Hospital consumable management system based on AIGC and VMI
The hospital consumables management system based on AIGC and VMI solves the problem of difficulty in accurately predicting the procurement model and quantity in traditional hospital consumables management, and realizes accurate procurement and visual management of high-value consumables, thereby improving management efficiency and inventory control.
Patent Information
- Application Number
- CN202510953078.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-11-11
AI Technical Summary
Traditional hospital consumables management systems struggle to accurately predict the models and quantities of high-value consumables to be purchased, resulting in low management visibility, complex inventory management, and tied-up capital.
The hospital consumables management system based on AIGC and VMI is adopted, including modules for basic information management, procurement plan formulation, warehousing management, outbound management and in-warehouse management. AIGC is used to build a procurement prediction model, and big data statistics and artificial intelligence technology are combined to achieve visualized management.
It enables accurate prediction of the procurement models and quantities of high-value consumables, reduces inventory backlog, improves management efficiency and scientific decision-making, and reduces the capital tied up by suppliers.
Smart Images
Figure CN120932833A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to consumables management, specifically to a hospital consumables management system based on AIGC and VMI. Background Technology
[0002] Hospital consumables management is a core component of the healthcare operations system, and its efficiency and accuracy directly impact the quality of medical services, cost control, and patient safety. With the rapid development of medical technology, the types and usage of high-value consumables (such as cardiac stents and orthopedic implants) and ordinary consumables (such as syringes and sutures) are growing exponentially.
[0003] High-value consumables refer to items like cardiac stents and cerebrovascular guidewires / catheters, which are very expensive (usually over ten thousand yuan) and are typically used in complex cardiovascular, cerebrovascular, orthopedic, and other major surgeries. A key characteristic of high-value consumables is that due to differences in each patient's physical condition, surgical plans, surgical preferences, and consumable usage habits among surgeons, there are many different models of the same high-value consumable. Furthermore, the surgeon must determine the appropriate model based on the specific circumstances during the operation. This necessitates hospitals purchasing large quantities of various models of high-value consumables in advance for surgical needs, resulting in significant capital tied up for suppliers.
[0004] Because traditional hospital consumables management relies on the experience of purchasing staff for procurement, issuance, and inventory management, the procurement quantities of high-value consumables are often inaccurate, sometimes even resulting in extreme situations where suppliers deliver supplies while the patient is still in surgery. Furthermore, traditional hospital consumables management systems use tables and text for display, which is not visual, making tasks such as locating goods, issuing them, and inventory counting extremely complex and user-unfriendly. Summary of the Invention
[0005] (a) Technical problems to be solved
[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a hospital consumables management system based on AIGC and VMI, which can effectively overcome the shortcomings of the existing technology, such as the difficulty in accurately predicting the procurement model and quantity of high-value consumables and the low degree of visualization of hospital consumables management.
[0007] (II) Technical Solution
[0008] To achieve the above objectives, the present invention provides the following technical solution:
[0009] The hospital consumables management system based on AIGC and VMI includes a basic information management module, a procurement plan formulation module, an inbound management module, an outbound management module, an in-warehouse management module, and a management visualization module.
[0010] The basic information management module covers basic information management including units, suppliers, hospitals and warehouses, and supports the maintenance of key information including custody accounts, settlement dates, departments, positions and users. It also provides user password modification function, as well as the definition of rules for data import and export.
[0011] The procurement plan development module utilizes AIGC to build a procurement prediction model. Based on big data statistics, it predicts the procurement models and quantities of various consumables by combining various surgical indicators in the hospital with the selection of surgical plans, surgical procedures, historical surgical records, and the usage habits of different departments and doctors, and develops corresponding procurement plans.
[0012] The inbound management module is responsible for the entire procurement process control, covering the generation, query, review and arrival notification management of replenishment orders, as well as handling sales returns, receiving and acceptance, overage processing and inbound accounting related work;
[0013] The outbound management module coordinates the requisition and fulfillment processes, and is responsible for the generation and review of requisition orders and the management of fulfillment orders. It comprehensively controls the outbound process, covering the creation, review, accounting and confirmation of outbound orders, and also handles matters related to transfer orders and damage reports.
[0014] The warehouse management module revolves around inventory counting, supports both dynamic and static inventory counting operations, has inbound and outbound details and inventory query functions, and can perform warehouse adjustments and invoice settlement management.
[0015] The management visualization module utilizes AIGC to automatically generate diverse media content, including text, images, audio, and video, using artificial intelligence technology to achieve visualized management of hospital consumables.
[0016] Preferably, the basic information management module includes unit information management, supplier information management, hospital information management, warehouse management, storage ledger, settlement date management, department management, job management, user management, user password modification, import definition and export definition.
[0017] Preferably, the procurement plan formulation module utilizes AIGC to construct a procurement prediction model. Leveraging big data statistics, and based on various in-hospital surgical indicators combined with the selection of surgical plans, surgical procedures, historical surgical records, and the usage habits of different departments and doctors, it predicts the procurement models and quantities of various consumables and formulates corresponding procurement plans, including:
[0018] S1. Collect historical data;
[0019] S2. Clean the historical data and integrate it into a unified database through data standardization and structuring. Then, label the data to build a training dataset.
[0020] S3. Use the AIGC large model based on the Transformer architecture to extract key features from the training dataset and construct feature vectors;
[0021] S4. Use the training dataset to train the AIGC large model to obtain the pre-trained AIGC large model.
[0022] S5. Input the surgical plan information, including the surgical plan, surgical procedure, surgical department and surgeon, collected in real time into the pre-trained AIGC large model, and output the procurement model and quantity prediction results of various consumables.
[0023] S6. Considering multiple factors including cost, inventory, and supply stability, optimize the forecast results with multiple objectives and formulate corresponding procurement plans.
[0024] S7. Collect actual procurement and usage data, compare them with the forecast results, and update the model parameters regularly based on the feedback results to optimize the forecast accuracy.
[0025] Preferably, historical data is collected in S1, including:
[0026] Surgical data, including surgical type, surgical plan, surgical procedure, consumable type and quantity, are collected through the Hospital Information System (HIS).
[0027] The Electronic Medical Record (EMR) system collects departmental consumable usage records, as well as departmental and physician data including surgical preferences and consumable usage habits.
[0028] The Supplies Management System (SPD) collects information on consumables, including model, specifications, purchase price, supplier, and inventory status.
[0029] External data, including market price fluctuations and policy adjustments, are collected through publicly available data sources or third-party platforms.
[0030] Preferably, in S3, a large AIGC model based on the Transformer architecture is used to extract key features from the training dataset and construct feature vectors, including:
[0031] The AIGC large model based on the Transformer architecture is used to extract key features from the training dataset, including surgical type, surgical procedure, surgical department, surgeon, consumable model and quantity used, and to construct feature vectors to fuse multi-source data into a unified feature space.
[0032] Preferably, in step S4, the AIGC large model is trained using the training dataset to obtain a pre-trained AIGC large model, including:
[0033] S41. Divide the training dataset into a training set, a validation set, and a test set according to a preset ratio;
[0034] S42. Set the loss function and optimizer for the AIGC large model;
[0035] S43. Input the training set into the AIGC large model for model training;
[0036] S44. Calculate the loss value based on the loss function, and the optimizer updates the model parameters based on the loss value and network gradient information;
[0037] S45. If the loss value is less than the preset threshold, the model training ends and the current AIGC large model is the trained AIGC large model. Otherwise, return to S43 and continue to train the model using the training set.
[0038] S46. Input the validation set into the trained AIGC large model, evaluate the model's generalization ability by observing its performance on the validation set, and fine-tune the model's hyperparameters and structure.
[0039] S47. Input the test set into the optimized AIGC large model, evaluate the model performance, and finally obtain the pre-trained AIGC large model.
[0040] Preferably, the warehouse management module includes replenishment order management, replenishment order not generated history query, replenishment order approval, arrival notification management, pre-sale return application form, return approval, receiving and acceptance form, overage report management, warehouse entry order accounting, and warehouse entry order detailed accounting.
[0041] Preferably, the outbound management module includes requisition order management, requisition order approval, packing order management, outbound order management, outbound order approval, outbound order accounting, outbound order detail accounting, outbound order confirmation, outbound order detail confirmation, transfer order management, and damage report management.
[0042] Preferably, the warehouse management module includes dynamic inventory counting, static inventory counting, detailed inquiry of inbound and outbound slips, inventory inquiry, warehouse adjustment management, and invoice settlement management.
[0043] Preferably, the management visualization module utilizes AIGC to automatically generate diverse media content, including text, images, audio, and video, using artificial intelligence technology to achieve visualized management of hospital consumables, including:
[0044] The dynamic dashboard generates a real-time inventory dashboard for consumables, displaying the inventory levels, consumption trends, and warning thresholds of consumables in each department. It also embeds a 3D model of consumables automatically generated by AIGC, supporting rotation and zoom interaction.
[0045] Intelligent reporting automatically generates weekly / monthly consumables analysis reports, including procurement recommendations, waste rate analysis, and comparison of doctors' consumables usage habits, and inserts dynamic charts automatically generated by AIGC;
[0046] Contextualized videos are generated to recreate surgical scenarios for complex consumables, with annotations of consumable usage steps and precautions, and demonstrations combined with AIGC-generated voice narration.
[0047] Voice interaction and reminders: It uses speech synthesis technology to broadcast inventory warnings and supports voice queries. AIGC automatically generates answers and displays them visually.
[0048] (III) Beneficial Effects
[0049] Compared with the prior art, the hospital consumables management system based on AIGC and VMI provided by the present invention has the following beneficial effects:
[0050] 1) By accurately predicting the models and quantities of consumables through the procurement plan formulation module, hospitals can arrange procurement plans more rationally, avoid over-purchasing or under-purchasing, and enable suppliers to arrange production and supply based on more accurate orders, reduce inventory backlog, and reduce the capital tied up by suppliers.
[0051] 2) The procurement plan formulation module utilizes AIGC to build a procurement prediction model. It combines big data statistics, various in-hospital surgical indicators, surgical plans and procedures, historical surgical records, and usage habits of different departments and doctors to achieve accurate prediction of the procurement models and quantities of various consumables, especially high-value consumables. This helps hospitals formulate scientific and reasonable procurement plans and improve the accuracy of procurement.
[0052] 3) The management visualization module utilizes AIGC to automatically generate diverse media content such as text, images, audio, and video using artificial intelligence technology. This presents various information in the hospital consumables management process in an intuitive and diverse manner, enabling visualized management of hospital consumables. This allows managers to comprehensively and promptly grasp the usage of consumables, improving management efficiency and the scientific nature of decision-making. Attached Figure Description
[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0054] Figure 1 This is a schematic diagram of the system of the present invention;
[0055] Figure 2 This is a flowchart illustrating the process by which the procurement scheme formulation module in this invention uses AIGC to predict the procurement models and quantities of various consumables and formulate corresponding procurement schemes. Detailed Implementation
[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0057] A hospital consumables management system based on AIGC and VMI, such as Figure 1 As shown, it includes a basic information management module, a procurement plan formulation module, an inbound management module, an outbound management module, an in-warehouse management module, and a management visualization module;
[0058] The basic information management module covers basic information management including units, suppliers, hospitals and warehouses, and supports the maintenance of key information including custody accounts, settlement dates, departments, positions and users. It also provides user password modification function, as well as the definition of rules for data import and export.
[0059] The procurement plan development module utilizes AIGC to build a procurement prediction model. Based on big data statistics, it predicts the procurement models and quantities of various consumables by combining various surgical indicators in the hospital with the selection of surgical plans, surgical procedures, historical surgical records, and the usage habits of different departments and doctors, and develops corresponding procurement plans.
[0060] The inbound management module is responsible for the entire procurement process control, covering the generation, query, review and arrival notification management of replenishment orders, as well as handling sales returns, receiving and acceptance, overage processing and inbound accounting related work;
[0061] The outbound management module coordinates the requisition and fulfillment processes, and is responsible for the generation and review of requisition orders and the management of fulfillment orders. It comprehensively controls the outbound process, covering the creation, review, accounting and confirmation of outbound orders, and also handles matters related to transfer orders and damage reports.
[0062] The warehouse management module revolves around inventory counting, supports both dynamic and static inventory counting operations, has inbound and outbound details and inventory query functions, and can perform warehouse adjustments and invoice settlement management.
[0063] The management visualization module utilizes AIGC to automatically generate diverse media content, including text, images, audio, and video, using artificial intelligence technology to achieve visualized management of hospital consumables.
[0064] The basic information management module includes unit information management, supplier information management, hospital information management, warehouse management, storage ledger, settlement date management, department management, job management, user management, user password modification, import definition and export definition.
[0065] The procurement plan development module utilizes AIGC to build a procurement forecasting model. Leveraging big data statistics, it predicts the types and quantities of various consumables to be procured based on in-hospital surgical indicators, surgical plan selection, surgical procedures, historical surgical records, and the usage habits of different departments and doctors. It then develops corresponding procurement plans. Figure 2 As shown, it includes:
[0066] S1. Collect historical data;
[0067] S2. Clean the historical data and integrate it into a unified database through data standardization and structuring. Then, label the data to build a training dataset.
[0068] S3. Use the AIGC large model based on the Transformer architecture to extract key features from the training dataset and construct feature vectors;
[0069] S4. Use the training dataset to train the AIGC large model to obtain the pre-trained AIGC large model.
[0070] S5. Input the surgical plan information, including the surgical plan, surgical procedure, surgical department and surgeon, collected in real time into the pre-trained AIGC large model, and output the procurement model and quantity prediction results of various consumables.
[0071] S6. Considering multiple factors including cost, inventory, and supply stability, optimize the forecast results with multiple objectives and formulate corresponding procurement plans.
[0072] S7. Collect actual procurement and usage data, compare them with the forecast results, and update the model parameters regularly based on the feedback results to optimize the forecast accuracy.
[0073] Specifically, historical data is collected in S1, including:
[0074] Surgical data, including surgical type, surgical plan, surgical procedure, consumable type and quantity, are collected through the Hospital Information System (HIS).
[0075] The Electronic Medical Record (EMR) system collects departmental consumable usage records, as well as departmental and physician data including surgical preferences and consumable usage habits.
[0076] The Supplies Management System (SPD) collects information on consumables, including model, specifications, purchase price, supplier, and inventory status.
[0077] External data, including market price fluctuations and policy adjustments, are collected through publicly available data sources or third-party platforms.
[0078] Specifically, S3 uses a large AIGC model based on the Transformer architecture to extract key features from the training dataset and construct feature vectors, including:
[0079] The AIGC large model based on the Transformer architecture is used to extract key features from the training dataset, including surgical type, surgical procedure, surgical department, surgeon, consumable model and quantity used, and to construct feature vectors to fuse multi-source data into a unified feature space.
[0080] Specifically, in S4, the AIGC large model is trained using the training dataset to obtain a pre-trained AIGC large model, including:
[0081] S41. Divide the training dataset into a training set, a validation set, and a test set according to a preset ratio;
[0082] S42. Set the loss function and optimizer for the AIGC large model;
[0083] S43. Input the training set into the AIGC large model for model training;
[0084] S44. Calculate the loss value based on the loss function, and the optimizer updates the model parameters based on the loss value and network gradient information;
[0085] S45. If the loss value is less than the preset threshold, the model training ends and the current AIGC large model is the trained AIGC large model. Otherwise, return to S43 and continue to train the model using the training set.
[0086] S46. Input the validation set into the trained AIGC large model, evaluate the model's generalization ability by observing its performance on the validation set, and fine-tune the model's hyperparameters and structure.
[0087] S47. Input the test set into the optimized AIGC large model, evaluate the model performance, and finally obtain the pre-trained AIGC large model.
[0088] The above technical solution, through the procurement planning module, accurately predicts the models and quantities of consumables to be purchased, enabling hospitals to arrange procurement plans more rationally, avoiding over-purchasing or under-purchasing, and allowing suppliers to arrange production and supply based on more accurate orders, reducing inventory backlog and reducing the capital tied up by suppliers.
[0089] Meanwhile, the procurement planning module utilizes AIGC to build a procurement forecasting model. By combining big data statistics, various in-hospital surgical indicators, surgical plans and procedures, historical surgical records, and the usage habits of different departments and doctors, it can accurately predict the procurement models and quantities of various consumables, especially high-value consumables. This helps hospitals develop scientific and reasonable procurement plans and improve the accuracy of procurement.
[0090] The following example demonstrates how AIGC technology can be used to predict the procurement model and quantity of consumables.
[0091] For specialized hospitals, such as Tiantan Hospital, whose strength lies in neurosurgery, the most common neurosurgical diseases are cerebral vascular stenosis and cerebral aneurysm, which require the use of special neurosurgical coils for surgery.
[0092] How is AIGC technology applied? First, angiography, specifically digital subtraction angiography (DSA), is performed before the procedure. The result is an X-ray medical image. From the DSA results, doctors determine the following indicators:
[0093] For cerebral vascular stenosis, attention should be paid to the length of the stenosis cutoff in DSA, the degree of stenosis, and the different anatomical locations of the stenosis.
[0094] For aneurysms, attention should be paid to the aneurysm size, neck width, parent artery width, local vessel morphology (whether there is tortuosity), and surrounding perforating arteries from the DSA.
[0095] The above provides the indicators reflected in DSA images for two different diseases in neurosurgery. These indicators help doctors make preliminary judgments: what size coil to choose, how to shape the microcatheter, etc. Then, based on the different sizes selected, the manufacturer and model of the coil to use are determined.
[0096] The inbound management module includes replenishment order management, replenishment order not generated history query, replenishment order approval, arrival notification management, pre-sale return application form, return approval, receiving and acceptance form, overage report management, inbound order accounting, and inbound order detailed accounting.
[0097] The outbound management module includes requisition order management, requisition order approval, packing order management, outbound order management, outbound order approval, outbound order accounting, outbound order detail accounting, outbound order confirmation, outbound order detail confirmation, transfer order management, and damage report management.
[0098] The warehouse management module includes dynamic inventory counting, static inventory counting, detailed inquiry of inbound and outbound orders, inventory inquiry, warehouse adjustment management, and invoice settlement management.
[0099] The management visualization module utilizes AIGC to automatically generate diverse media content, including text, images, audio, and video, using artificial intelligence technology to achieve visualized management of hospital consumables, including:
[0100] The dynamic dashboard generates a real-time inventory dashboard for consumables, displaying the inventory levels, consumption trends, and warning thresholds of consumables in each department. It also embeds a 3D model of consumables automatically generated by AIGC, supporting rotation and zoom interaction.
[0101] Intelligent reporting automatically generates weekly / monthly consumables analysis reports, including procurement recommendations, waste rate analysis, and comparison of doctors' consumables usage habits, and inserts dynamic charts automatically generated by AIGC;
[0102] Contextualized videos are generated to recreate surgical scenarios for complex consumables, with annotations of consumable usage steps and precautions, and demonstrations combined with AIGC-generated voice narration.
[0103] Voice interaction and reminders: It uses speech synthesis technology to broadcast inventory warnings and supports voice queries. AIGC automatically generates answers and displays them visually.
[0104] The above technical solution utilizes AIGC to automatically generate diverse media content such as text, images, audio, and video using artificial intelligence technology. This presents various information in the hospital consumables management process in an intuitive and diverse manner, enabling visualized management of hospital consumables. This allows managers to comprehensively and promptly grasp the usage of consumables, improving management efficiency and the scientific nature of decision-making.
[0105] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A hospital consumables management system based on AIGC and VMI, characterized in that: It includes modules for basic information management, procurement plan development, warehousing management, outbound management, warehouse management, and management visualization. The basic information management module covers basic information management including units, suppliers, hospitals and warehouses, and supports the maintenance of key information including custody accounts, settlement dates, departments, positions and users. It also provides user password modification function, as well as the definition of rules for data import and export. The procurement plan development module utilizes AIGC to build a procurement prediction model. Based on big data statistics, it predicts the procurement models and quantities of various consumables by combining various surgical indicators in the hospital with the selection of surgical plans, surgical procedures, historical surgical records, and the usage habits of different departments and doctors, and develops corresponding procurement plans. The inbound management module is responsible for the entire procurement process control, covering the generation, query, review and arrival notification management of replenishment orders, as well as handling sales returns, receiving and acceptance, overage processing and inbound accounting related work; The outbound management module coordinates the requisition and fulfillment processes, and is responsible for the generation and review of requisition orders and the management of fulfillment orders. It comprehensively controls the outbound process, covering the creation, review, accounting and confirmation of outbound orders, and also handles matters related to transfer orders and damage reports. The warehouse management module revolves around inventory counting, supports both dynamic and static inventory counting operations, has inbound and outbound details and inventory query functions, and can perform warehouse adjustments and invoice settlement management. The management visualization module utilizes AIGC to automatically generate diverse media content, including text, images, audio, and video, using artificial intelligence technology to achieve visualized management of hospital consumables.
2. The hospital consumables management system based on AIGC and VMI according to claim 1, characterized in that: The basic information management module includes unit information management, supplier information management, hospital information management, warehouse management, storage ledger, settlement date management, department management, job management, user management, user password modification, import definition and export definition.
3. The hospital consumables management system based on AIGC and VMI according to claim 1, characterized in that: The procurement plan formulation module utilizes AIGC to construct a procurement prediction model. Leveraging big data statistics, it predicts the procurement models and quantities of various consumables based on in-hospital surgical indicators, surgical plan selection, surgical procedure selection, historical surgical records, and the usage habits of different departments and doctors. It then formulates corresponding procurement plans, including: S1. Collect historical data; S2. Clean the historical data and integrate it into a unified database through data standardization and structuring. Then, label the data to build a training dataset. S3. Use the AIGC large model based on the Transformer architecture to extract key features from the training dataset and construct feature vectors; S4. Use the training dataset to train the AIGC large model to obtain the pre-trained AIGC large model. S5. Input the surgical plan information, including the surgical plan, surgical procedure, surgical department and surgeon, collected in real time into the pre-trained AIGC large model, and output the procurement model and quantity prediction results of various consumables. S6. Considering multiple factors including cost, inventory, and supply stability, optimize the forecast results with multiple objectives and formulate corresponding procurement plans. S7. Collect actual procurement and usage data, compare them with the forecast results, and update the model parameters regularly based on the feedback results to optimize the forecast accuracy.
4. The hospital consumables management system based on AIGC and VMI according to claim 3, characterized in that: Historical data is collected in S1, including: Surgical data, including surgical type, surgical plan, surgical procedure, consumable type and quantity, are collected through the Hospital Information System (HIS). The Electronic Medical Record (EMR) system collects departmental consumable usage records, as well as departmental and physician data including surgical preferences and consumable usage habits. The Supplies Management System (SPD) collects information on consumables, including model, specifications, purchase price, supplier, and inventory status. External data, including market price fluctuations and policy adjustments, are collected through publicly available data sources or third-party platforms.
5. The hospital consumables management system based on AIGC and VMI according to claim 3, characterized in that: S3 employs a large AIGC model based on the Transformer architecture to extract key features from the training dataset and construct feature vectors, including: The AIGC large model based on the Transformer architecture is used to extract key features from the training dataset, including surgical type, surgical procedure, surgical department, surgeon, consumable model and quantity used, and to construct feature vectors to fuse multi-source data into a unified feature space.
6. The hospital consumables management system based on AIGC and VMI according to claim 3, characterized in that: In S4, the AIGC large model is trained using the training dataset to obtain a pre-trained AIGC large model, including: S41. Divide the training dataset into a training set, a validation set, and a test set according to a preset ratio; S42. Set the loss function and optimizer for the AIGC large model; S43. Input the training set into the AIGC large model for model training; S44. Calculate the loss value based on the loss function, and the optimizer updates the model parameters based on the loss value and network gradient information; S45. If the loss value is less than the preset threshold, the model training ends and the current AIGC large model is the trained AIGC large model. Otherwise, return to S43 and continue to train the model using the training set. S46. Input the validation set into the trained AIGC large model, evaluate the model's generalization ability by observing its performance on the validation set, and fine-tune the model's hyperparameters and structure. S47. Input the test set into the optimized AIGC large model, evaluate the model performance, and finally obtain the pre-trained AIGC large model.
7. The hospital consumables management system based on AIGC and VMI according to claim 1, characterized in that: The warehouse management module includes replenishment order management, replenishment order not generated history query, replenishment order approval, arrival notification management, pre-sale return application form, return approval, receiving and acceptance form, overage report management, warehouse entry order accounting, and warehouse entry order detailed accounting.
8. The hospital consumables management system based on AIGC and VMI according to claim 1, characterized in that: The outbound management module includes requisition order management, requisition order approval, packing order management, outbound order management, outbound order approval, outbound order accounting, outbound order detail accounting, outbound order confirmation, outbound order detail confirmation, transfer order management, and damage report management.
9. The hospital consumables management system based on AIGC and VMI according to claim 1, characterized in that: The warehouse management module includes dynamic inventory counting, static inventory counting, detailed inquiry of inbound and outbound slips, inventory inquiry, warehouse adjustment management, and invoice settlement management.
10. The hospital consumables management system based on AIGC and VMI according to claim 1, characterized in that: The management visualization module utilizes AIGC to automatically generate diverse media content, including text, images, audio, and video, using artificial intelligence technology to achieve visualized management of hospital consumables, including: The dynamic dashboard generates a real-time inventory dashboard for consumables, displaying the inventory levels, consumption trends, and warning thresholds of consumables in each department. It also embeds a 3D model of consumables automatically generated by AIGC, supporting rotation and zoom interaction. Intelligent reporting automatically generates weekly / monthly consumables analysis reports, including procurement recommendations, waste rate analysis, and comparison of doctors' consumables usage habits, and inserts dynamic charts automatically generated by AIGC; Contextualized videos are generated to recreate surgical scenarios for complex consumables, with annotations of consumable usage steps and precautions, and demonstrations combined with AIGC-generated voice narration. Voice interaction and reminders: It uses voice synthesis technology to announce inventory warnings and supports voice queries. AIGC automatically generates answers and displays them visually.