Medical supplies allocation method and system based on deep learning

Abstract

The present invention relates to the technical field of medical material allocation, and in particular to a medical material allocation method and system based on deep learning. The method comprises: by counting the existing medical material stocks in the hospital and the existing medical material stocks in the warehouse, and at the same time counting the transportation cost of the warehouse and the inventory turnover rate of the warehouse; constructing a medical material stock prediction model to predict the medical material stocks in the next cycle; setting a stock detection threshold for each type of medical material, when the predicted value of the stock of any type of medical material is lower than the given stock detection threshold, it is determined that the stock of this type of medical material is at risk of shortage; when the medical material stock is at risk of shortage, selecting the warehouse with the highest matching degree for medical material allocation.

Description

Background Art

[0002] In the current medical system, the allocation of medical supplies is a complex and critical process involving multiple links and factors, such as storage, forecasting, risk assessment, and allocation strategies.

[0003] Traditional methods of allocating medical supplies often rely on simple statistical and forecasting models, which often lack the accuracy to predict when dealing with complex and changing demands for medical supplies. This results in hospitals being unable to obtain the required medical supplies in a timely manner when faced with emergencies; existing risk assessment methods usually only consider a single factor, such as material inventory or transportation costs, while ignoring other important factors that may affect allocation decisions. This results in allocation strategies that may not fully reflect the actual situation, increasing the risk and uncertainty of allocation; traditional allocation methods often adopt fixed strategies that lack flexibility and adaptability. When the needs of the hospital or external conditions change, these strategies may not be adjusted in time, resulting in inefficient allocation or failure to meet the needs of the hospital. Summary of the invention

[0004] The main purpose of the present invention is to provide a medical supply allocation method based on deep learning, and further to provide a medical supply allocation system based on deep learning that can run and implement the above method, so as to effectively solve the above problems mentioned in the background technology.

[0005] The technical solution of the present invention is as follows:

[0006] First, a medical supplies allocation method based on deep learning is proposed, which includes the following steps:

[0007] S1. Count the medical supplies in the hospital and the warehouse, and also count the transportation cost and inventory turnover rate of the warehouse;

[0008] S2. Build a medical supplies inventory forecasting model to predict the inventory of medical supplies in the next cycle;

[0009] S3. Set a stock detection threshold for each type of medical supplies. When the predicted stock value of any type of medical supplies is lower than the given stock detection threshold, it is determined that there is a risk of shortage of the stock of this type of medical supplies.

[0010] S4. When there is a risk of shortage of medical supplies, select the warehouse with the highest matching degree to distribute medical supplies.

[0011] A further improvement of the present invention is that the specific content of S1 is: a preset statistical period and a preset time interval are set at the same time, and the existing m types of medical supplies in the hospital, the existing m types of medical supplies in the warehouse, the m types of medical supplies of the supplier, and the transportation costs and inventory turnover rates of different warehouses are counted once every preset time interval within the preset statistical period. Then, the medical supplies inventory of the i-th hospital can be expressed as the matrix X i ∈R m×n , n is the number of statistics within the preset statistical period, where the matrix X i The jth column of It represents the inventory of m types of medical supplies counted for the jth time in the i-th hospital within the preset statistical period.

[0012] A further improvement of the present invention is that S2 comprises the following specific steps:

[0013] S21, constructing a medical supplies inventory prediction model, wherein the medical supplies inventory prediction model comprises a convolutional neural network part and a recursive neural network part, wherein the convolutional neural network part is composed of a convolutional neural network layer, and a one-dimensional convolution kernel is used for convolution, and the height of the one-dimensional convolution kernel is the same as the input data X. i The number of rows m is equal, and the output matrix V i ∈R k×n , k is the number of convolution kernels in the convolutional neural network layer;

[0014] S22, the recursive neural network part adopts the LSTM network structure, which includes five LSTM network layers. Let the zth column of Vi be Will Input LSTM network structure and output the corresponding hidden state As the prediction result of the recurrent neural network part, the model is based on the medical resource inventory statistics of the current period. Generate forecasts

[0015] A further improvement of the present invention is that the training method of the medical supplies inventory prediction model in S2 is: using the medical resource inventory statistics of a hospital in three periods to construct a training set, and using the loss function Where L represents the loss function, represents the true value of the i-th hospital at the j+n time step, represents the predicted value of the ith hospital at the j+n time step, ‖.‖ F represents the F-norm.

[0016] A further improvement of the present invention is that the calculation formula of the matching degree in S4 is:

[0017]

[0018] Among them, Fitness is the matching value, s is the distance from the warehouse to the hospital, g is the inventory of medical supplies of the corresponding type in the warehouse, C represents the transportation cost, R represents the inventory turnover rate of the warehouse, and α, β, γ, and δ represent weight factors respectively.

[0019] Secondly, a medical material allocation system based on deep learning is proposed, which includes: information collection module, inventory prediction module, risk judgment module, and material allocation module;

[0020] The information collection module is used to count the existing medical supplies in the hospital and the existing medical supplies in the warehouse, and also count the transportation cost and inventory turnover rate of the warehouse;

[0021] The inventory prediction module is used to build a medical supplies inventory prediction model to predict the inventory of medical supplies in the next cycle;

[0022] The risk judgment module is used to set a stock detection threshold for each type of medical supplies. When the predicted value of the stock of any type of medical supplies is lower than the given stock detection threshold, it is determined that there is a shortage risk of the stock of this type of medical supplies;

[0023] The material allocation module is used to select the warehouse with the highest matching degree to allocate medical materials when there is a risk of shortage of medical materials in stock.

[0024] The technical effects of the present invention are as follows:

[0025] A medical material allocation method based on deep learning has been constructed. By building a medical material inventory prediction model and using deep learning technology, the inventory of medical materials in the next cycle can be predicted more accurately. This helps hospitals understand material needs in advance and formulate reasonable procurement and allocation plans; in the allocation process, a variety of factors are taken into consideration, such as material inventory, transportation costs, inventory turnover rate, etc., and a matching calculation formula is set to comprehensively evaluate the allocation plans of different warehouses. This helps hospitals choose the best allocation strategy and reduce allocation risks; the allocation method is highly flexible and adaptable. When the hospital's needs or external conditions change, the system can update data in real time, recalculate the matching degree, and quickly adjust the allocation strategy. This helps hospitals respond quickly to various emergencies and ensure the timely supply of medical materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0027] Figure 1This is a flowchart of a medical supplies allocation method based on deep learning according to Embodiment 1 of the present invention;

[0028] Figure 2 This is a structural diagram of a medical supplies allocation system based on deep learning according to Example 2 of the present invention. DETAILED DESCRIPTION

[0029] Embodiment 1:

[0030] This embodiment constructs a medical material allocation method based on deep learning. By constructing a medical material inventory prediction model and using deep learning technology, it can more accurately predict the inventory of medical materials in the next cycle. This helps hospitals understand material needs in advance and formulate reasonable procurement and allocation plans; in the allocation process, a variety of factors are taken into consideration, such as material inventory, transportation costs, inventory turnover rate, etc., and a matching calculation formula is set to comprehensively evaluate the allocation plans of different warehouses. This helps hospitals choose the best allocation strategy and reduce allocation risks; the allocation method is highly flexible and adaptable. When the hospital's needs or external conditions change, the system can update the data in real time, recalculate the matching degree, and quickly adjust the allocation strategy. This helps hospitals respond quickly to various emergencies and ensure the timely supply of medical supplies.

[0031] Medical supplies deployment methods based on deep learning, such as Figure 1 As shown, the specific steps include:

[0032] S1. Count the medical supplies in the hospital and the warehouse, and also count the transportation cost and inventory turnover rate of the warehouse;

[0033] S2. Build a medical supplies inventory forecasting model to predict the inventory of medical supplies in the next cycle;

[0034] S3. Set a stock detection threshold for each type of medical supplies. When the predicted stock value of any type of medical supplies is lower than the given stock detection threshold, it is determined that there is a risk of shortage of the stock of this type of medical supplies.

[0035] S4. When there is a risk of shortage of medical supplies, select the warehouse with the highest matching degree to distribute medical supplies.

[0036] In this embodiment, the specific content of S1 is: preset a statistical period and a time interval, and count the existing m types of medical supplies in the hospital, the existing m types of medical supplies in the warehouse, the m types of medical supplies of the supplier, and the transportation costs and inventory turnover rates of different warehouses at preset time intervals within the preset statistical period. Then, the medical supplies inventory of the i-th hospital can be expressed as the matrix X i ∈R m×n, n is the number of statistics within the preset statistical period, where the matrix X i The jth column of It represents the inventory of m types of medical supplies counted for the jth time in the i-th hospital within the preset statistical period.

[0037] In this embodiment, S2 includes the following specific steps:

[0038] S21, constructing a medical supplies inventory prediction model, wherein the medical supplies inventory prediction model comprises a convolutional neural network part and a recursive neural network part, wherein the convolutional neural network part is composed of a convolutional neural network layer, and a one-dimensional convolution kernel is used for convolution, and the height of the one-dimensional convolution kernel is the same as the input data X. i The number of rows m is equal, and the output matrix V i ∈R k×n , k is the number of convolution kernels in the convolutional neural network layer;

[0039] S22, the recursive neural network part adopts the LSTM network structure, which includes five LSTM network layers. Let the zth column of Vi be Will Input LSTM network structure and output the corresponding hidden state As the prediction result of the recurrent neural network part, the model is based on the medical resource inventory statistics of the current period. Generate forecasts

[0040] In this embodiment, the training method of the medical supplies inventory prediction model in S2 is: use the medical resource inventory of a hospital in three periods counted by the hospital medical resource inventory statistics equipment to construct a training set, and use the loss function Where L represents the loss function, represents the true value of the i-th hospital at the j+n time step, represents the predicted value of the ith hospital at the j+n time step, ‖.‖ F represents the F-norm.

[0041] In this embodiment, the calculation formula of the matching degree in S4 is:

[0042]

[0043] Among them, Fitness is the matching value, s is the distance from the warehouse to the hospital, g is the inventory of medical supplies of the corresponding type in the warehouse, C represents the transportation cost, R represents the inventory turnover rate of the warehouse, and α, β, γ, and δ represent weight factors respectively.

[0044] Embodiment 2:

[0045] This embodiment proposes a medical supplies allocation system based on deep learning, such as Figure 2 As shown, it includes: information collection module, inventory prediction module, risk judgment module, and material allocation module;

[0046] The information collection module is used to count the existing medical supplies in the hospital and the existing medical supplies in the warehouse, and also count the transportation cost and inventory turnover rate of the warehouse;

[0047] The inventory prediction module is used to build a medical supplies inventory prediction model to predict the inventory of medical supplies in the next cycle;

[0048] The risk judgment module is used to set a stock detection threshold for each type of medical supplies. When the predicted value of the stock of any type of medical supplies is lower than the given stock detection threshold, it is determined that there is a shortage risk of the stock of this type of medical supplies;

[0049] The material allocation module is used to select the warehouse with the highest matching degree to allocate medical materials when there is a risk of shortage of medical materials in stock.

[0050] Embodiment 3:

[0051] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-mentioned deep learning-based medical supplies allocation method by calling the computer program stored in the memory.

[0052] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein at least one computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the medical supplies allocation method based on deep learning provided by the above method embodiment. The electronic device may also include other components for realizing the functions of the device. For example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0053] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), and it can also be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0054] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0055] The present invention is described with reference to flowcharts and block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process or block in the flowchart and block diagram, as well as the combination of processes and blocks in the flowchart or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 Process or multiple processes and boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 Process or multiple processes and boxes Figure 1 The steps for the functions specified in one or more boxes.

[0057] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A medical supplies allocation method based on deep learning, characterized by: The specific steps include: S1. Count the medical supplies in the hospital and the warehouse, and also count the transportation cost and inventory turnover rate of the warehouse; S2. Build a medical supplies inventory forecasting model to predict the inventory of medical supplies in the next cycle; S3. Set a stock detection threshold for each type of medical supplies. When the predicted stock value of any type of medical supplies is lower than the given stock detection threshold, it is determined that there is a risk of shortage of the stock of this type of medical supplies. S4. When there is a risk of shortage of medical supplies, select the warehouse with the highest matching degree to distribute medical supplies.

2. The medical supplies allocation method based on deep learning according to claim 1, characterized in that: The specific content of S1 is: preset a statistical period and a time interval, and count the existing m types of medical supplies in the hospital, the existing m types of medical supplies in the warehouse, the m types of medical supplies of the supplier, and the transportation costs and inventory turnover rates of different warehouses at preset time intervals within the preset statistical period. The medical supplies inventory of the i-th hospital can be expressed as the matrix X i ∈R m×n , n is the number of statistics within the preset statistical period, where the matrix X i The jth column of It represents the inventory of m types of medical supplies counted for the jth time in the i-th hospital within the preset statistical period.

3. The medical supplies allocation method based on deep learning according to claim 2, characterized in that: The S2 comprises the following specific steps: S21, constructing a medical supplies inventory prediction model, wherein the medical supplies inventory prediction model comprises a convolutional neural network part and a recursive neural network part, wherein the convolutional neural network part is composed of a convolutional neural network layer, and a one-dimensional convolution kernel is used for convolution, and the height of the one-dimensional convolution kernel is the same as the input data X. i The number of rows m is equal, and the output matrix V i ∈R k×n , k is the number of convolution kernels in the convolutional neural network layer; S22, the recursive neural network part adopts the LSTM network structure, which includes five LSTM network layers. Let the zth column of Vi be Will Input LSTM network structure and output the corresponding hidden state As the prediction result of the recurrent neural network part, the model is based on the medical resource inventory statistics of the current period. Generate forecasts 4. The medical supplies allocation method based on deep learning according to claim 3 is characterized in that: The training method of the medical supplies inventory prediction model in S2 is: use the medical resource inventory statistics equipment of a hospital in three periods to construct a training set, and use the loss function Among them, L represents the loss function, represents the true value of the i-th hospital at the j+n time step, represents the predicted value of the ith hospital at the j+n time step, ‖.‖ F represents the F-norm.

5. The medical supplies allocation method based on deep learning according to claim 4 is characterized in that: The calculation formula of the matching degree in S4 is: Among them, Fitness is the matching value, s is the distance from the warehouse to the hospital, g is the inventory of medical supplies of the corresponding type in the warehouse, C represents the transportation cost, R represents the inventory turnover rate of the warehouse, and α, β, γ, and δ represent weight factors respectively.

6. A medical material allocation system based on deep learning, which is implemented based on the medical material allocation method based on deep learning according to any one of claims 1 to 5, characterized in that: The system includes: an information collection module, an inventory prediction module, a risk judgment module, and a material allocation module; The information collection module is used to count the existing medical supplies in the hospital and the existing medical supplies in the warehouse, and also count the transportation cost and inventory turnover rate of the warehouse; The inventory prediction module is used to build a medical supplies inventory prediction model to predict the inventory of medical supplies in the next cycle; The risk judgment module is used to set a stock detection threshold for each type of medical supplies. When the predicted value of the stock of any type of medical supplies is lower than the given stock detection threshold, it is determined that there is a shortage risk of the stock of this type of medical supplies; The material allocation module is used to select the warehouse with the highest matching degree to allocate medical materials when there is a risk of shortage of medical materials in stock.

Images