Quality control method of medical service, electronic equipment and storage medium
By receiving medical execution data and using large language models to generate and update quality control rules, the problem of manual maintenance cannot be updated in time is solved, and the automated quality control and accuracy of medical services are improved.
Patent Information
- Application Number
- CN202411675012.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-21
- Publication Date
- 2025-03-07
AI Technical Summary
The manual maintenance business quality control rules cannot be updated in time, resulting in poor accuracy of medical process monitoring and evaluation.
By receiving medical execution data uploaded by medical staff, using preset large language models and standardized publishing platforms to generate and update business quality control rules, conduct on-site quality monitoring, and push quality control reminders to medical terminals.
It realizes automated quality control of medical business, improves the efficiency of generation and update of quality control rules, and enhances the accuracy of real-time monitoring and evaluation of the medical process.
Smart Images
Figure CN119580971A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a quality control method, electronic device and storage medium for medical services. Background Art
[0002] Using information technology to better serve the improvement of medical quality and ensure the safe diagnosis and treatment of patients is a topic that must be paid attention to at every stage of medical informatization. Among them, achieving closed-loop management of hospital business is the key focus to ensure medical quality and patient safety.
[0003] At present, manually maintained business quality control plans are often used to monitor and evaluate whether the medical process complies with regulations in order to achieve closed-loop management of medical services.
[0004] However, manually maintained business quality control rules cannot be updated in a timely manner, resulting in poor accuracy in monitoring and evaluating the medical process. Summary of the invention
[0005] In view of this, the embodiments of the present application provide a quality control method, electronic device and storage medium for medical services to solve the problem that manually maintained service quality control rules cannot be updated in a timely manner, resulting in poor accuracy in monitoring and evaluating medical processes.
[0006] In a first aspect, an embodiment of the present application provides a quality control method for medical services, including:
[0007] Receiving medical execution data uploaded by medical staff through a medical terminal, wherein the medical execution data is data generated by the medical staff performing a business operation corresponding to a target business node under a preset medical business;
[0008] According to the pre-generated business quality control rules for the target business node, the medical execution data is subjected to in-process quality monitoring to obtain in-process quality control results, wherein the business quality control rules are obtained by using a preset large language model based on the current medical specification information for the preset medical business stored in a preset specification publishing platform;
[0009] If the in-process quality control result indicates that the medical execution data does not meet the business quality control rules of the target business node, an in-process quality control reminder is pushed to the medical terminal.
[0010] In an optional implementation, the method further includes: performing in-process quality monitoring on the medical execution data according to the pre-generated business quality control rules for the target business node, before obtaining the in-process quality control results:
[0011] Acquire the current medical standard information from the preset standard publishing platform;
[0012] According to the current medical standard information, a preset large language model is used to generate business quality control rules for multiple business nodes under the preset medical business.
[0013] In an optional implementation, the generating of the service quality control rules of the plurality of service nodes under the preset medical service by using the preset large language model according to the current medical specification information includes:
[0014] Generate a prompt word according to the current medical specification information and preset rule type configuration information, wherein the preset rule type configuration information is used to indicate the quality control rule type;
[0015] Inputting the prompt word into the preset large language model;
[0016] The preset large language model is used to extract medical specification data for each business node for the quality control rule type from the current medical specification information based on the prompt word, and the business quality control rules for each business node for the quality control rule type are generated based on the medical specification data.
[0017] In an optional implementation, generating a prompt word according to the current medical specification information and preset rule type configuration information includes:
[0018] Generate the prompt word according to the current medical standard information, the preset rule type configuration information and the preset quality control rule data model;
[0019] The method of using the preset large language model based on the prompt word to extract medical specification data for each business node for the quality control rule type from the current medical specification information, and generating a business quality control rule for each business node for the quality control rule type based on the medical specification data, includes:
[0020] The preset large language model is used to extract the medical specification data of each business node for the quality control rule type from the current medical specification information based on the prompt word, and the business quality control rules for each business node for the quality control rule type are generated based on the medical specification data and the preset quality control rule data model.
[0021] In an optional implementation, the preset rule type configuration information includes: necessary rule type configuration information, and / or timely rule type configuration information, and the generating of prompt words according to the current medical specification information and the preset rule type configuration information includes:
[0022] generating the prompt word according to the current medical specification information, the necessary rule type configuration information, and / or the timely rule type configuration information;
[0023] The method of using the preset large language model based on the prompt word to extract medical specification data for each business node for the quality control rule type from the current medical specification information, and generating a business quality control rule for each business node for the quality control rule type based on the medical specification data, includes:
[0024] The preset large language model is used to extract the medical standard data of the necessary rule type and / or the medical standard data of the timely rule type for each service node from the current medical standard information based on the prompt word;
[0025] The preset large language model is used to generate business quality control rules for the necessary rule type and / or the timely rule type for each business node based on the medical standard data for the necessary rule type and / or the medical standard data for the timely rule type for each business node.
[0026] In an optional embodiment, the method further comprises:
[0027] Displaying the business quality control rules of the multiple business nodes on the quality control rule configuration page;
[0028] According to the rule adjustment operation for any service node among the multiple service nodes input through the quality control rule configuration page, the service quality control rule of the any service node is adjusted.
[0029] In an optional embodiment, the method further comprises:
[0030] According to the interface configuration operations for the multiple service nodes input through the interface configuration page, respectively configuring the model calling interface of the preset large language model for the multiple service nodes;
[0031] The method of generating business quality control rules for multiple business nodes under the preset medical business by using a preset large language model according to the current medical standard information includes:
[0032] According to the current medical specification information, the preset large language model is called by using the model calling interface to generate business quality control rules for the multiple business nodes.
[0033] In an optional embodiment, the method further comprises:
[0034] According to the post-quality control analysis operation input through the closed-loop quality control analysis page, according to the in-process quality control results corresponding to each business node under the preset medical business within the preset time period, post-quality control analysis is performed on each business node, and the post-quality control results of each business node are generated and displayed.
[0035] In a second aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory through the bus, and the processor executes the machine-readable instructions to execute any one of the methods described in the first aspect.
[0036] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described in any one of the first aspects is executed.
[0037] The present application provides a quality control method, electronic device and storage medium for medical services, wherein the method includes: receiving medical execution data uploaded by medical staff through a medical terminal, the medical execution data being the data generated by the medical staff when performing the business operation corresponding to the target business node under the preset medical service, and performing in-process quality monitoring on the medical execution data according to the pre-generated business quality control rules for the target business node, and obtaining in-process quality control results, wherein the business quality control rules are obtained by using a preset large language model and based on the current medical specification information for the preset medical service stored in a preset specification publishing platform, and if the in-process quality control results indicate that the medical execution data does not meet the business quality control rules of the target business node, then pushing an in-process quality control reminder to the medical terminal. The preset large language model is used to automatically maintain the business quality control rules and perform quality monitoring on the medical execution data, so as to realize closed-loop quality control of medical services and improve the accuracy of quality control. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0039] Figure 1 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 1 ;
[0040] Figure 2 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 2 ;
[0041] Figure 3 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 3 ;
[0042] Figure 4Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 4 ;
[0043] Figure 5 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 5 ;
[0044] Figure 6 A schematic diagram of a quality control rule configuration page provided in an embodiment of the present application;
[0045] Figure 7 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 6 ;
[0046] Figure 8 A schematic diagram of an interface configuration page provided in an embodiment of the present application;
[0047] Fig. 9 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 7 ;
[0048] Fig.10 A schematic diagram of the structure of a quality control device for medical services provided in an embodiment of the present application;
[0049] Fig.11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0050] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme 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 only part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.
[0051] The business quality control rules of traditional manually maintained business quality control solutions need to be manually maintained and updated to adapt to current policies and requirements of different hospitals. They cannot be timely and comprehensive, and are not suitable for hospitals to rely on them completely. Based on this, this application is equipped with a preset large language model to perform learning, training and optimization of relevant medical standards, which can automatically generate business quality control rules, automatically evaluate whether the medical process complies with the standards, and realize seamless closed-loop management of medical services.
[0052] Figure 1 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 1 The execution subject of this embodiment can be an electronic device of a hospital management department, such as a desktop computer, a laptop computer, a tablet computer, etc.
[0053] like Figure 1 As shown, the method may include:
[0054] S101. Receive medical execution data uploaded by medical staff through a medical terminal.
[0055] Among them, medical staff may be doctors, nurses and other personnel who perform medical services, and medical services may include, for example, blood transfusion services, medical advice services, and hospitalization services.
[0056] Each medical service includes multiple business nodes. Taking the blood transfusion service as an example, the business nodes of the blood transfusion service include blood bank issuance, blood collection and verification, double verification before transfusion, blood transfusion, blood transfusion inspection, end of transfusion, blood bag recovery in the ward, and blood bag destruction. Taking the medical order service as an example, the business nodes of the medical order service include checking the patient's condition, confirming the patient's contraindicated drugs, formulating treatment plans, etc.
[0057] Among them, the target business node is any business node under the preset medical business. When the medical staff performs the business operation corresponding to the target business node under the preset medical business, medical execution data is generated, that is, the medical execution data is the data generated by the medical staff performing the business operation corresponding to the target business node under the preset medical business. The medical staff inputs the medical execution data through the medical terminal and uploads it to the electronic equipment of the hospital management department. Correspondingly, the electronic equipment of the hospital management department receives the medical execution data uploaded by the medical terminal.
[0058] The medical execution data may include execution information of the business operation corresponding to the target business node. For example, the target business node is a blood collection and verification node, the medical staff is a nurse, and the business operation corresponding to the target business node is the operation of the nurse collecting blood and verifying it. The medical execution data may be, for example, "completed 1 blood collection and verification". For another example, the target business node is blood transfusion, the medical staff is a nurse, and the business operation corresponding to the target business node is the nurse performing a blood transfusion operation on the patient. The medical execution data may be, for example, "the infusion rate is 1.5 ml per minute, and the time difference between the end of blood transfusion and the start of blood transfusion is 2.5 hours". For another example, the target business node is blood transfusion inspection, the medical staff is a nurse, and the business operation corresponding to the target business node is that the nurse performs a blood transfusion inspection operation during the patient's blood transfusion process. The medical execution data may be, for example, "inspections have been made twice".
[0059] It is worth noting that the medical terminal can be a handheld terminal, an inpatient workstation terminal, or an outpatient workstation terminal. The handheld terminal can be, for example, a PDA scanning terminal. When a nurse performs corresponding business operations, the medical execution data can be uploaded to the electronic device of the hospital management department through the handheld terminal. When a resident doctor performs corresponding business operations, the medical execution data can be uploaded to the electronic device of the hospital management department through the inpatient workstation terminal. When an outpatient doctor performs business operations, the medical execution data can be uploaded to the electronic device of the medical management department through the outpatient workstation terminal.
[0060] S102. According to the pre-generated business quality control rules for the target business node, the medical execution data is subjected to in-process quality control to obtain in-process quality control results.
[0061] Among them, the business quality control rules are obtained by using a preset large language model based on the current medical specification information for the preset medical business stored in the preset specification publishing platform.
[0062] The preset specification publishing platform stores the current medical specification information for the preset medical business, that is, the latest medical specification information, such as the "Eighteen Core Systems for Medical Quality and Safety". The current medical specification information is usually in text form. The preset large language model can be used to extract information from the current medical specification information for the preset medical business stored in the preset specification publishing platform to obtain the business quality control rules for the target business node under the preset medical business. The preset large language model can be, for example, a GPT model.
[0063] Taking the target business node as a blood transfusion node as an example, the quality control rules of the blood transfusion node are as follows:
[0064] 1. Red blood cells
[0065] Infusion rate: 1 to 2 ml / kg body weight per minute. Average time difference between the end of transfusion and the start of transfusion: usually 2-4 hours, including blood typing, cross-matching, blood product preparation and infusion process.
[0066] 2. Platelets
[0067] Infusion rate: 10 to 20 ml per minute. Average time difference between the end of transfusion and the start of transfusion: Due to the limited shelf life of platelets, they are usually transfused as soon as possible after collection. The time difference may be from tens of minutes to several hours.
[0068] 3. Fresh frozen plasma (FFP)
[0069] Infusion rate: 10 to 20 ml per minute, average time difference between end of transfusion and start of transfusion: usually 1-2 hours, depending on the medical treatment process and the patient's needs.
[0070] 4. Cryoprecipitation
[0071] Infusion rate: 10 to 20 ml per minute. Average time difference between the end of transfusion and the start of transfusion: Due to the short shelf life of cryoprecipitate, it is usually infused as soon as possible after collection. The time difference may be from tens of minutes to several hours.
[0072] 5. Leukocyte concentrate
[0073] Infusion rate: no more than 5 ml per minute. Average time difference between the end of transfusion and the start of transfusion: Since the preparation and infusion of leukocyte concentrate are relatively simple, the time difference may be shorter, usually within a few hours.
[0074] 6. Plasma substitutes
[0075] Infusion rate: 500-1000 ml per hour, average time difference between end of transfusion and start of transfusion: Since these products do not require complex cross-matching like whole blood or blood cells, the time difference may be shorter, usually between a few minutes and a few hours.
[0076] According to the business quality control rules for the target business node, the medical execution data is subjected to in-process quality monitoring to obtain the in-process quality control results. The in-process quality control results are the evaluation results of the medical process of the medical staff at the target business node, which are used to determine whether the medical process complies with the specifications.
[0077] S103. If the in-process quality control result indicates that the medical execution data does not meet the business quality control rules of the target business node, an in-process quality control reminder is pushed to the medical terminal.
[0078] If the in-process quality control results indicate that the medical execution data does not meet the business quality control rules of the target business node, which means that the medical staff's medical process under the target business node does not comply with the current medical standard information, an in-process quality control reminder will be pushed to the medical terminal so that the medical staff can adjust the business operations corresponding to the target business node, thereby promptly correcting the medical staff's business operations to avoid errors and ensure patient safety. The in-process quality control reminder can be displayed and / or broadcasted through the medical terminal so that the medical staff can be informed of the standardization of the medical process in a timely manner.
[0079] For example, the medical execution data of the blood transfusion inspection node is "two inspections were carried out during the blood transfusion process", and the business quality control rule of the blood transfusion inspection node is "three inspections are required for blood transfusion". In this way, a quality control reminder that the blood transfusion inspection time is less than three times is pushed to the medical terminal, so that the medical staff can perform the blood transfusion inspection operation again.
[0080] For example, the medical execution data of the blood transfusion node is "the infusion speed is 3 ml per minute", and the business quality control rule of the blood transfusion node is "1-2 ml per minute". If the blood transfusion speed does not meet 1-2 ml per minute, a quality control reminder is pushed to the medical terminal so that medical staff can adjust the infusion speed in time.
[0081] In this embodiment, at the product code level, a preset large language model is used to autonomously learn current medical specification information and generate the latest business quality control rules to apply current policies. There is no need for continuous customization and research and development, which improves the efficiency of generating business quality control rules. At the product delivery level, flexible delivery is achieved without the need for the R&D team to configure and manually maintain business quality control rules, which meets hospital requirements, responds quickly, and shortens the implementation cycle. At the user usage level, business quality control rules are used to remind medical staff of the rationality of their operations, and they can correct them in time to avoid errors, thereby ensuring patient safety and improving work efficiency.
[0082] Figure 2 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 2 In an optional embodiment, as Figure 2 As shown, in the above step S102, according to the pre-generated business quality control rules for the target business node, the medical execution data is subjected to in-process quality control, and before the in-process quality control results are obtained, the method further includes:
[0083] S201. Obtain current medical standard information from a preset standard publishing platform.
[0084] Among them, the preset specification publishing platform stores current medical specification information for preset medical services. The current medical specification information is usually in text form. The current medical specification information can be the latest released medical specification information for preset medical services, for example, it can be core system specifications related to medical quality and patient safety.
[0085] It is worth noting that the preset specification publishing platform can be an internal specification publishing platform of the hospital or a specification publishing platform of the medical industry. The internal specification publishing platform of the hospital can store the current medical specification information of the hospital for the preset medical services, and the specification publishing platform of the medical industry can store the current medical specification information of the medical industry for the preset medical services. Among them, the current medical specification information of the hospital for the preset medical services can be adjusted according to the hospital's own situation based on the current medical specification information of the medical industry for the preset medical services.
[0086] S202. Based on the current medical standard information, a preset large language model is used to generate business quality control rules for multiple business nodes under the preset medical business.
[0087] According to the current medical standard information, a preset large language model is used to extract business quality control rules for multiple business nodes under the preset medical business from the current medical standard information.
[0088] In this embodiment, a preset large language model is used to learn the current medical standard information to generate business quality control rules for multiple business nodes under the preset medical business, wherein the current medical standard information is the current medical standard information within the hospital for the preset medical business, or the current medical standard information of the medical industry for the preset medical business, thereby giving the latest quality control rules based on the artificial intelligence model and realizing customized generation of business quality control rules.
[0089] Figure 3 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 3 In an optional implementation, the above step S202, based on the current medical specification information, uses a preset large language model to generate business quality control rules for multiple business nodes under the preset medical business, including:
[0090] S301. Generate prompt words according to current medical standard information and preset rule type configuration information.
[0091] S302: Input the prompt word into a preset large language model.
[0092] Among them, the preset rule type configuration information is used to indicate the quality control rule type, and the quality control rule type includes a necessary rule type and / or a timely rule type. The necessary rule type is used to indicate that the quality control rule is a necessary quality control rule, which is used to evaluate whether the business node is executed completely and without omissions. The timely rule type is used to indicate that the quality control rule is a timely quality control rule, which is used to evaluate whether the business node is executed in a timely manner and without timeout.
[0093] Generate prompt words based on current medical specification information and preset rule type configuration information, and use the prompt words as model input and input them into the preset large language model.
[0094] S303, using a preset large language model based on prompt words, extracting medical specification data for each business node for a quality control rule type from the current medical specification information, and generating business quality control rules for each business node for a quality control rule type based on the medical specification data.
[0095] A preset large language model is used to extract medical standard data for quality control rule types at each business node from current medical standard information based on prompt words, wherein the medical standard data for quality control rule types at each business node include: medical standard data for necessary rule types and / or medical standard data for timely rule types at each business node.
[0096] The medical specification data for necessary rule types are medical specification data related to necessity under each business node. The medical specification data for necessary rule types may include: the number of business operations that must be performed under each business node. For example, the number of business operations that must be performed under the double check before blood transfusion node may include the number of double check operations before blood transfusion, such as 1 time.
[0097] The medical standard data for the timely rule type is the medical standard data related to the time limit under each business node. The medical standard data for the timely rule type may include: the operating parameters of the business operations related to the time limit under each business node. For example, the operating parameters of the business operations related to the time limit under the blood transfusion node may include: the blood transfusion infusion speed, and the average time difference between the end of blood transfusion and the start of blood transfusion.
[0098] A preset large language model is used to generate business quality control rules for each business node based on medical standard data for the quality control rule type, including business quality control rules of necessary rule types and / or business quality control rules of timely rule types.
[0099] In an optional implementation, the above step S301 generates prompt words according to the current medical specification information and the preset rule type configuration information, including:
[0100] Prompt words are generated based on current medical practice information, necessary rule type configuration information, and / or timely rule type configuration information.
[0101] The preset rule type configuration information includes: necessary rule type configuration information, and / or timely rule type configuration information.
[0102] Generate prompt words based on current medical standard information, necessary rule type configuration information, and / or timely rule type configuration information, and input the prompt words into a preset large language model.
[0103] In an optional implementation, the above step S303 uses a preset large language model based on prompt words to extract medical specification data for each business node for a quality control rule type from the current medical specification information, and generates a business quality control rule for each business node for a quality control rule type based on the medical specification data, including:
[0104] Using a preset large language model based on prompt words, extract medical specification data for necessary rule types and / or medical specification data for timely rule types for each business node from current medical specification information;
[0105] A preset large language model is used to generate business quality control rules for necessary rule types and / or timely rule types for each business node based on the medical standard data for necessary rule types and / or medical standard data for timely rule types for each business node.
[0106] A preset large language model is used to extract medical standard data for necessary rule types and / or medical standard data for timely rule types from the current medical standard information based on prompt words, wherein the medical standard data for quality control rule types at each business node include: medical standard data for necessary rule types and / or medical standard data for timely rule types at each business node.
[0107] The medical standard data for the necessary rule type is medical standard data related to necessity under each business node, and the medical standard data for the necessary rule type may include: the number of business operations that must be performed under each business node.
[0108] The medical standard data for the timely rule type is the medical standard data related to the time limit under each business node, and the medical standard data for the timely rule type may include: operation parameters of business operations related to the time limit under each business node.
[0109] Then, a preset large language model is used to generate business quality control rules for necessary rule types and / or business quality control rules for timely rule types for each business node based on the medical standard data for necessary rule types and / or the medical standard data for timely rule types for each business node, wherein the business quality control rules for necessary rule types for each business node may include: medical standard data for necessary rule types for each business node, and the business quality control rules for timely rule types for each business node may include: medical standard data for timely rule types for each business node.
[0110] For example, the medical execution data is "the end time of blood transfusion is 10 o'clock, and the start time of blood transfusion is 7 o'clock". The business quality control rules of the blood transfusion node for the timely rule type include "the average time difference between the end of blood transfusion and the start of blood transfusion: usually 1-2 hours". It can be seen that the time difference between the end of blood transfusion and the start of blood transfusion in the actual medical process determined based on the medical execution data is 3 hours. If it does not meet 1-2 hours, it is determined that the blood transfusion process takes a long time, so it does not meet the business quality control rules of the timely rule type.
[0111] For another example, the medical execution data is "completed 1 blood collection check", and the business quality control rules of the blood collection check node for the necessary rule type include "the number of blood collection check operations is 1 time". It can be seen that the number of times the nurse performs the "blood collection check" operation in the actual medical process is 1 time, so the business quality control rules of the necessary rule type are met.
[0112] In this embodiment, by generating business quality control rules for different quality control rule types for each business node, necessary and timely quality monitoring is performed on medical execution data to evaluate whether the medical staff has performed the business operations corresponding to the target business node completely and timely during the medical process. The business quality control rules are used to remind medical staff of the rationality of their operations, and timely corrections are made to avoid errors, thereby ensuring patient safety and improving work efficiency.
[0113] Figure 4 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 4 In an optional implementation, the above step S301 generates prompt words according to the current medical specification information and the preset rule type configuration information, including:
[0114] S401, generating prompt words according to current medical standard information, preset rule type configuration information and preset quality control rule data model.
[0115] In the above step S303, the preset large language model is used based on the prompt word to extract the medical standard data of the quality control rule type under each business node from the current medical standard information, and the business quality control rules for the quality control rule type of each business node are generated based on the medical standard data, including:
[0116] S402. Using a preset large language model based on prompt words, extract medical specification data for each business node for a quality control rule type from the current medical specification information, and generate business quality control rules for each business node for a quality control rule type based on the medical specification data and the preset quality control rule data model.
[0117] The preset quality control rule data model is used to indicate the data format of the business quality control rules for the quality control rule type.
[0118] According to the current medical specification information, the preset rule type configuration information and the preset quality control rule data model, prompt words are generated, and the preset large language model is used to extract the medical specification data for the quality control rule type of each business node from the current medical specification information based on the prompt words. Based on the medical specification data and the preset quality control rule data model, the medical specification data is filled with data in the data format indicated by the preset quality control rule data model to generate business quality control rules for the quality control rule type of each business node.
[0119] Among them, the preset quality control rule data model can be a quality control rule table, Table 1 is the quality control rule table, as shown in Table 1, the table header includes field name, field description, type, length, non-empty, and description, wherein the field name is the operation identifier of the business operation of each business node for the quality control rule type, including the operation identifier of the business operation that must be executed under each business node, and / or, the operation identifier of the business operation related to timeliness under each business node.
[0120] The field description is a specific description of the field name. The type is the type of the field name, including string type and numeric type. The length is the length of the field name. Non-empty indicates whether the field value corresponding to the field name is empty. For necessary rule types, it is empty. For timely rule types, it is non-empty (there are time limit-related operation parameters). The description is the quality control description for the field name.
[0121] Table 1
[0122] Field Name Field Description type length Not empty illustrate
[0123] In this embodiment, by defining a quality control rule model, the preset large language model outputs business quality control rules as required according to the defined quality control rule model, adapting to current policy requirements, without the need for continuous customization of research and development, and without the need for the R&D team to configure manual maintenance of business quality control rules, thus meeting hospital requirements, responding quickly, and shortening the implementation cycle.
[0124] Figure 5 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 5 ,like Figure 5 As shown, in an optional implementation, the method may further include:
[0125] S501. Display business quality control rules for multiple business nodes on a quality control rule configuration page.
[0126] S502: Adjust the service quality control rules of any service node according to the rule adjustment operation for any service node among the multiple service nodes input through the quality control rule configuration page.
[0127] After using the preset large language model to generate business quality control rules for multiple business nodes, the business quality control rules for multiple business nodes can be displayed on the quality control rule configuration page. The management personnel can input rule adjustment operations for any business node through the quality control rule configuration page, and the electronic device adjusts the business quality control rules of any business node according to the operation.
[0128] Figure 6 A schematic diagram of a quality control rule configuration page provided in an embodiment of the present application, such as Figure 6As shown, the business quality control rules of the blood transfusion node are "for plasma, the time difference between the end of blood transfusion and the start of blood transfusion is less than 30 minutes", "for platelets, the time difference between the end of blood transfusion and the start of blood transfusion is less than 10 minutes", "for cryoprecipitate, the time difference between the end of blood transfusion and the start of blood transfusion is less than 10 minutes", "for whole blood, the time difference between the end of blood transfusion and the start of blood transfusion is less than 60 minutes", and "for red blood cells, the time difference between the end of blood transfusion and the start of blood transfusion is less than 60 minutes".
[0129] Through the rule adjustment operation, "For platelets, the time difference between the end of blood transfusion and the start of blood transfusion is less than 10 minutes" can be adjusted to "For platelets, the time difference between the end of blood transfusion and the start of blood transfusion is less than 5 minutes". The specific adjustment can be made according to the actual needs of the hospital, and this embodiment does not make any special limitations on this.
[0130] In this embodiment, by displaying the business quality control rules generated based on the preset large language model on the front-end page, it can also be adjusted according to the actual situation at different stages of the hospital, which is more flexible. Moreover, the maintenance workload is relatively small because the adjustments are made through the front-end page.
[0131] Figure 7 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 6 ,like Figure 7 As shown, in an optional implementation, the method may further include:
[0132] S601. According to interface configuration operations for multiple business nodes input through an interface configuration page, a model calling interface of a preset large language model is configured for each of the multiple business nodes.
[0133] The administrator inputs interface configuration operations for multiple business nodes through the interface configuration page, and configures the model calling interface of the preset large language model for the multiple business nodes respectively. The model calling interface can be, for example, a GPT calling interface.
[0134] Figure 8 A schematic diagram of an interface configuration page provided in an embodiment of the present application, such as Figure 8 As shown, the interface configuration page displays the enablement option of the model call interface of the preset large language model and the options of multiple business nodes. Click a business node (such as blood transfusion inspection) and the enablement option to configure the model call interface of the preset large language model for the business node.
[0135] In the above step S202, based on the current medical standard information, a preset large language model is used to generate business quality control rules for multiple business nodes under the preset medical business, including:
[0136] S602. Based on the current medical standard information, a preset large language model is called using a model calling interface to generate business quality control rules for multiple business nodes.
[0137] The model calling interface is used to call the preset large language model, and the business quality control rules for multiple business nodes under the preset medical business are generated according to the current medical specification information. The business quality control rules are returned to each business node by calling the model calling interface, and the returned business quality control rules are called during the business execution process for quality control reminders.
[0138] In this embodiment, for a business node that needs to perform a medical process standard evaluation, a preset large language model can be called through a model calling interface to generate business quality control rules for the business node.
[0139] Fig. 9 Schematic diagram of the process of the quality control method of medical services provided in the embodiment of the present application Figure 7 ,like Fig. 9 As shown, in an optional implementation, the method may further include:
[0140] S701. According to the post-quality control analysis operation input through the closed-loop quality control analysis page, and according to the in-process quality control results corresponding to each business node under the preset medical business within the preset time period, a post-quality control analysis is performed on each business node, and the post-quality control results of each business node are generated and displayed.
[0141] After performing in-process quality monitoring on the medical execution data according to the pre-generated business quality control rules for the target business node and obtaining the in-process quality control results corresponding to the target business node, the management personnel can also input the post-quality control analysis operation through the closed-loop quality control analysis page. According to the post-quality control analysis operation, based on the in-process quality control results corresponding to each business node under the preset medical business within the preset time period, perform post-quality control analysis on each business node, generate and display the post-quality control results of each business node, wherein the preset time period can be the preset time period before the current moment, and the time granularity of the preset time period can be day, day, or quarter.
[0142] Among them, the post-quality control results include: missing rate or timely rate. The missing rate is used to indicate whether the business node is executed completely, and the timely rate is used to indicate whether the business node is executed in time. Then, the business nodes with high missing rates and business nodes with long time consumption (i.e., low timely rate) are determined to conduct closed-loop quality control analysis. Among them, the missing rate is the ratio of the number of operations that are missing (the difference between the total number of operations corresponding to the business quality control rules and the number of operations that have been executed) to the total number of operations corresponding to the business quality control rules, and the timely rate is the ratio of the number of operations that meet the time corresponding to the business quality control rules to the total number of operations corresponding to the business quality control rules.
[0143] Taking the blood transfusion inspection node as an example, the inspection rate (missing rate) 15 minutes before the start of blood transfusion can be calculated based on the in-process quality control results 15 minutes before the start of blood transfusion, and the inspection rate (missing rate) 15 minutes before the end of blood transfusion can be calculated based on the in-process quality control results 15 minutes before the end of blood transfusion. In addition, the inspection rate (missing rate) 15 minutes before the start of blood transfusion and 15 minutes before the end of blood transfusion can be calculated based on the in-process quality control results 15 minutes before the start of blood transfusion and 15 minutes before the end of blood transfusion, so that management personnel can be informed of any irregularities in the process of blood transfusion inspection nodes in a timely manner.
[0144] Taking the blood transfusion application node as an example, the blood transfusion application execution rate (1-missing rate) can also be calculated and displayed, so that management personnel can promptly learn about irregularities in the blood transfusion application node process.
[0145] Taking the blood transfusion node as an example, the timeliness rate corresponding to the blood transfusion node within the preset time period can also be calculated to reflect whether the infusion of the blood transfusion node is timely and whether the infusion speed meets the standard.
[0146] It is worth mentioning that it is also possible to analyze the number of untimely infusions for transfusion nodes in each clinical department within a preset time period, and calculate the proportion of untimely infusions in each clinical department, the name of the clinical department with the largest number of untimely infusions, and the obvious untimely transfusion orders.
[0147] Of course, it is also possible to automatically trigger a timed post-quality control analysis of each business node based on the in-process quality control results corresponding to each business node under the preset medical business within the preset time period, generate and display the post-quality control results of each business node, and the post-quality control results can be displayed in text or graphic form.
[0148] In this embodiment, post-event quality control is performed based on the in-event quality control results to obtain post-event quality control results for post-event analysis and summary to improve the medical execution strategy, thereby truly realizing seamless closed-loop management of medical services.
[0149] Fig.10 This is a schematic diagram of the structure of a quality control device for medical services provided in an embodiment of the present application, which can be integrated into an electronic device.
[0150] like Fig.10 As shown, the device may include:
[0151] The receiving module 801 is used to receive the medical execution data uploaded by the medical staff through the medical terminal. The medical execution data is the data generated by the medical staff when performing the business operation corresponding to the target business node under the preset medical business;
[0152] Processing module 802, used to perform in-process quality monitoring on medical execution data according to pre-generated business quality control rules for target business nodes to obtain in-process quality control results, wherein the business quality control rules are obtained by using a preset large language model based on current medical specification information for preset medical services stored in a preset specification publishing platform;
[0153] The push module 803 is used to push an in-process quality control reminder to the medical terminal if the in-process quality control result indicates that the medical execution data does not meet the business quality control rules of the target business node.
[0154] In an optional implementation, the processing module 802 is specifically configured to:
[0155] Obtain current medical standard information from the preset standard publishing platform;
[0156] Based on the current medical standard information, a preset large language model is used to generate business quality control rules for multiple business nodes under the preset medical business.
[0157] In an optional implementation, the processing module 802 is specifically configured to:
[0158] Generate prompt words according to current medical standard information and preset rule type configuration information, where the preset rule type configuration information is used to indicate the quality control rule type;
[0159] Input the prompt word into the preset large language model;
[0160] A preset large language model is used based on prompt words to extract medical specification data for each business node for a quality control rule type from the current medical specification information, and business quality control rules for each business node for a quality control rule type are generated based on the medical specification data.
[0161] In an optional implementation, the processing module 802 is specifically configured to:
[0162] Generate prompt words according to current medical standard information, preset rule type configuration information and preset quality control rule data model;
[0163] A preset large language model is used based on prompt words to extract medical specification data for each business node for a quality control rule type from the current medical specification information, and business quality control rules for each business node for a quality control rule type are generated based on the medical specification data and the preset quality control rule data model.
[0164] In an optional implementation, the preset rule type configuration information includes: necessary rule type configuration information, and / or timely rule type configuration information. The processing module 802 is specifically used to:
[0165] Generate prompt words according to current medical standard information, necessary rule type configuration information, and / or timely rule type configuration information;
[0166] The preset large language model is used based on prompt words to extract the medical specification data for each business node for the quality control rule type from the current medical specification information, and the business quality control rules for each business node for the quality control rule type are generated based on the medical specification data, including:
[0167] Using a preset large language model based on prompt words, extract medical specification data for necessary rule types and / or medical specification data for timely rule types for each business node from current medical specification information;
[0168] A preset large language model is used to generate business quality control rules for necessary rule types and / or timely rule types for each business node based on the medical standard data for necessary rule types and / or medical standard data for timely rule types for each business node.
[0169] In an optional embodiment, it also includes:
[0170] A display module 804 is used to display the business quality control rules of multiple business nodes on the quality control rule configuration page;
[0171] The processing module 802 is further configured to adjust the service quality control rules of any service node according to the rule adjustment operation for any service node among the multiple service nodes input through the quality control rule configuration page.
[0172] In an optional implementation, the processing module 802 is further configured to:
[0173] According to the interface configuration operations for multiple business nodes input through the interface configuration page, a model calling interface of a preset large language model is configured for each of the multiple business nodes;
[0174] The processing module 802 is specifically configured to:
[0175] According to the current medical standard information, the model calling interface is used to call the preset large language model to generate business quality control rules for multiple business nodes.
[0176] In an optional implementation, the processing module 802 is further configured to:
[0177] According to the post-quality control analysis operation entered through the closed-loop quality control analysis page, based on the in-process quality control results corresponding to each business node under the preset medical business within the preset time period, post-quality control analysis is performed on each business node, and the post-quality control results of each business node are generated and displayed.
[0178] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.
[0179] Fig.11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Fig.11 As shown, the device may include: a processor 901, a memory 902 and a bus 903. The memory 902 stores machine-readable instructions executable by the processor 901. When the electronic device is running, the processor 901 communicates with the memory 902 through the bus 903, and the processor 901 executes the machine-readable instructions to perform the above method.
[0180] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. The computer program is executed when a processor is running, and the processor executes the above method.
[0181] In the embodiment of the present application, the computer program can also execute other machine-readable instructions when run by the processor to execute other methods described in the embodiment. For the specific execution method steps and principles, please refer to the description of the embodiment, which will not be repeated here.
[0182] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.
[0183] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0184] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0185] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0186] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.
[0187] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. They should all 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 quality control method for medical services, characterized in that: include: Receiving medical execution data uploaded by medical staff through a medical terminal, wherein the medical execution data is data generated by the medical staff performing a business operation corresponding to a target business node under a preset medical business; According to the pre-generated business quality control rules for the target business node, the medical execution data is subjected to in-process quality monitoring to obtain in-process quality control results, wherein the business quality control rules are obtained by using a preset large language model based on the current medical specification information for the preset medical business stored in a preset specification publishing platform; If the in-process quality control result indicates that the medical execution data does not meet the business quality control rules of the target business node, an in-process quality control reminder is pushed to the medical terminal.
2. The method according to claim 1, characterized in that Before performing in-process quality monitoring on the medical execution data according to the pre-generated service quality control rules for the target service node and obtaining the in-process quality control results, the method further includes: Acquire the current medical standard information from the preset standard publishing platform; According to the current medical standard information, a preset large language model is used to generate business quality control rules for multiple business nodes under the preset medical business.
3. The method according to claim 2, characterized in that The method of generating business quality control rules for multiple business nodes under the preset medical business by using a preset large language model according to the current medical standard information includes: Generate a prompt word according to the current medical specification information and preset rule type configuration information, wherein the preset rule type configuration information is used to indicate the quality control rule type; Inputting the prompt word into the preset large language model; The preset large language model is used to extract medical specification data for each business node for the quality control rule type from the current medical specification information based on the prompt word, and the business quality control rules for each business node for the quality control rule type are generated based on the medical specification data.
4. The method according to claim 3, characterized in that The generating of prompt words according to the current medical specification information and the preset rule type configuration information includes: Generate the prompt word according to the current medical standard information, the preset rule type configuration information and the preset quality control rule data model; The method of using the preset large language model based on the prompt word to extract medical specification data for each business node for the quality control rule type from the current medical specification information, and generating a business quality control rule for each business node for the quality control rule type based on the medical specification data, includes: The preset large language model is used to extract the medical specification data of each business node for the quality control rule type from the current medical specification information based on the prompt word, and the business quality control rules for each business node for the quality control rule type are generated based on the medical specification data and the preset quality control rule data model.
5. The method according to claim 3, characterized in that: The preset rule type configuration information includes: necessary rule type configuration information, and / or timely rule type configuration information. The prompt word is generated according to the current medical specification information and the preset rule type configuration information, including: generating the prompt word according to the current medical specification information, the necessary rule type configuration information, and / or the timely rule type configuration information; The method of using the preset large language model based on the prompt word to extract medical specification data for each business node for the quality control rule type from the current medical specification information, and generating a business quality control rule for each business node for the quality control rule type based on the medical specification data, includes: The preset large language model is used to extract the medical standard data of the necessary rule type and / or the medical standard data of the timely rule type for each service node from the current medical standard information based on the prompt word; The preset large language model is used to generate business quality control rules for the necessary rule type and / or the timely rule type for each business node based on the medical standard data for the necessary rule type and / or the medical standard data for the timely rule type for each business node.
6. The method according to claim 2, characterized in that The method further comprises: Displaying the business quality control rules of the multiple business nodes on the quality control rule configuration page; According to the rule adjustment operation for any service node among the multiple service nodes input through the quality control rule configuration page, the service quality control rule of the any service node is adjusted.
7. The method according to claim 2, characterized in that: The method further comprises: According to the interface configuration operations for the multiple service nodes input through the interface configuration page, respectively configuring the model calling interface of the preset large language model for the multiple service nodes; The method of generating business quality control rules for multiple business nodes under the preset medical business by using a preset large language model according to the current medical standard information includes: According to the current medical specification information, the preset large language model is called by using the model calling interface to generate business quality control rules for the multiple business nodes.
8. The method according to claim 1, characterized in that The method further comprises: According to the post-quality control analysis operation input through the closed-loop quality control analysis page, according to the in-process quality control results corresponding to each business node under the preset medical business within the preset time period, post-quality control analysis is performed on each business node, and the post-quality control results of each business node are generated and displayed.
9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and the processor executes the machine-readable instructions to perform the method described in any one of claims 1 to 8.
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 method according to any one of claims 1 to 8 is executed.