity vectors between the target symptom and each disease, and obtaining the disease to which the patient's current physical symptom data belongs according to the disease to which the maximum value belongs.

8. The analysis and control method for comprehensive medical insurance management and medical quality supervision based on DRG according to claim 1 is characterized in that: The provision of multiple treatment options based on the current symptoms and related symptoms is to propose multiple treatment options for the current symptoms based on medical records combined with the patient's physical data, and to obtain each treatment option including corresponding medical time and cost requirements based on the medical expenses and medical insurance data in the symptom data. At the same time, based on the incidence of related symptoms, prevention plans for related symptoms and corresponding medical time and cost requirements are provided.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements the steps of the analysis and control method for DRG-based comprehensive medical insurance management and medical quality supervision as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the analysis and control method for DRG-based comprehensive medical insurance management and medical quality supervision are implemented.

Citation Information

Patent Citations

  • Associated disease category analysis method and device, equipment and storage medium

    CN113972008A

  • Modeled DRGs clinical path planning management information system and method

    CN115482921A

  • DRG-based medical expense supervision and management and diagnosis and treatment analysis and control method

    CN117274007A

  • Price comparison method, system and device for DRG paid drug treatment scheme and storage medium

    CN117314490A

  • Medical scheme recommendation method and device based on feature analysis

    CN117809823A