Performance scheduling method and device based on health record, equipment and storage medium

Abstract

The invention relates to the field of intelligent decision making, is applied to the field of financial science and technology businesses, and discloses a performance scheduling method, device and equipment based on health archives, and a storage medium, the method comprises the steps: scoring each to-be-selected hospital through a pre-trained target scoring model, and determining the hospital grade of each to-be-selected hospital; determining target DRG information through a preset DRG model and the user information; and generating a target performance scheduling strategy based on the hospital level of the to-be-selected hospital and the target DRG information. Through the above mode, the to-be-selected hospitals are scored through the pre-trained target scoring model, and by presetting the diagnosis related grouping DRG model and the user information, the medical requirements of the target user can be determined more accurately, the medical resources and service quality of the hospitals can be evaluated more accurately, the medical service process can be arranged more efficiently, and the medical service efficiency can be improved. Therefore, the performance efficiency of the user in the financial insurance field is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent decision-making and is applied to the field of financial technology business, and in particular to a performance scheduling method, device, equipment and storage medium based on health records. Background Art

[0002] In the modern financial insurance industry, contract fulfillment scheduling is a key link for insurance companies to provide high-quality services. With the increasing shortage of medical resources and the diversification of customer needs, traditional contract fulfillment scheduling methods have been unable to meet market demand. The existing contract fulfillment scheduling platform is mainly responsible for handling customers' outpatient appointment services. At present, when customers need outpatient appointment services, the contract fulfillment scheduling platforms of various insurance companies will select the nearest hospitals that meet the needs of customers according to their current geographical location. If it is a chronic disease, the operator will select the most authoritative and best hospital for appointment according to the type of disease of the customer. This method cannot perform specific analysis based on different populations, different diseases, different hospitals, and different service qualities to achieve the goal of reducing corporate operating costs. Therefore, how to improve the efficiency of contract fulfillment for users in the field of financial insurance has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides a method, device, equipment and storage medium for scheduling contract performance based on health records, so as to improve the contract performance efficiency for users in the financial insurance field.

[0004] In a first aspect, the present application provides a method for scheduling a contract based on a health record, the method comprising:

[0005] Scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital;

[0006] Acquire user information of a target user, and determine target DRG information of the target user by using a preset diagnosis related group (DRG) model and the user information;

[0007] Based on the hospital grade of each of the selected hospitals and the target DRG information, a target fulfillment scheduling strategy is generated.

[0008] In a second aspect, the present application further provides a health record-based performance scheduling device, the device comprising:

[0009] A hospital grade determination module, used to score each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model, and determine the hospital grade of each candidate hospital;

[0010] A target DRG information determination module is used to obtain user information of a target user and determine the target DRG information of the target user by using a preset diagnosis-related grouping DRG model and the user information;

[0011] The fulfillment scheduling strategy generation module is used to generate a target fulfillment scheduling strategy based on the hospital grade of each of the selected hospitals and the target DRG information.

[0012] In a third aspect, the present application also provides a computer device, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the above-mentioned health record-based performance scheduling method when executing the computer program.

[0013] In a fourth aspect, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the above-mentioned health record-based performance scheduling method.

[0014] The present application discloses a performance scheduling method, device, equipment and storage medium based on health records, wherein the performance scheduling method based on health records includes scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; obtaining the user information of the target user, and determining the target DRG information of the target user through a preset diagnosis-related grouping DRG model and the user information; generating a target performance scheduling strategy based on the hospital grade of each candidate hospital and the target DRG information. Through the above-mentioned method, the present application scores each candidate hospital through a pre-trained target scoring model, and through the preset diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the performance efficiency of users in the financial insurance field. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 is a schematic flow chart of a method for scheduling a contract based on a health record provided in the first embodiment of the present application;

[0017] Figure 2is a schematic flow chart of a method for scheduling contract performance based on health records provided in the second embodiment of the present application;

[0018] Figure 3 is a schematic flow chart of a method for scheduling contract performance based on health records provided in the third embodiment of the present application;

[0019] Figure 4 A schematic block diagram of a health record-based fulfillment scheduling device provided in an embodiment of the present application;

[0020] Figure 5 A schematic block diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0023] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0024] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0025] The embodiments of the present application provide a method, device, equipment and storage medium for fulfillment scheduling based on health records. Among them, the fulfillment scheduling method based on health records can be applied to a server, and each candidate hospital is scored through a pre-trained target scoring model. Through the preset diagnosis-related grouping DRG model and user information, the medical needs of the target user can be more accurately determined, the medical resources and service quality of each hospital can be more accurately evaluated, and the medical service process can be arranged more efficiently, thereby improving the performance efficiency of users in the financial insurance field. Among them, the server can be an independent server or a server cluster.

[0026] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0027] See also Figure 1 , Figure 1 This is a schematic flow chart of a health record-based performance scheduling method provided by the first embodiment of the present application. The health record-based performance scheduling method can be applied to a server to score each candidate hospital through a pre-trained target scoring model, and through the preset diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the performance efficiency of users in the financial insurance field.

[0028] like Figure 1 As shown, the health record-based fulfillment scheduling method specifically includes steps S10 to S30.

[0029] Step S10: Score each candidate hospital according to at least one preset evaluation dimension using a pre-trained target scoring model to determine the hospital grade of each candidate hospital;

[0030] Specifically, the pre-trained target scoring model is used to score each candidate hospital according to preset evaluation dimensions (such as medical quality, service level, facilities and equipment, etc.), and the grade of each hospital is determined based on the scoring results, which can refer to relevant standards.

[0031] Step S20: acquiring user information of a target user, and determining target DRG information of the target user by using a preset diagnosis related group (DRG) model and the user information;

[0032] Specifically, the user information of the target user is obtained, including but not limited to personal health records, historical medical records, etc.

[0033] Through the preset diagnosis-related group (DRG) model, combined with user information, the DRG information of the target user is determined. This can refer to the relevant standards in the hospital information interconnection and interoperability standardization maturity assessment to ensure the standardization and accuracy of the information.

[0034] Step S30: Generate a target fulfillment scheduling strategy based on the hospital grade of each of the to-be-selected hospitals and the target DRG information.

[0035] In one embodiment, the optimal fulfillment strategy is generated by an optimization algorithm (such as a genetic algorithm, simulated annealing, etc.) taking into account factors such as the hospital's service capabilities, the urgency of the user's needs, and the geographical location.

[0036] The present embodiment discloses a performance scheduling method, apparatus, device and storage medium based on health records, wherein the performance scheduling method based on health records includes scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; obtaining the user information of the target user, and determining the target DRG information of the target user through a preset diagnosis-related grouping DRG model and the user information; generating a target performance scheduling strategy based on the hospital grade of each candidate hospital and the target DRG information. Through the above-mentioned method, the present application scores each candidate hospital through a pre-trained target scoring model, and through the preset diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the performance efficiency for users in the financial insurance field.

[0037] based on Figure 1 In the embodiment shown, in this embodiment, before step S20, the following steps are included:

[0038] Acquiring historical medical records, wherein the historical medical records include historical diagnosis records, historical surgery records, historical operation codes and / or historical rehabilitation records;

[0039] Extracting historical medical key features of the historical medical records, encoding the historical medical key features, and generating medical record codes in a standardized format;

[0040] The initial DRG model is trained by cross-validation with the medical record coding to determine the preset DRG model.

[0041] In one embodiment, first, the patient's medical records need to be collected, which may include diagnosis records, surgical records, operation codes, and rehabilitation records. These data are usually stored in the hospital's electronic health record (EHR) system. Key features are extracted from historical medical records, which may include but are not limited to the patient's diagnosis information, treatment process, drug use, surgical details, etc.

[0042] The extracted key features are coded to generate medical record codes in a standardized format. This process can be automated by artificial intelligence medical coding software, such as using natural language processing (NLP) technology to identify and convert medical terms into standardized codes, such as I CD-10 or CPT codes. The initial DRG model is trained using a cross-validation method. Cross-validation is a statistical method used to evaluate and improve the generalization ability of a model. Common cross-validation methods include k-fold cross-validation (such as 10-fold cross-validation), which divides the dataset into k subsets, uses one of the subsets as a test set in turn, and the rest as a training set, repeats k times, and takes the average result as an estimate of the model performance.

[0043] Based on the results of cross-validation, adjust the parameters of the DRG model until the optimal model configuration is found. This process may involve adjusting the model's hyperparameters, such as learning rate, batch size, number of network layers, etc. After determining the optimal DRG model, the model needs to be evaluated to ensure its accuracy and reliability. Evaluation indicators may include accuracy, recall, F1 score, mean square error, etc.

[0044] The present embodiment discloses a fulfillment scheduling method, apparatus, device and storage medium based on health records. The fulfillment scheduling method based on health records includes scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; obtaining historical medical records, wherein the historical medical records include historical diagnosis records, historical surgical records, historical operation codes and / or historical rehabilitation records; extracting historical medical key features of the historical medical records, encoding the historical medical key features, and generating a medical record code in a standardized format; adjusting and training the initial DRG model through cross-validation with the medical record code to determine the preset DRG model; obtaining user information of the target user, and determining the target DRG information of the target user through a preset diagnosis-related grouping DRG model and the user information; generating a target fulfillment scheduling strategy based on the hospital grade of each candidate hospital and the target DRG information. Through the above method, the present application scores each candidate hospital through a pre-trained target scoring model. By presetting the diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the efficiency of fulfilling contracts with users in the financial insurance field.

[0045] In a more preferred embodiment, step S20 includes:

[0046] Performing data cleaning and data preprocessing operations on the user information, extracting valid information from the user information, and converting the valid information into valid information features;

[0047] Performing clustering processing on the effective information features through the clustering analysis algorithm built into the preset DRG model to determine the type of the effective information features;

[0048] Determine at least one DRG grouping information of the user information according to the type of the valid information feature;

[0049] The target DRG information is determined based on each of the DRG grouping information.

[0050] In one embodiment, effective information such as age, gender, diagnosis information, treatment process, etc. is extracted from the cleaned data, and this information will be used as features for subsequent DRG grouping. The extracted features are converted into a standardized format, such as converting the diagnosis information into the International Classification of Diseases (ICD) code. This step can be achieved through natural language processing technology to convert unstructured text information into structured codes.

[0051] The encoded features are clustered using the clustering algorithm in the preset DRG model. This can be achieved through machine learning or deep learning models, which divide users into different groups based on the similarity of features. Based on the clustering results and combined with the DRG grouping rules, the DRG grouping information corresponding to each user information is determined. This may involve consideration of factors such as resource consumption, treatment complexity, and case severity.

[0052] Finally, the DRG information of the target user is determined based on the DRG grouping information of all users. This can be done by comparing the resource consumption and treatment results of different DRG groups and selecting the DRG grouping that best suits the user's situation.

[0053] The present embodiment discloses a method, apparatus, device and storage medium for fulfillment scheduling based on health records. The method for fulfillment scheduling based on health records includes scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; obtaining historical medical records, wherein the historical medical records include historical diagnosis records, historical operation records, historical operation codes and / or historical rehabilitation records; extracting historical medical key features of the historical medical records, encoding the historical medical key features, and generating a medical record code in a standardized format; and cross-validating the medical record code with the medical record code to obtain the historical medical key features of the historical medical records. The initial DRG model is trained and adjusted to determine the preset DRG model; the user information is cleaned and preprocessed to extract valid information from the user information, and the valid information is converted into valid information features; the valid information features are clustered by the built-in clustering analysis algorithm in the preset DRG model to determine the type of the valid information features; at least one DRG grouping information of the user information is determined according to the type of the valid information features; the target DRG information is determined based on each DRG grouping information; based on the hospital grade of each of the selected hospitals and the target DRG information, a target fulfillment scheduling strategy is generated. Through the above-mentioned method, the present application scores each selected hospital through a pre-trained target scoring model, and through the preset diagnosis-related grouping DRG model and user information, the medical needs of the target user can be determined more accurately, the medical resources and service quality of each hospital can be evaluated more accurately, and the medical service process can be arranged more efficiently, thereby improving the performance efficiency of users in the financial insurance field.

[0054] See also Figure 2 , Figure 2This is a schematic flow chart of a health record-based performance scheduling method provided by the second embodiment of the present application. The health record-based performance scheduling method can be applied to a server to score each candidate hospital through a pre-trained target scoring model, and through the preset diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the performance efficiency of users in the financial insurance field.

[0055] Based on the above embodiments, this embodiment Figure 2 As shown, clustering processing is performed on the valid information features by using the built-in clustering analysis algorithm in the preset DRG model, and determining the type of the valid information features includes steps S2021 to S2025.

[0056] Step S2021, determining the number of types of the effective information features, and using the number of types as the K value of the K-Means algorithm;

[0057] Step S2022: randomly select one of the valid information features as the initial cluster center of the K-Means algorithm, and calculate the Euclidean distance between each of the valid information features and the initial cluster center;

[0058] Step S2023, iteratively updating the latest cluster center of the K-Means algorithm based on each of the Euclidean distances until the latest cluster center stops updating or the number of iterations is equal to the K value;

[0059] Step S2024, calculating the intra-cluster sum of squares of each of the latest cluster centers, and determining the latest cluster center with the smallest intra-cluster sum of squares as the target cluster center;

[0060] Step S2025: Determine the type of each cluster according to the target cluster center, and use the type of each cluster as the type of the effective information feature.

[0061] In one embodiment, the valid information features are analyzed to determine the number of feature types. This number of types will be used as the K value in the K-Means algorithm, that is, the number of clusters to be formed. A valid information feature is randomly selected as the initial cluster center of the K-Means algorithm. This selection is random, but it should be ensured that the selected feature can represent the diversity in the data set. For each valid information feature in the data set, the Euclidean distance between it and the initial cluster center is calculated. The Euclidean distance is the straight-line distance between feature vectors and can be calculated using the following formula: Among them, a and b are two eigenvectors, a i and b i is the i-th element in the vector.

[0062] According to the calculated Euclidean distance, each valid information feature is assigned to the nearest cluster center to form K clusters. For each cluster, the average value of all feature vectors in the cluster is calculated and the position of the cluster center is updated. This iterative process is repeated until the position of the cluster center no longer changes or the number of iterations reaches the preset K value. During the iteration process, the intra-cluster sum of squares (SSE) of each cluster center is calculated, that is, the sum of the squares of the distances from all points in the cluster to the cluster center. According to the target cluster center, the type of each cluster is determined. These types represent different patterns or groupings in the data set.

[0063] Assign the type of each cluster to each valid information feature in the cluster, so that each feature has a type label. The effectiveness of the clustering results can be evaluated by calculating indicators such as the silhouette coefficient and the Davies-Bouldin index. If the result is not satisfactory, you may need to return to step 1 to adjust the K value or reselect the initial cluster center and repeat the entire clustering process.

[0064] The present embodiment discloses a performance scheduling method, apparatus, device and storage medium based on health records. The performance scheduling method based on health records includes scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; obtaining historical medical records, wherein the historical medical records include historical diagnosis records, historical surgical records, historical operation codes and / or historical rehabilitation records; extracting historical medical key features of the historical medical records, encoding the historical medical key features, and generating medical record codes in a standardized format; adjusting and training the initial DRG model through cross-validation with the medical record codes to determine the preset DRG model; performing data cleaning and data preprocessing operations on the user information, extracting valid information from the user information, and converting the valid information into valid information features; determining the type number of the types of the valid information features. The method comprises the following steps: randomly selecting one of the valid information features as the initial cluster center of the K-Means algorithm, and calculating the Euclidean distance between each of the valid information features and the initial cluster center; iteratively updating the latest cluster center of the K-Means algorithm based on each of the Euclidean distances until the latest cluster center stops updating or the number of iterations is equal to the K value; calculating the intra-cluster sum of squares of each of the latest cluster centers, and determining the latest cluster center with the smallest intra-cluster sum of squares as the target cluster center; determining the type of each cluster according to the target cluster center, and using the type of each cluster as the type of the valid information feature; determining at least one DRG grouping information of the user information according to the type of the valid information feature; determining the target DRG information based on each of the DRG grouping information; and generating a target fulfillment scheduling strategy based on the hospital grade of each of the selected hospitals and the target DRG information. Through the above method, the present application scores each candidate hospital through a pre-trained target scoring model. By presetting the diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the efficiency of fulfilling contracts with users in the financial insurance field.

[0065] See also Figure 3 , Figure 3This is a schematic flow chart of a health record-based performance scheduling method provided by the third embodiment of the present application. The health record-based performance scheduling method can be applied to a server to score each candidate hospital through a pre-trained target scoring model, and through the preset diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the performance efficiency of users in the financial insurance field.

[0066] based on Figure 1 The embodiment shown, this embodiment Figure 3 As shown, step S30 includes steps S301 to S304.

[0067] Step S301, disassemble the target DRG information to determine the target medical needs of the target user;

[0068] Step S302: determining the medical resources of each of the hospitals to be selected based on the hospital grade of each of the hospitals to be selected;

[0069] Step S303: determine the hospital to be selected by Brother Yin that is the medical resource with the highest adaptability to the target medical need as the target hospital;

[0070] Step S304: Generate the target fulfillment scheduling strategy according to the preset fulfillment rules, the target hospital and the target medical needs.

[0071] In one embodiment, the DRG information of the target user is disassembled in detail to determine the user's medical needs. This includes analyzing key factors such as the types of diseases covered by the DRG grouping, the complexity of treatment, and resource consumption. Based on the hospital grade of each candidate hospital, the medical resources of each hospital are evaluated and determined. This includes the availability and quality of resources such as the hospital's facilities, professionals, number of beds, and medical equipment. The target user's medical needs are matched with the medical resources of the candidate hospitals, and the adaptability of each hospital to the user's medical needs is evaluated. The adaptability assessment can take into account factors such as the hospital's professional capabilities, past treatment experience, and patient satisfaction.

[0072] Based on the results of the suitability assessment, select the hospital with the highest suitability as the target hospital. This selection should ensure that users can get the services that best suit their medical needs. Develop a detailed fulfillment scheduling strategy based on the preset fulfillment rules, the resources of the target hospital, and the target medical needs. This includes appointment time, treatment plan, resource allocation, cost estimation, etc.

[0073] The present embodiment discloses a fulfillment scheduling method, apparatus, device and storage medium based on health records. The fulfillment scheduling method based on health records includes scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; obtaining user information of the target user, and determining the target DRG information of the target user through a preset diagnosis-related grouping DRG model and the user information; disassembling the target DRG information to determine the target medical needs of the target user; determining the medical resources of each candidate hospital based on the hospital grade of each candidate hospital; determining the candidate hospital with the medical resources that are most compatible with the target medical needs as the target hospital; and generating the target fulfillment scheduling strategy according to the preset fulfillment rules, the target hospital and the target medical needs. Through the above method, the present application scores each candidate hospital through a pre-trained target scoring model. By presetting the diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the efficiency of fulfilling contracts with users in the financial insurance field.

[0074] based on Figure 1 In the embodiment shown, in this embodiment, before step S10, the following steps are included:

[0075] Acquire at least one item of historical hospital data of each of the to-be-selected hospitals, wherein the historical hospital data includes historical operation data, historical evaluation data, historical medical quality data and / or historical financial data;

[0076] Based on each of the historical hospital data, determining at least one of the preset evaluation dimensions of the initial scoring model;

[0077] The initial scoring model is trained by using a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate the pre-trained target scoring model.

[0078] The present embodiment discloses a performance scheduling method, apparatus, device and storage medium based on health records. The performance scheduling method based on health records includes obtaining at least one historical hospital data of each of the candidate hospitals, wherein the historical hospital data includes historical operation data, historical evaluation data, historical medical quality data and / or historical financial data; based on each of the historical hospital data, determining at least one of the preset evaluation dimensions of the initial scoring model; training the initial scoring model through a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate a pre-trained target scoring model; scoring each of the candidate hospitals according to at least one preset evaluation dimension through the pre-trained target scoring model to determine the hospital grade of each of the candidate hospitals; obtaining user information of the target user, and determining the target DRG information of the target user through a preset diagnosis-related grouping DRG model and the user information; generating a target performance scheduling strategy based on the hospital grade of each of the candidate hospitals and the target DRG information. Through the above method, the present application scores each candidate hospital through a pre-trained target scoring model. By presetting the diagnosis-related grouping DRG model and user information, it can more accurately determine the medical needs of the target user, more accurately evaluate the medical resources and service quality of each hospital, and more efficiently arrange the medical service process, thereby improving the efficiency of fulfilling contracts with users in the financial insurance field.

[0079] In a more preferred embodiment, the initial scoring model is trained by a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate the pre-trained target scoring model, including:

[0080] Dividing each of the historical hospital data into training set data and validation set data, and determining initial parameters of the initial scoring model through the preset machine learning algorithm;

[0081] Based on each of the preset evaluation dimensions, the training set data and the initial scoring model, generating initial model output data;

[0082] The initial parameters are adjusted through the validation set data and the initial model output data to generate the pre-trained target scoring model.

[0083] See also Figure 4 , Figure 4 The embodiment of the present application provides a schematic block diagram of a health record-based performance scheduling device, which is used to execute the aforementioned health record-based performance scheduling method. The health record-based performance scheduling device can be configured on a server.

[0084] like Figure 4As shown, the health record-based fulfillment scheduling device 400 includes:

[0085] The hospital grade determination module 410 is used to score each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital;

[0086] The target DRG information determination module 420 is used to obtain the user information of the target user, and determine the target DRG information of the target user by using a preset diagnosis-related group DRG model and the user information;

[0087] The fulfillment scheduling strategy generation module 430 is used to generate a target fulfillment scheduling strategy based on the hospital grade of each of the candidate hospitals and the target DRG information.

[0088] Furthermore, the health record-based contract performance scheduling device 400 includes:

[0089] A historical medical record acquisition module, used to acquire historical medical records, wherein the historical medical records include historical diagnosis records, historical operation records, historical operation codes and / or historical rehabilitation records;

[0090] A medical record code generation module, used for extracting historical medical key features of the historical medical record, encoding the historical medical key features, and generating a medical record code in a standardized format;

[0091] The preset DRG model determination module is used to adjust the parameters of the initial DRG model through cross-validation and the medical record coding to determine the preset DRG model.

[0092] Furthermore, the target DRG information determination module 420 includes:

[0093] An effective information feature generating unit, used for performing data cleaning and data preprocessing operations on the user information, extracting effective information from the user information, and converting the effective information into effective information features;

[0094] An effective information feature type determination unit, configured to perform clustering processing on the effective information feature by using a clustering analysis algorithm built into the preset DRG model to determine the type of the effective information feature;

[0095] A DRG group information determining unit, configured to determine at least one DRG group information of the user information according to the type of the valid information feature;

[0096] A target DRG information determination unit is used to determine the target DRG information based on each of the DRG grouping information.

[0097] Furthermore, the effective information feature type determination unit includes:

[0098] A K value determination subunit, used to determine the number of types of the effective information features, and use the number of types as the K value of the K-Means algorithm;

[0099] A Euclidean distance calculation subunit, used for randomly selecting one of the effective information features as the initial cluster center of the K-Means algorithm, and calculating the Euclidean distance between each of the effective information features and the initial cluster center;

[0100] An iterative updating subunit, used for iteratively updating the latest cluster center of the K-Means algorithm based on each of the Euclidean distances, until the latest cluster center stops updating or the number of iterations is equal to the K value;

[0101] a target cluster center determination subunit, configured to calculate the intra-cluster sum of squares of the latest cluster centers, and determine the latest cluster center with the smallest intra-cluster sum of squares as the target cluster center;

[0102] The effective information feature type determination subunit is used to determine the type of each cluster according to the target cluster center, and use the type of each cluster as the type of the effective information feature.

[0103] Furthermore, the fulfillment scheduling strategy generation module 430 includes:

[0104] A target medical need determination unit, used to decompose the target DRG information and determine the target medical need of the target user;

[0105] A medical resource determination unit, configured to determine the medical resources of each of the hospitals to be selected based on the hospital grade of each of the hospitals to be selected;

[0106] A target hospital determination unit, configured to determine the hospital to be selected by Brother Yin that has the highest adaptability to the target medical need as the target hospital;

[0107] The target fulfillment scheduling strategy unit is used to generate the target fulfillment scheduling strategy according to the preset fulfillment rules, the target hospital and the target medical needs.

[0108] Furthermore, the health record-based contract performance scheduling device 400 includes:

[0109] A historical hospital data acquisition module, used to acquire at least one item of historical hospital data of each of the to-be-selected hospitals, wherein the historical hospital data includes historical operation data, historical evaluation data, historical medical quality data and / or historical financial data;

[0110] A preset evaluation dimension determination module, used to determine at least one preset evaluation dimension of an initial scoring model based on each of the historical hospital data;

[0111] The target scoring model generation module is used to train the initial scoring model through a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate a pre-trained target scoring model.

[0112] Furthermore, the target scoring model generation module includes:

[0113] An initial parameter determination unit, used for dividing each of the historical hospital data into training set data and validation set data, and determining the initial parameters of the initial scoring model through the preset machine learning algorithm;

[0114] An initial model output data generating unit, configured to generate initial model output data based on each of the preset evaluation dimensions, the training set data and the initial scoring model;

[0115] The target scoring model generating unit is used to adjust the initial parameters through the validation set data and the initial model output data to generate the pre-trained target scoring model.

[0116] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0117] The above-mentioned device can be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.

[0118] See also Figure 5 , Figure 5 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0119] See also Figure 5 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0120] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any one of the health record-based performance scheduling methods.

[0121] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0122] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any one of the performance scheduling methods based on the health record.

[0123] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0124] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0125] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0126] Scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital;

[0127] Acquire user information of a target user, and determine target DRG information of the target user by using a preset diagnosis related group (DRG) model and the user information;

[0128] Based on the hospital grade of each of the selected hospitals and the target DRG information, a target fulfillment scheduling strategy is generated.

[0129] In one embodiment, before obtaining the user information of the target user and determining the target DRG information of the target user by presetting the diagnosis related group (DRG) model and the user information, it is used to implement:

[0130] Acquiring historical medical records, wherein the historical medical records include historical diagnosis records, historical surgery records, historical operation codes and / or historical rehabilitation records;

[0131] Extracting historical medical key features of the historical medical records, encoding the historical medical key features, and generating medical record codes in a standardized format;

[0132] The initial DRG model is trained by cross-validation with the medical record coding to determine the preset DRG model.

[0133] In one embodiment, user information of a target user is obtained, and target DRG information of the target user is determined by using a preset diagnosis related group (DRG) model and the user information, so as to achieve:

[0134] Performing data cleaning and data preprocessing operations on the user information, extracting valid information from the user information, and converting the valid information into valid information features;

[0135] Performing clustering processing on the effective information features through the built-in clustering analysis algorithm in the preset DRG model to determine the type of the effective information features;

[0136] Determine at least one DRG grouping information of the user information according to the type of the valid information feature;

[0137] The target DRG information is determined based on each of the DRG grouping information.

[0138] In one embodiment, clustering processing is performed on the effective information features through the built-in clustering analysis algorithm in the preset DRG model to determine the type of the effective information features, so as to achieve:

[0139] Determine the number of types of the effective information features, and use the number of types as the K value of the K-Means algorithm;

[0140] Randomly select one of the effective information features as the initial cluster center of the K-Means algorithm, and calculate the Euclidean distance between each of the effective information features and the initial cluster center;

[0141] Iteratively updating the latest cluster center of the K-Means algorithm based on each of the Euclidean distances until the latest cluster center stops updating or the number of iterations is equal to the K value;

[0142] Calculating the intra-cluster sum of squares of the latest cluster centers, and determining the latest cluster center with the smallest intra-cluster sum of squares as the target cluster center;

[0143] The type of each cluster is determined according to the target cluster center, and the type of each cluster is used as the type of the effective information feature.

[0144] In one embodiment, based on the hospital grade of each of the selected hospitals and the target DRG information, a target fulfillment scheduling strategy is generated to achieve:

[0145] Decomposing the target DRG information to determine the target medical needs of the target user;

[0146] Determining the medical resources of each of the hospitals to be selected based on the hospital grade of each of the hospitals to be selected;

[0147] Determine the hospital to be selected by Brother Yin that has the highest adaptability to the target medical needs as the target hospital;

[0148] The target fulfillment scheduling strategy is generated according to the preset fulfillment rules, the target hospital and the target medical needs.

[0149] In one embodiment, the pre-trained target scoring model is used to score each candidate hospital according to at least one preset evaluation dimension, and before determining the hospital grade of each candidate hospital, it is used to achieve:

[0150] Acquire at least one item of historical hospital data of each of the hospitals to be selected, wherein the historical hospital data includes historical operation data, historical evaluation data, historical medical quality data and / or historical financial data;

[0151] Based on each of the historical hospital data, determining at least one of the preset evaluation dimensions of the initial scoring model;

[0152] The initial scoring model is trained by using a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate the pre-trained target scoring model.

[0153] In one embodiment, the initial scoring model is trained by a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate the pre-trained target scoring model, which is used to achieve:

[0154] Dividing each of the historical hospital data into training set data and validation set data, and determining initial parameters of the initial scoring model through the preset machine learning algorithm;

[0155] Based on each of the preset evaluation dimensions, the training set data and the initial scoring model, generating initial model output data;

[0156] The initial parameters are adjusted through the validation set data and the initial model output data to generate the pre-trained target scoring model.

[0157] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement any one of the health record-based performance scheduling methods provided in the embodiments of the present application.

[0158] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Medi a Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card (Flash Card), etc., equipped on the computer device.

[0159] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for scheduling contract performance based on health records, characterized in that: include: Scoring each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model to determine the hospital grade of each candidate hospital; Acquire user information of a target user, and determine target DRG information of the target user by using a preset diagnosis related group (DRG) model and the user information; Based on the hospital grade of each of the selected hospitals and the target DRG information, a target fulfillment scheduling strategy is generated.

2. The method for scheduling contract performance based on health records according to claim 1, characterized in that: The method of obtaining the user information of the target user and determining the target DRG information of the target user by using a preset diagnosis related group (DRG) model and the user information includes: Acquiring historical medical records, wherein the historical medical records include historical diagnosis records, historical surgery records, historical operation codes and / or historical rehabilitation records; Extracting historical medical key features of the historical medical records, encoding the historical medical key features, and generating medical record codes in a standardized format; The initial DRG model is trained by cross-validation with the medical record coding to determine the preset DRG model.

3. The method for scheduling contract performance based on health records according to claim 2, characterized in that: The acquiring the user information of the target user and determining the target DRG information of the target user by using a preset diagnosis related group (DRG) model and the user information includes: Performing data cleaning and data preprocessing operations on the user information, extracting valid information from the user information, and converting the valid information into valid information features; Performing clustering processing on the effective information features through the built-in clustering analysis algorithm in the preset DRG model to determine the type of the effective information features; Determine at least one DRG grouping information of the user information according to the type of the valid information feature; The target DRG information is determined based on each of the DRG grouping information.

4. The method for scheduling contract performance based on health records according to claim 3, characterized in that: The cluster analysis algorithm includes a K-Means algorithm, and the clustering process is performed on the effective information features by using the built-in cluster analysis algorithm in the preset DRG model to determine the type of the effective information features, including: Determine the number of types of the effective information features, and use the number of types as the K value of the K-Means algorithm; Randomly select one of the effective information features as the initial cluster center of the K-Means algorithm, and calculate the Euclidean distance between each of the effective information features and the initial cluster center; Iteratively updating the latest cluster center of the K-Means algorithm based on each of the Euclidean distances until the latest cluster center stops updating or the number of iterations is equal to the K value; Calculating the intra-cluster sum of squares of the latest cluster centers, and determining the latest cluster center with the smallest intra-cluster sum of squares as the target cluster center; The type of each cluster is determined according to the target cluster center, and the type of each cluster is used as the type of the effective information feature.

5. The method for scheduling contract performance based on health records according to claim 1, characterized in that: The generating of a target fulfillment scheduling strategy based on the hospital grade of each of the to-be-selected hospitals and the target DRG information includes: Decomposing the target DRG information to determine the target medical needs of the target user; Determining the medical resources of each of the hospitals to be selected based on the hospital grade of each of the hospitals to be selected; Determine the hospital to be selected by Brother Yin that has the highest adaptability to the target medical needs as the target hospital; The target fulfillment scheduling strategy is generated according to the preset fulfillment rules, the target hospital and the target medical needs.

6. The method for scheduling contract performance based on health records according to claim 1, characterized in that: Scoring each candidate hospital according to at least one preset evaluation dimension by the pre-trained target scoring model and determining the hospital grade of each candidate hospital includes: Acquire at least one item of historical hospital data of each of the hospitals to be selected, wherein the historical hospital data includes historical operation data, historical evaluation data, historical medical quality data and / or historical financial data; Based on each of the historical hospital data, determining at least one of the preset evaluation dimensions of the initial scoring model; The initial scoring model is trained by using a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate the pre-trained target scoring model.

7. The method for scheduling contract performance based on health records according to claim 6, characterized in that: The initial scoring model is trained by using a preset machine learning algorithm, each of the historical hospital data and each of the preset evaluation dimensions to generate the pre-trained target scoring model, including: Dividing each of the historical hospital data into training set data and validation set data, and determining initial parameters of the initial scoring model through the preset machine learning algorithm; Based on each of the preset evaluation dimensions, the training set data and the initial scoring model, generating initial model output data; The initial parameters are adjusted through the validation set data and the initial model output data to generate the pre-trained target scoring model.

8. A contract performance scheduling device based on health records, characterized in that: include: A hospital grade determination module, used to score each candidate hospital according to at least one preset evaluation dimension through a pre-trained target scoring model, and determine the hospital grade of each candidate hospital; A target DRG information determination module is used to obtain user information of a target user and determine the target DRG information of the target user by using a preset diagnosis-related grouping DRG model and the user information; The fulfillment scheduling strategy generation module is used to generate a target fulfillment scheduling strategy based on the hospital grade of each of the selected hospitals and the target DRG information.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the health record-based performance scheduling method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the health record-based performance scheduling method as described in any one of claims 1 to 7.

Images