Medical resource management method and system based on large model

Abstract

The invention relates to the technical field of medical resource management, in particular to a medical resource management method and system based on a large model, and the method comprises the steps: carrying out the prediction according to the historical data of the patient amount, and obtaining the future patient amount; predicting according to the use frequency of the target equipment in the current target department and the future patient quantity to obtain the future additional demand quantity of the target equipment in the target department; judging and scheduling the target equipment according to the dispatchable number of the target equipment and the future extra demand quantity of the target equipment in the target department; according to the invention, cross-department medical equipment dynamic scheduling management is carried out based on the number of used equipment, the historical data of the number of patients and the comprehensive maintenance, purchase and use status of multiple groups of data, and the technical problem of deficiencies in medical resource management in smart hospitals is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical resource management, and in particular to a medical resource management method and system based on a large model. Background Art

[0002] With the rapid development of science and technology, smart hospitals have become a development trend in the medical field. By integrating medical resources and optimizing medical service processes, they have significantly improved the quality and efficiency of medical services. In the process of building smart hospitals, advanced information systems are applied to the management and configuration of medical resources to achieve the optimal configuration and refined management of medical equipment and materials, ensuring the efficient and stable operation of hospitals. However, in the process of building smart hospitals, there are still some shortcomings in the management of medical equipment and materials.

[0003] First, in the construction of smart hospitals, although equipped with advanced information systems, the failure of some high-value medical equipment cannot be predicted, which not only easily leads to waste of medical resources, but also seriously affects the normal operation of clinical departments. When equipment fails, how to effectively allocate maintenance personnel and quickly determine the location of the equipment failure becomes the key to affecting the timely restoration of equipment.

[0004] Furthermore, the distribution of medical supplies and equipment often faces unreasonable problems. The reason is that the material demands among departments vary significantly, and in the actual distribution process, there is often a lack of scientific and precise basis, which directly leads to waste or shortage of materials. Especially in special periods, when hospitals are faced with tight supplies and equipment, how to distribute materials efficiently and fairly has become a difficult problem that needs to be solved urgently.

[0005] In addition, the existing medical resource management system is relatively complex in terms of usage costs, and its functions are mainly limited to data storage management, lacking the data processing capabilities to effectively respond to emergencies; in terms of data integration, the system faces compatibility and consistency issues with data from different sources, formats and structures, which makes it difficult for the system to ensure the accuracy and integrity of the data, which in turn causes serious signal island phenomena, making it impossible for various devices to effectively communicate information, thus affecting the management system's predictive maintenance and fault diagnosis of medical equipment.

[0006] Therefore, it is of great significance to solve the technical problems of insufficient medical resource management in smart hospitals. Summary of the invention

[0007] The purpose of this application is to provide a medical resource management method and system based on a large model, which solves the shortcomings of medical resource management in smart hospitals, including unpredictable failures of high-value medical equipment, unreasonable material allocation, loopholes in medical waste management, complex system usage costs, and difficulties in data integration. The current technology is unable to effectively integrate multi-source data and lacks emergency data processing capabilities, resulting in waste and shortages of resources, affecting the quality and efficiency of medical services.

[0008] In order to solve the above technical problems, the solution adopted by this application is as follows: The present invention provides a medical resource management method based on a large model, characterized in that it comprises the following steps: Step S1: Obtain historical data of the number of patients in the target department, and make a prediction based on the historical data of the number of patients to obtain the number of patients in the future; Step S2: Obtain the usage frequency of the target equipment in the target department; perform a prediction based on the current usage frequency of the target equipment in the target department and the future patient volume to obtain the future additional demand for the target equipment in the target department; Step S3: Obtain the dispatchable number of target devices; determine and dispatch the target devices according to the dispatchable number of target devices and the future additional demand for target devices in the target department; The dispatchable number of target devices includes the number of target devices that can be used by the target department, the number of target devices that can be repaired in the future, and the number of target devices to be used.

[0009] In some embodiments, in step S2, the future additional demand for the target equipment in the target department is predicted by the formula: Among them, N is the future additional demand for the target equipment; W is the number of available target equipment in the target department; X is the number of future patients; is the frequency of use of the i-th type of equipment; The above formula is used to predict the future additional demand for the target equipment in the target department.

[0010] In some embodiments, the usage frequency of the target device is obtained by the formula: in, is the frequency of use of the i-th type of equipment; s is the number of equipment used; p is the number of patients; The usage frequency of the target device is obtained by the above formula.

[0011] In some embodiments, in step S3, the formula: Among them, y is the judgment value of whether the target equipment needs to be dispatched; T is the number of target equipment to be used; N is the future additional demand for target equipment in the target department; n is the number of target equipment that can be repaired in the future; when When , it means that there is no need to dispatch the target equipment to the target department; when , it means that the target equipment needs to be dispatched to the target department.

[0012] In some embodiments, obtaining the number of target devices that can be repaired in the future includes the following steps: Step S401: Obtain relevant data of the equipment to be repaired, and sort the relevant data of the equipment to be repaired to obtain sorted data: the relevant data includes the recent use frequency of the equipment and the future additional demand for the target equipment of the recent target department; Step S402: Obtain the number of maintenance personnel and the working hours of each maintenance personnel; and predict the number of equipment that can be repaired on the current day and in the future based on the number of maintenance personnel, the working hours of each maintenance personnel, and the sorting data.

[0013] In some embodiments, in step S402, the formula: Among them, A is the number of maintenance personnel; B is the working hours of maintenance personnel; n is the number of Class A equipment expected to be completed on the same day; m is the number of Class A equipment expected to be completed on the next day; Indicates the jth device to be repaired among the class a devices.

[0014] This can be used to predict the number of target equipment that can be repaired today and in the future.

[0015] The present invention also provides a medical resource management system based on a large model, characterized in that it includes an application input module, a monitoring module, a feature extraction module, a data processing module, a database, and an output module; The application input module is used to read the report filled in by the staff, organize the information on the report into raw data, package it and send it to the feature extraction module for further processing; the report includes a material application report, an equipment repair report, and a fault repair report; the report format includes text and picture formats; the application input module includes a material application unit and a fault repair unit; The feature extraction module is used to receive the data packet transmitted by the application input module, extract the raw data in the data packet and organize it into standard data, then package it and send it to the data processing module for further processing; the standard data includes the name code of the equipment, the department code, the working status data, and the fault data; The monitoring module is used to monitor and obtain the operating status data of the target device in real time, process the operating status data into standard data, package it and send it to the data processing module for further processing; the operating status data includes the frequency of use of the device, the number and time of maintenance, the current value of the device, and the voltage value of the device; A data processing module, used to receive data packets transmitted by the feature extraction module and the monitoring module, extract standard data in the data packets, retrieve the data and model of the target department in the database according to the standard data, process the standard data and the data of the target department according to the model, predict the future demand of the target equipment, and schedule the target equipment according to the future demand; the model includes a remaining life model of the equipment, a future patient volume model, a future demand model of the equipment, a scheduling allocation model, and a fault judgment model; The database is used to store the above-mentioned models, the historical patient volume data of each department, the recent patient volume data of each department, and the equipment data of each department.

[0016] In some embodiments, an output module is also included, and the output module is used to integrate the data provided by the data processing module into a result report and output it; the result report includes a fault diagnosis report, a fault repair report, a material application report, and a scrap report.

[0017] In some embodiments, the data processing module processes the data transmitted by the monitoring module, including predicting the remaining life of the target device through the formula: in, is the remaining lifetime risk rate of the target device, is the base risk rate of the target device, is the i-th type of failure risk factor of the target device; yes The regression coefficient of ; from this, the remaining life risk rate of the target equipment is obtained.

[0018] In some embodiments, the data packet includes a label, and each module receives the data packet according to the label; the label is obtained by the formula: in, , , , The initial value of is 0. When each module is labeled, the corresponding k value becomes 1. This is how the data packets are numbered.

[0019] The technical solution of the present application has at least the following advantages and beneficial effects: 1. The present invention uses the historical data of the number of equipment used and the number of patients to derive the forecast results of future equipment demand. It also monitors the equipment status based on a large model, combines the maintenance time, the number of personnel and the frequency of equipment use, and comprehensively integrates the maintenance, procurement and use status to perform dynamic scheduling management of cross-department medical equipment; it incorporates the maintenance results and status of the equipment into the hospital inventory management framework, avoiding the idle waste of equipment that is put back into use after maintenance, especially in hospitals with a large number of equipment. This method can use equipment more efficiently and at a low cost.

[0020] 2. In the present invention, for the equipment to be repaired, the system prioritizes the data according to the emergency needs of the equipment and the time required for repair, and then the maintenance personnel will systematically handle each faulty equipment according to this sorting order. This approach not only facilitates maintenance personnel to manage time and allocate resources more effectively, but also ensures the orderliness and efficiency of maintenance work.

[0021] 3. When applying for materials and repairing equipment, the system will also determine the demand for materials or equipment based on the number of patients, and limit the quantity of materials and dispatch equipment based on the demand. This not only ensures the work efficiency of each department, but also effectively avoids the waste caused by excessive accumulation of materials and operational obstructions caused by shortages.

[0022] 4. The present invention uses large model technology to receive and process data. It can receive multimodal data and extract features to generate a standard data format, breaking the data island phenomenon and ensuring data privacy and security, which strongly supports the optimal allocation and efficient use of medical resources. The technology uses standard data to generate a unified report format, which helps maintenance personnel to view and understand the report, avoiding the situation where the equipment maintenance time is extended due to unclear descriptions by medical staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is the overall working flow diagram of the system of the present invention; Figure 2 It is a signal transmission flow chart of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] It should be noted that similar numbers and letters represent similar items in the following figures, so once an item is defined in one figure, it does not need to be further defined and explained in the subsequent figures. If the terms "center", "upper", "lower", "inner", "outer", etc. indicate an orientation or position relationship based on the orientation or position relationship shown in the figure, or the orientation or position relationship in which the product of the application is usually placed when in use, it is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present application. It should also be noted that unless otherwise clearly specified and limited, if the terms "set", "install", and "connect" appear, they should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be a connection between the two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0026] Example Please refer to Figure 1-Figure 2 The present invention provides a medical resource management method and system based on a large model, comprising the following steps: Step S1: Obtain historical data of the number of patients in the target department, and make a prediction based on the historical data of the number of patients to obtain the number of patients in the future; Step S2: Obtain the usage frequency of the target equipment in the target department; perform a prediction based on the current usage frequency of the target equipment in the target department and the future patient volume to obtain the future additional demand for the target equipment in the target department; Step S3: Obtain the dispatchable number of target devices; determine and dispatch the target devices according to the dispatchable number of target devices and the future additional demand for target devices in the target department; The dispatchable number of target devices includes the number of target devices that can be used by the target department, the number of target devices that can be repaired in the future, and the number of target devices to be used.

[0027] In step S1, prediction is performed based on the historical data of the patient volume to obtain the future patient volume, which specifically includes the following steps: Obtain historical patient volume data of the target department in the past period of time from the hospital's information management system; the time granularity of the data may be one of daily, weekly, and monthly; collect frequency data of each patient's use of relevant equipment or materials; The historical patient volume and usage frequency data were cleaned, and then the patient volume data set and the usage frequency data set were normalized respectively, so that data of different magnitudes can be calculated on the same scale; the normalization formula is as follows: Where x is the original data of historical patient volume or usage frequency; x' is the normalized historical patient volume or usage frequency data; min(x) and max(x) are the minimum and maximum values ​​of the original data of historical patient volume or usage frequency, respectively; The LSTM model is constructed based on the normalized data, including the following: Input layer: The normalized historical patient volume data and the frequency of use of equipment or supplies are used as input features; the time step is set to t, then the input vector is .

[0028] LSTM layer: handles long-term dependencies in time series data through special gating mechanisms (input gate, forget gate, and output gate); The input gate determines what information to update to the cell state. It consists of two parts: input gate activation and cell state candidate. The input gate activation formula is as follows:

[0029] The candidate formulas for cell states are as follows: in, and All are weight matrices of historical patient volume or usage frequency data; and All are bias items for historical patient volume or frequency of use data; Cell state candidate data for historical patient volume or usage frequency data; Activate data for input of historical patient volume or frequency of use data; is the hidden state of the previous time step of the historical patient volume or usage frequency data; and They are all activation functions; the cell state update formula is as follows:

[0030] Forget gate formula: in, The forget gate weight matrix for historical patient volume or usage frequency data; Enter data for current historical patient volume or frequency of use data; is a bias term for historical patient volume or frequency of use data; Historical patient volume or cell status update data using frequency data.

[0031] The output gate determines what information is output from the cell state, using the formula: The final hidden state is: in, Output data for historical patient volume or frequency of use data; It is the hidden state of historical patient volume or usage frequency data; A weight matrix for historical patient volume or frequency of use data; It is the bias item of historical patient quantity or usage frequency data; Output layer: outputs the predicted demand for equipment or materials in a certain period of time in the future.

[0032] It should be noted that the output of the predicted demand for equipment or materials in a certain period of time in the future also needs to be error-checked to ensure the accuracy of the model; in this embodiment, the mean square error is selected as the loss function to test the model, and the formula is: in, is the actual demand value; is the demand value predicted by the model, and n is the number of samples.

[0033] It should be noted that in each time step, the LSTM model will receive an input vector, which usually contains the feature information of the current time step, that is, the data set of historical patient volume or usage frequency data for this prediction; the LSTM model will update the memory state through the above-mentioned gating mechanism based on the input vector and the current memory state; the updated memory state will be used to generate the prediction result of the current time step, that is, the future demand for equipment or materials of the target department.

[0034] It should be noted that the LSTM model in this embodiment can be trained for the first time using the hospital's historical data to obtain the memory state of the model; each subsequent prediction will use the current data as new training data to optimize the prediction model, and at this time a new memory state of the model will be obtained, which can be used as the current memory state for the next prediction.

[0035] It should be noted that the usage frequency of the target device is obtained through formula (1): (1) in, is the frequency of use of the i-th type of equipment; s is the number of equipment used; p is the number of patients; The usage frequency of the target device is obtained by the above formula.

[0036] In step S2, the future additional demand for the target equipment in the target department is predicted by formula (2): (2) Among them, N is the future additional demand for the target equipment; W is the number of available target equipment in the target department; X is the number of future patients; is the usage frequency of the i-th type of equipment; the above formula is used to predict the future additional demand for the target equipment in the target department; when When , it means that the target equipment has a certain demand in the target department, and N equipment can be allocated to the target department; when , it means that the target equipment is sufficient in the target department, and the remaining equipment in the target department can also cope with the future patient volume.

[0037] exist In this case, it is determined whether additional scheduling equipment is needed; in step S3, by formula (3): (3) Among them, y is the judgment value of whether the target equipment needs to be dispatched; T is the number of target equipment to be used; N is the future additional demand for target equipment in the target department; n is the number of target equipment that can be repaired in the future; when When , it means that there is no need to dispatch the target equipment to the target department; when , it means that the target equipment needs to be dispatched to the target department.

[0038] exist Under the condition of , the system starts to search the database for the use of the target equipment by other departments, and judges the demand of other departments for the target equipment based on this use and the number of patients in other departments in the near future; the demand of other departments for the target equipment can be judged by the above formula (1), formula (2), and formula (3); when it is judged that It means that other departments can provide redundant equipment to the target department to relieve the pressure of equipment usage.

[0039] In order to make the equipment demand more accurate, the equipment demand for several days in the future can be calculated, and a curve can be drawn based on the predicted equipment demand for several days in the future; the system records this data as historical data to provide a reference for similar situations in the future and to facilitate verification and evaluation.

[0040] It should be noted that when it is predicted that other departments cannot provide excess target equipment, the hospital can consider purchasing or borrowing target equipment from other hospitals to cope with the increase in patient volume in the future.

[0041] It should be noted that when the recent patient growth rate of a department is greater than a certain threshold, the system will automatically retrieve the equipment usage frequency of the target department and formulate an allocation plan in advance. The process of formulating this allocation plan is the same as the process of the allocation plan for equipment application.

[0042] The acquisition of the number of the above target equipment that can be repaired in the future includes the following steps: Step S401: Obtain relevant data of the equipment to be repaired, and sort the relevant data of the equipment to be repaired to obtain sorted data: the relevant data includes the recent use frequency of the equipment and the future additional demand for the target equipment of the recent target department; Step S402: Obtain the number of maintenance personnel and the working hours of each maintenance personnel; and predict the number of equipment that can be repaired on the current day and in the future based on the number of maintenance personnel, the working hours of each maintenance personnel, and the sorting data.

[0043] Specifically, the equipment is sorted according to its related data, that is, the equipment with more urgent demand is placed in front, and the equipment with less urgent demand is placed in the back; it is expressed as: …, …, ; Among them, a, b, and c are different types of equipment; after completion, they are sorted according to the working hours in the same type, that is, the equipment with short working hours is placed in front and the equipment with long working hours is placed in the back, and the result is: , It should be explained that the man-hours for equipment maintenance are determined by the relevant model; It needs to be explained that the higher the recent frequency of use of the equipment, the more urgent the demand for the equipment; the greater the future additional demand for the target equipment in the target department in the near future, the more urgent the demand for the equipment; equipment with more urgent demand should be given priority for repair to ensure the work efficiency of the hospital.

[0044] In step S402, by formula: Among them, A is the number of maintenance personnel; B is the working hours of maintenance personnel; n is the number of Class A equipment expected to be completed on the same day; m is the number of Class A equipment expected to be completed on the next day; Indicates the jth device to be repaired among the class a devices.

[0045] This can be used to predict the number of target equipment that can be repaired today and in the future.

[0046] It should be noted that if it is expected that after completing the maintenance of Class A equipment on a certain day, the transition to Class B equipment maintenance will be made, then the formula: Where m is the number of Class B equipment that is expected to be completed on the day; i+m is the number of equipment repaired on the day; Indicates the jth device to be repaired in category a; Indicates the jth device to be repaired among the class B devices.

[0047] This can be used to determine the number of Class B equipment that can be repaired after the Class A equipment is repaired on the same day.

[0048] The present invention also provides a medical resource management system based on a large model to implement the above-mentioned medical resource management method based on a large model, including an application input module, a monitoring module, a feature extraction module, a data processing module, a database, and an output module; The application input module is used to read the report filled in by the staff, organize the information on the report into raw data, package it and send it to the feature extraction module for further processing; the report includes a material application report, an equipment repair report, and a fault repair report; the report format includes text and picture formats; the application input module includes a material application unit and a fault repair unit; The feature extraction module is used to receive the data packet transmitted by the application input module, extract the raw data in the data packet and organize it into standard data, then package it and send it to the data processing module for further processing; the standard data includes the name code of the equipment, the department code, the working status data, and the fault data; The monitoring module is used to monitor and obtain the operating status data of the target device in real time, process the operating status data into standard data, package it and send it to the data processing module for further processing; the operating status data includes the frequency of use of the device, the number and time of maintenance, the current value of the device, and the voltage value of the device; A data processing module, used to receive data packets transmitted by the feature extraction module and the monitoring module, extract standard data in the data packets, retrieve the data and model of the target department in the database according to the standard data, process the standard data and the data of the target department according to the model, predict the future demand of the target equipment, and schedule the target equipment according to the future demand; the model includes a remaining life model of the equipment, a future patient volume model, a future demand model of the equipment, a scheduling allocation model, and a fault judgment model; The database is used to store the above-mentioned models, the historical patient volume data of each department, the recent patient volume data of each department, and the equipment data of each department.

[0049] In this embodiment, an output module is also included, and the output module is used to integrate the data provided by the data processing module into a result report and output it; the result report includes a fault diagnosis report, a fault repair report, a material application report, and a scrap report.

[0050] It should be explained that the monitoring module determines whether the device is in normal working condition and the energy consumption of the device by monitoring the current and voltage of the device in real time to see if they are abnormal; If the current value or voltage value exceeds the threshold continuously or exceeds the threshold multiple times, the system determines that the device is abnormal, and the system automatically generates a fault maintenance report; If the system determines that the device does not exceed the threshold based on the current and voltage of the device, it determines that the device is normal; The system can also comprehensively predict the remaining service life of the equipment based on the equipment's energy consumption, maintenance times and time, and usage frequency; if the remaining service life of the equipment exceeds a reference threshold, the system provides a reference solution, which includes equipment maintenance and equipment scrapping.

[0051] Specifically, the remaining life of the target equipment is predicted using the Cox proportional hazard model, using the formula: in, is the remaining lifetime risk rate of the target device, is the base risk rate of the target device, is the i-th type of failure risk factor of the target device; yes The regression coefficient of ; from this, the remaining life risk rate of the target equipment is obtained.

[0052] It should be noted that the regression coefficient It can be estimated by maximum likelihood estimation method.

[0053] For ease of understanding, let's take the usage frequency as an example: Let the usage frequency of the target device be , with other covariates unchanged, if , indicating that the frequency of use is positively correlated with the risk of equipment failure. The risk of equipment failure at some point increases will increase, and For each unit increase, the risk function of the device will be multiplied times; if , indicating that the frequency of use is negatively correlated with the risk of equipment failure. That is, the higher the frequency of use, the lower the risk of equipment failure. if , indicating that the covariate of usage frequency has no effect on the risk function of the equipment, and the impact of this factor on the equipment life can be ignored in the model.

[0054] The Cox proportional risk model is used to predict the probability of equipment failure at a certain point in the future. When the failure probability exceeds a certain threshold, the equipment is judged to be unusable. At this time, the corresponding time point can be obtained, and the remaining life of the equipment can be obtained by calculation based on the time point.

[0055] It should be noted that the monitoring module can also be used to monitor the usage of hospital environment auxiliary equipment terminals, predict faults based on the usage, and ensure that the hospital environment reaches the optimal state. Its monitoring method is consistent with the medical equipment monitoring method. Hospital environment auxiliary equipment includes but is not limited to lights, air conditioners, and water heaters.

[0056] During the use of medical equipment, equipment failures due to other factors may occur from time to time. For example, the display screen of the medical equipment fails and medical staff cannot read the data. In this state, the system cannot determine the equipment failure on its own. At this time, medical staff can submit a maintenance application to the system, and the system will arrange the maintenance.

[0057] Medical staff submit maintenance application information, which can be in the form of text or images. The application module and feature extraction module preprocess, extract features, identify, and classify the text or image information in sequence to finally obtain standardized data. The system then generates a fault report based on the standardized data.

[0058] It should be noted that the fault repair report includes the fault time, fault problem, fault degree, and reference method for solving the problem; the fault problem is automatically generated and filled in by the system, the fault degree is determined by the system, and the reference method for solving the problem is generated by the system based on historical records. The generation of fault reports not only facilitates technicians to quickly locate problems, but also helps hospitals identify potential risk factors. For example, the large model analyzes fault reports. The analyzed information will affect the allocation plan of the equipment in the target department, making the plan more in line with the actual situation of each department and maximizing the efficiency of the equipment in the target department.

[0059] It should be explained that the fault severity includes low, medium and high. The system determines the maintenance time required to repair the equipment based on the fault problem, historical records and related models, and determines the fault severity based on the maintenance time.

[0060] It should be explained that, in this embodiment, the large model regards the time period from when the maintenance personnel receive the order to when they submit the fault repair report as the maintenance time required for the equipment.

[0061] The specific steps for medical staff to apply for supplies, medical staff to report faults, or maintenance personnel to fill out fault reports are similar. The application module transmits data to the feature extraction module for data standardization, and the standardized data is transmitted to the data processing module for processing. After the processing is completed, it is transmitted to the output module for output; Taking medical staff applying for supplies as an example, the specific steps are: Medical staff submit a material application form. The application module reads the information on the material application form and packages the number to form a material application form data packet, and transmits the material application form data packet to the feature extraction module. The material application form includes the name, current quantity, and department of the applied material. After the feature extraction module decompresses the data packet of the material application form, it extracts the text information or picture information to obtain standard data; the standard data includes the name code, the existing quantity, and the department code of the material; it is then packaged and numbered, and the data packet is transmitted to the data processing module; The data processing module receives and decompresses the data packet, retrieves the relevant model and recent data from the database according to the label information, name code, and department code, and then processes the model and the recent data of the corresponding department to predict the demand for the material; reasonably arranges the quantity of applied materials based on the demand and existing stock, and packages the data again and transmits it to the output module.

[0062] The difference between medical staff's fault reporting and medical staff's application for supplies is that the application module reads the information on the fault report form and packages the label to form a fault report form data packet; the fault report form data packet includes the equipment name, department code, and fault problem description; The difference between maintenance personnel filling out fault reports and medical personnel applying for supplies is that the application module reads the information on the fault report and packages it with a number to form a fault report data packet; the fault report data packet includes the equipment name, maintenance time, fault problem description, and fault problem solution; the maintenance time can provide a reference for the estimated completion time of subsequent equipment fault repairs; the fault problem solution can provide a reference solution for the same fault problem of the same equipment in the future; In this embodiment, the data packet includes a label, and each module receives the data packet according to the label; The label is expressed by the formula: in, , , , The initial value of is 0. When each module is labeled, the corresponding k value becomes 1. This is how the data packets are numbered.

[0063] Specifically, Indicates that the data packet is packaged by the application module; Indicates that the data packet is packaged by the feature extraction module; Indicates that the data packet is packaged by the monitoring module; or Indicates that the data packet is packaged by a data processing module.

[0064] Furthermore, when the label When , the data packet can only be received by the feature extraction module; when the label or When the data packet can only be received by the data processing module; when the label or When , the data packet can only be received by the output module.

[0065] It should be noted that the significance of the data packet label is to facilitate each module to determine whether to receive the module, and the data type can be determined based on the label, so as to retrieve the model corresponding to the data in the database.

[0066] So far, various embodiments of the present invention have been described in detail. In order to avoid obscuring the concept of the present invention, some details known in the art are not described. Those skilled in the art can fully understand how to implement the technical solution invented here based on the above description, and the scope of the present invention is defined by the attached claims.

Claims

1. A medical resource management method based on a large model, characterized in that: The following steps are involved: Step S1: Obtain historical data of the number of patients in the target department, and make a prediction based on the historical data of the number of patients to obtain the number of patients in the future; Step S2: Obtain the usage frequency of the target equipment in the target department; perform a prediction based on the current usage frequency of the target equipment in the target department and the future patient volume to obtain the future additional demand for the target equipment in the target department; Step S3: Obtain the dispatchable number of target devices; determine and dispatch the target devices according to the dispatchable number of target devices and the future additional demand for target devices in the target department; The dispatchable number of target devices includes the number of target devices that can be used by the target department, the number of target devices that can be repaired in the future, and the number of target devices to be used.

2. A medical resource management method based on a large model according to claim 1, characterized in that: In step S2, the future additional demand for the target equipment in the target department is predicted by the formula: Among them, N is the future additional demand for the target equipment; W is the number of available target equipment in the target department; X is the number of future patients; is the usage frequency of the i-th type of equipment; the above formula is used to predict the future additional demand for the target equipment in the target department.

3. A medical resource management method based on a large model according to claim 2, characterized in that: The usage frequency of the target device is obtained by the formula: in, is the frequency of use of the i-th type of equipment; s is the number of equipment used; p is the number of patients; The usage frequency of the target device is obtained by the above formula.

4. The medical resource management method based on a large model according to claim 1, characterized in that: In step S3, the formula: Among them, y is the judgment value of whether the target equipment needs to be dispatched; T is the number of target equipment to be used; N is the future additional demand for target equipment in the target department; n is the number of target equipment that can be repaired in the future; when When , it means that there is no need to dispatch the target equipment to the target department; when , it means that the target equipment needs to be dispatched to the target department.

5. A medical resource management method based on a large model according to claim 4, characterized in that: The acquisition of the number of target devices that can be repaired in the future includes the following steps: Step S401: Obtain relevant data of the equipment to be repaired, and sort the relevant data of the equipment to be repaired to obtain sorted data: the relevant data includes the recent use frequency of the equipment and the future additional demand for the target equipment of the recent target department; Step S402: Obtain the number of maintenance personnel and the working hours of each maintenance personnel; and predict the number of equipment that can be repaired on the current day and in the future based on the number of maintenance personnel, the working hours of each maintenance personnel, and the sorting data.

6. A medical resource management method based on a large model according to claim 5, characterized in that: In step S402, by formula, Among them, A is the number of maintenance personnel; B is the working hours of maintenance personnel; n is the number of Class A equipment expected to be completed on the same day; m is the number of Class A equipment expected to be completed on the next day; It indicates the jth device to be repaired in category A. It predicts the number of target devices that can be repaired on the current day and in the future.

7. A medical resource management system based on a large model, characterized in that: To implement a large model-based medical resource management method as described in claims 1 to 6 above, comprising an application input module, a monitoring module, a feature extraction module, a data processing module, a database, and an output module; The application input module is used to read the report filled in by the staff, organize the information on the report into raw data, package it and send it to the feature extraction module for further processing; the report includes a material application report, an equipment repair report, and a fault repair report; the report format includes text and picture formats; the application input module includes a material application unit and a fault repair unit; The feature extraction module is used to receive the data packet transmitted by the application input module, extract the raw data in the data packet and organize it into standard data, then package it and send it to the data processing module for further processing; the standard data includes the name code of the equipment, the department code, the working status data, and the fault data; The monitoring module is used to monitor and obtain the operating status data of the target device in real time, process the operating status data into standard data, package it and send it to the data processing module for further processing; the operating status data includes the frequency of use of the device, the number and time of maintenance, the current value of the device, and the voltage value of the device; A data processing module, used to receive data packets transmitted by the feature extraction module and the monitoring module, extract standard data in the data packets, retrieve the data and model of the target department in the database according to the standard data, process the standard data and the data of the target department according to the model, predict the future demand of the target equipment, and schedule the target equipment according to the future demand; the model includes a remaining life model of the equipment, a future patient volume model, a future demand model of the equipment, a scheduling allocation model, and a fault judgment model; The database is used to store the above-mentioned models, the historical patient volume data of each department, the recent patient volume data of each department, and the equipment data of each department.

8. A medical resource management system based on a large model according to claim 7, characterized in that: It also includes an output module, which is used to integrate the data provided by the data processing module into a result report and output it; the result report includes a fault diagnosis report, a fault repair report, a material application report, and a scrap report.

9. A medical resource management system based on a large model according to claim 7, characterized in that: The data processing module processes the data transmitted by the monitoring module, including predicting the remaining life of the target device through the formula: in, is the remaining lifetime risk rate of the target device, is the base risk rate of the target device, is the i-th type of failure risk factor of the target device; yes The regression coefficient of ; from this, the remaining life risk rate of the target equipment is obtained.

10. A medical resource management system based on a large model according to claim 7, characterized in that: The data packet includes a label, and each module receives the data packet according to the label; the label is calculated by the formula: in, , , , The initial value of is 0. When each module completes the labeling, the corresponding k value becomes 1; this is how the data packets are labeled.

Images