Operating room high-value consumable management system based on artificial intelligence

Abstract

The invention provides an operating room high-value consumable management system based on artificial intelligence, and relates to the technical field of artificial intelligence, an intelligent sensing module collects position information, state information and use conditions of high-value consumables in real time through RFID tags and sensors, and identification and tracking of the consumables are achieved. And the artificial intelligence analysis module analyzes the use frequency of the consumables and the relevance between the operation type and the demand based on a deep learning algorithm so as to predict the future consumable demand. And the inventory management module dynamically adjusts the inventory according to the actual use condition and the predicted demand. The operation auxiliary module automatically generates a consumable list according to the operation type and the patient condition, and medical staff is reminded to prepare. And the compliance supervision module records full life cycle data of the consumables, provides a traceability report and ensures compliance. Through the technical means, the operation cost of a hospital is remarkably reduced, the operation safety and the satisfaction degree of a patient are improved, and wide application prospects and important social value are achieved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based high-value consumables management system for operating rooms. Background Art

[0002] High-value consumables in operating rooms (such as artificial joints, heart stents, etc.) have become the focus and difficulty of hospital management due to their high prices, wide variety, high frequency of use and significant impact on the success rate of surgery. At present, traditional consumables management methods mainly rely on manual records and inventory management, which has the following problems:

[0003] (1) Difficulty in tracking consumables: The use of consumables during surgery is difficult to track in real time, which can easily lead to omissions or errors.

[0004] (2) Inaccurate inventory management: Delayed updating of consumables inventory information may lead to insufficient inventory or backlogs.

[0005] (3) Difficulty in cost control: Lack of intelligent analysis of consumables usage data makes it difficult to optimize procurement and usage strategies.

[0006] (4) Insufficient regulatory compliance: Unable to meet the requirements of medical regulatory authorities for full life cycle tracking of high-value consumables.

[0007] Therefore, there is an urgent need for an intelligent management system based on artificial intelligence that can realize full-process intelligent management of high-value consumables. Summary of the invention

[0008] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide an artificial intelligence-based high-value consumables management system for operating rooms. Through various technical means such as intelligent perception, deep learning analysis, dynamic inventory management and compliance supervision, the present invention can significantly improve the hospital's management efficiency, reduce operating costs, optimize resource allocation, and improve surgical safety and patient satisfaction. It has broad application prospects and important social value.

[0009] To achieve the above object, the present invention provides the following solutions:

[0010] An artificial intelligence-based high-value consumables management system for operating rooms, comprising:

[0011] Intelligent sensing module, used to identify and track high-value consumables, and collect the location information, status information and usage of the high-value consumables in real time;

[0012] An artificial intelligence analysis module, for analyzing the usage frequency of the high-value consumables, the correlation between the type of surgery and the consumables demand based on the deep learning algorithm and the location information, the status information and the usage status, so as to predict the future consumables demand;

[0013] An inventory management module, used to dynamically adjust the inventory of the high-value consumables according to the usage and the future consumables demand;

[0014] The surgical assistance module is used to automatically generate a list of consumables based on the preset surgical type and patient condition to remind medical staff to prepare consumables;

[0015] The compliance supervision module is used to record the full life cycle data of the high-value consumables and provide a traceability report on the use of the high-value consumables.

[0016] Preferably, the intelligent sensing module includes:

[0017] An identification submodule, used to attach a unique radio frequency identification tag to each of the high-value consumables to identify the unique identity information of the high-value consumables;

[0018] A two-dimensional code submodule, used for printing a spare code scanning mark on each of the high-value consumables to identify the unique identity information of the high-value consumables; the spare code scanning mark includes a two-dimensional code and a barcode;

[0019] The reader / writer submodule is used to install the reader / writer at the entrance of the consumables warehouse and the door of the operating room, so as to monitor the entry and exit and circulation of the corresponding high-value consumables through the unique identity information, and obtain the usage and location information;

[0020] The image recognition submodule is used to install a camera in the operating room or consumables storage area to combine computer vision technology to determine the status information through the image information of the high-value consumables.

[0021] Preferably, the intelligent sensing module further includes:

[0022] The sensor submodule is used to embed a sensor device on the packaging of each of the high-value consumables, and collect environmental information of the high-value consumables in real time through the sensor device.

[0023] Preferably, the image recognition submodule includes:

[0024] An image denoising unit, used to perform image denoising on the image information to obtain a denoised image;

[0025] An image enhancement unit, used for performing image enhancement on the denoised image to obtain an enhanced image;

[0026] A feature extraction unit, used to extract features from the enhanced image to obtain image features;

[0027] The state recognition unit is used to input the image features into a trained state recognition model to obtain the state information.

[0028] Preferably, the image enhancement unit comprises:

[0029] A window construction subunit, used for constructing a moving image enhancement window according to the scale of the denoised image;

[0030] The pixel difference calculation subunit is used to calculate the pixel difference according to the formula F ij =|x ij -y ij |Get the pixel difference of each pixel under each image enhancement window; where F ij represents the pixel difference of the denoised image at position (i, j), x ij Represents the pixel value of the image information at position (i, j), y ij Represents the pixel value of the denoised image at position (i, j);

[0031] The correction difference calculation subunit is used to calculate the difference according to the formula Perform brightness uniformity correction on the pixel difference to obtain a corrected difference; wherein, D ij is the correction difference of the pixel point, D dark is the dark signal value of the pixel point, K(i) is the correction coefficient;

[0032] A feature point recording subunit, used to take the pixel points whose correction difference is greater than the feature threshold as feature points and record the number;

[0033] A proportion value determination subunit, used to take the proportion of the feature points to all the pixels under the image enhancement window as the proportion value;

[0034] A denoising subunit, configured to perform median denoising on the denoised image in the corresponding moving image enhancement window to obtain median denoising data when the proportion value in the moving image enhancement window is greater than a preset threshold;

[0035] The enhancement subunit is used to slide the moving image enhancement window until the entire median denoising data is traversed to obtain the enhanced image.

[0036] Preferably, the feature extraction unit comprises:

[0037] A normalization subunit, used for performing image scaling and image normalization processing on the enhanced image to obtain a normalized image;

[0038] The extraction subunit is used to perform feature extraction on the normalized image using a preset VGG16 network to obtain the image features.

[0039] Preferably, the artificial intelligence analysis module includes:

[0040] The sample data acquisition submodule is used to acquire a sample location information data set, a sample status information data set, a sample usage data set, and operation-related information; the operation-related information includes the operation type, operation duration, patient medical record information, and doctor's operation habits;

[0041] A labeling submodule is used to label the sample location information data set, the sample status information data set and the sample usage data set to obtain a training data set; the labeling content includes: the usage frequency of high-value consumables, the correlation between high-value consumables and surgical types, and the demand for high-value consumables;

[0042] A training submodule, used for inputting the training data set into the initial neural network for training to obtain a trained future demand forecasting model;

[0043] The prediction submodule is used to input the location information, the state information and the usage status into the future demand prediction model to obtain the future consumables demand.

[0044] Preferably, the initial neural network is a CNN-LSTM combined network model.

[0045] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0046] The present invention provides an operating room high-value consumables management system based on artificial intelligence, including: an intelligent perception module for identifying and tracking high-value consumables, and collecting the location information, status information and usage of the high-value consumables in real time; an artificial intelligence analysis module for analyzing the usage frequency of the high-value consumables, the correlation between the type of surgery and the consumables demand based on the location information, the status information and the usage, so as to predict the future consumables demand; an inventory management module for dynamically adjusting the inventory of the high-value consumables according to the usage and the future consumables demand; a surgical auxiliary module for automatically generating a list of consumables according to the preset surgical type and patient condition to remind medical staff to prepare consumables; a compliance supervision module for recording the full life cycle data of the high-value consumables and providing a traceability report on the use of the high-value consumables. The present invention can significantly improve the management efficiency of the hospital, reduce operating costs, optimize resource allocation, and improve surgical safety and patient satisfaction through a variety of technical means such as intelligent perception, deep learning analysis, dynamic inventory management and compliance supervision, and has broad application prospects and important social value. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0048] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] 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.

[0050] The purpose of this invention is to provide an artificial intelligence-based high-value consumables management system for operating rooms. Through various technical means such as intelligent perception, deep learning analysis, dynamic inventory management and compliance supervision, it can significantly improve the hospital's management efficiency, reduce operating costs, optimize resource allocation, and improve surgical safety and patient satisfaction. It has broad application prospects and important social value.

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides an artificial intelligence-based high-value consumables management system for operating rooms, comprising:

[0053] Intelligent sensing module, used to identify and track high-value consumables, and collect the location information, status information and usage of the high-value consumables in real time;

[0054] An artificial intelligence analysis module, for analyzing the usage frequency of the high-value consumables, the correlation between the type of surgery and the consumables demand based on the deep learning algorithm and the location information, the status information and the usage status, so as to predict the future consumables demand;

[0055] An inventory management module, used to dynamically adjust the inventory of the high-value consumables according to the usage and the future consumables demand;

[0056] The surgical assistance module is used to automatically generate a list of consumables based on the preset surgical type and patient condition to remind medical staff to prepare consumables;

[0057] The compliance supervision module is used to record the full life cycle data of the high-value consumables and provide a traceability report on the use of the high-value consumables.

[0058] Preferably, the intelligent sensing module includes:

[0059] An identification submodule, used to attach a unique radio frequency identification tag to each of the high-value consumables to identify the unique identity information of the high-value consumables;

[0060] A two-dimensional code submodule, used for printing a spare code scanning mark on each of the high-value consumables to identify the unique identity information of the high-value consumables; the spare code scanning mark includes a two-dimensional code and a barcode;

[0061] The reader / writer submodule is used to install the reader / writer at the entrance of the consumables warehouse and the door of the operating room, so as to monitor the entry and exit and circulation of the corresponding high-value consumables through the unique identity information, and obtain the usage and location information;

[0062] The image recognition submodule is used to install a camera in the operating room or consumables storage area to combine computer vision technology to determine the status information through the image information of the high-value consumables.

[0063] Specifically, the core function of the identification submodule of this embodiment is to attach a unique radio frequency identification (RFID) tag to each high-value consumable. Each RFID tag contains a unique identification code that can uniquely identify the consumable throughout the consumable life cycle. When the hospital purchases high-value consumables, the supplier will pre-attach an RFID tag to the packaging of each consumable. The hospital's management system will associate the unique identification code of each tag with the detailed information of the consumable (such as name, specification, production batch, expiration date, etc.) and store it in the database. In this way, the hospital can quickly identify and track each consumable to ensure the accuracy and completeness of the information during use and management.

[0064] Optionally, the present embodiment is responsible for printing spare scan code labels, including QR codes and bar codes, on the packaging of each high-value consumable. These QR codes and bar codes also contain the unique identity information of the consumables as a supplement to the RFID tags. When the consumables are put into storage, the hospital uses a printer to print the QR code and bar code on the consumable packaging, and synchronizes it with the information on the RFID tag to the management system. When using consumables, medical staff can quickly scan the QR code or bar code with a handheld scanning device to obtain detailed information on the consumables. This dual identification method improves the flexibility of consumable management and ensures that when the RFID signal is poor or the equipment fails, it can still be identified and tracked by scanning the code.

[0065] Furthermore, the reading and writing submodule of this embodiment realizes the monitoring of the entry and exit and circulation of high-value consumables by installing RFID readers at the entrance of the consumables warehouse and the door of the operating room. Whenever the consumables pass through the reader, the system will automatically read the information of the RFID tag and record the entry and exit time, location and status of the consumables. During the implementation process, the reader is connected to the hospital's management system in real time to ensure the real-time update of data. In this way, the hospital can grasp the inventory status and circulation of consumables in real time, avoiding errors and omissions caused by manual records. In addition, the reader can also be linked with the inventory management module to automatically adjust the inventory data to ensure the accuracy of inventory information.

[0066] Furthermore, the image recognition submodule of this embodiment monitors the status information of high-value consumables in real time by installing a camera in the operating room or consumables storage area, combined with computer vision technology. During the implementation process, the camera will regularly take images of consumables and transmit the image data to the image processing system. The system analyzes the status of consumables in the image (such as whether it is unsealed, whether it is expired, etc.) through techniques such as image denoising, enhancement and feature extraction. The trained state recognition model will classify the extracted features, determine the current state of the consumables, and feed the results back to the management system. This image recognition technology not only improves the accuracy of state monitoring, but also can monitor the use of consumables in real time during surgery to ensure surgical safety.

[0067] Preferably, the intelligent sensing module further includes:

[0068] The sensor submodule is used to embed a sensor device on the packaging of each of the high-value consumables, and collect environmental information of the high-value consumables in real time through the sensor device.

[0069] Specifically, in this embodiment, a small sensor device is embedded in the packaging of each high-value consumable, and these sensors can monitor and collect environmental information of the consumables in real time, such as temperature, humidity, vibration and light. In the specific implementation process, when the hospital purchases high-value consumables, the supplier will embed sensors in the packaging of the consumables, and these sensors are connected to the hospital's management system through wireless communication technology (such as Bluetooth, ZigBee or Wi-Fi). The sensor regularly collects environmental data and uploads the data to the central management system in real time to ensure the timeliness and accuracy of the data. The system will store and analyze the collected environmental information for subsequent use. And this embodiment takes into account that the real-time collection of environmental information is crucial for the management of high-value consumables. First of all, many high-value consumables (such as biomaterials, medicines, etc.) have strict requirements for the storage environment, and changes in temperature and humidity may affect their effectiveness and safety. By monitoring these environmental parameters, the hospital can promptly detect and deal with potential environmental anomalies to ensure that the consumables are kept in safe storage conditions before use. Secondly, the analysis of environmental information can help hospitals optimize storage conditions and management processes. For example, by analyzing historical environmental data, hospitals can identify which storage areas have unstable environmental conditions and take measures to improve the management of these areas, reduce the risk of consumables loss and expiration, and ultimately improve the hospital's operational efficiency and patient safety.

[0070] Preferably, the image recognition submodule includes:

[0071] An image denoising unit, used to perform image denoising on the image information to obtain a denoised image;

[0072] An image enhancement unit, used for performing image enhancement on the denoised image to obtain an enhanced image;

[0073] A feature extraction unit, used to extract features from the enhanced image to obtain image features;

[0074] The state recognition unit is used to input the image features into a trained state recognition model to obtain the state information.

[0075] Specifically, this embodiment can also select other filtering methods, such as median filtering, which is a nonlinear filtering technology that is particularly suitable for removing salt and pepper noise. This method effectively retains edge information and reduces the impact of noise by replacing the value of each pixel with the median of the pixel values ​​in its neighborhood. Median filtering performs well when processing images with sharp noise. Bilateral filtering can also be selected. Bilateral filtering is a denoising method that considers both spatial distance and pixel value similarity. It performs weighted averaging of pixels in the neighborhood, and the weight is determined by the spatial distance and the pixel value difference. This method can retain edge details while denoising, and is suitable for applications that need to maintain image structure. This embodiment can also select wavelet transform, which is a frequency domain denoising technology that separates components of different frequencies by converting the image to the wavelet domain. Then, the high-frequency components can be thresholded to remove noise, and finally the processed components are inversely transformed back to the spatial domain. Wavelet denoising is effective in processing images with multi-scale features. In addition, this embodiment can also select non-local mean (NLM), and the non-local mean denoising method denoises by considering the similarity of all pixels in the image. This method calculates a weighted average of each pixel, with the weights based on the similarity of other pixels similar to that pixel. This method can effectively preserve image details while removing noise and is applicable to various types of noise.

[0076] Preferably, the image enhancement unit comprises:

[0077] A window construction subunit, used for constructing a moving image enhancement window according to the scale of the denoised image;

[0078] The pixel difference calculation subunit is used to calculate the pixel difference according to the formula F ij =|x ij -y ij |Get the pixel difference of each pixel under each image enhancement window; where F ij represents the pixel difference of the denoised image at position (i, j), x ij Represents the pixel value of the image information at position (i, j), y ij Represents the pixel value of the denoised image at position (i, j);

[0079] The correction difference calculation subunit is used to calculate the difference according to the formula Perform brightness uniformity correction on the pixel difference to obtain a corrected difference; wherein, D ij is the correction difference of the pixel point, D dark is the dark signal value of the pixel point, K(i) is the correction coefficient;

[0080] A feature point recording subunit, used to take the pixel points whose correction difference is greater than the feature threshold as feature points and record the number;

[0081] A proportion value determination subunit, used to take the proportion of the feature points to all the pixels under the image enhancement window as the proportion value;

[0082] A denoising subunit, configured to perform median denoising on the denoised image in the corresponding moving image enhancement window to obtain median denoising data when the proportion value in the moving image enhancement window is greater than a preset threshold;

[0083] The enhancement subunit is used to slide the moving image enhancement window until the entire median denoising data is traversed to obtain the enhanced image.

[0084] Preferably, the feature extraction unit comprises:

[0085] A normalization subunit, used for performing image scaling and image normalization processing on the enhanced image to obtain a normalized image;

[0086] The extraction subunit is used to perform feature extraction on the normalized image using a preset VGG16 network to obtain the image features.

[0087] Preferably, the artificial intelligence analysis module includes:

[0088] The sample data acquisition submodule is used to acquire a sample location information data set, a sample status information data set, a sample usage data set, and operation-related information; the operation-related information includes the operation type, operation duration, patient medical record information, and doctor's operation habits;

[0089] A labeling submodule is used to label the sample location information data set, the sample status information data set and the sample usage data set to obtain a training data set; the labeling content includes: the usage frequency of high-value consumables, the correlation between high-value consumables and surgical types, and the demand for high-value consumables;

[0090] A training submodule, used for inputting the training data set into the initial neural network for training to obtain a trained future demand forecasting model;

[0091] The prediction submodule is used to input the location information, the state information and the usage status into the future demand prediction model to obtain the future consumables demand.

[0092] Specifically, the main function of the sample data acquisition submodule is to collect relevant data from the intelligent perception module and the hospital management system to build a basic data set for training and prediction. This embodiment will regularly extract sample location information data sets (such as storage location and circulation records of consumables), sample status information data sets (such as environmental monitoring data and consumable status) and sample usage data sets (such as the use time, quantity and department of consumables) from the database. In addition, surgery-related information (including surgery type, surgery duration, patient medical record information and doctor's operating habits) will also be collected. These data will be integrated into a comprehensive data set to provide a basis for subsequent analysis and model training.

[0093] Optionally, the labeling submodule is responsible for labeling the acquired sample data to generate a training data set. During the implementation process, this embodiment will classify and mark the sample location information, status information and usage according to preset labeling rules. The labeling content includes the frequency of use of high-value consumables (such as high frequency, medium frequency, low frequency), the correlation between high-value consumables and surgical types (such as the types and quantities of consumables commonly used in a certain surgery), and the demand for high-value consumables (such as demand forecasts within a specific time period). In this way, the labeling submodule converts the original data into a structured training data set, which is convenient for subsequent model training and analysis.

[0094] Specifically, the core function of the training submodule is to input the labeled training data set into the initial neural network for training to generate a future demand forecasting model. This embodiment selects a suitable deep learning framework (such as TensorFlow or PyTorch), builds an initial neural network model (such as a CNN-LSTM combined network model), and sets training parameters (such as learning rate, batch size, and number of training rounds). During the training process, the system will continuously adjust the model parameters through the back propagation algorithm to minimize the prediction error. After the training is completed, the system will save the trained model for future demand forecasting.

[0095] Furthermore, the function of the prediction submodule is to use the trained future demand prediction model to analyze the real-time collected location information, status information and usage to predict future consumables demand. In this embodiment, real-time data is input into the trained model, and the model will generate prediction results for future consumables demand based on historical data and current status. These prediction results will be used to dynamically adjust inventory, optimize procurement plans and improve surgical preparation efficiency. The system will also regularly update the prediction model to adapt to new data and changing needs to ensure the accuracy and reliability of the prediction.

[0096] It should be noted that the main function of the inventory management module of this embodiment is to dynamically adjust the inventory of high-value consumables based on real-time usage and future consumables demand. The module will regularly obtain the latest consumables demand forecast data and actual usage data from the artificial intelligence analysis module. Based on these data, the system will calculate the safety stock level and optimal inventory of each consumable and compare them with the current inventory. When the inventory is lower than the safety level, the system will automatically generate replenishment suggestions and notify the purchasing department to replenish; when the inventory is in excess, the system will prompt the manager to clean up the inventory or adjust the purchasing plan. In addition, the inventory management module will also record the detailed information of each inventory change for subsequent analysis and auditing.

[0097] The function of the surgical assistance module is to automatically generate a list of consumables based on the preset surgical type and patient condition to remind medical staff to prepare consumables. The module will obtain information about upcoming surgeries from the hospital's scheduling system, including the type of surgery, surgery time, and basic patient information. The system will automatically match the required high-value consumables and generate a detailed list of consumables based on historical data and the doctor's operating habits. The list will be presented to medical staff through a visual interface to ensure that they can prepare the required consumables in time before the operation. In addition, the surgical assistance module can also update the use of consumables in real time according to the progress of the operation to ensure that any sudden needs can be responded to quickly during the operation.

[0098] The main function of the compliance supervision module is to record the full life cycle data of high-value consumables and provide traceability reports to meet the requirements of medical supervision. The module automatically collects and stores relevant information for each high-value consumable, including procurement records, storage conditions, usage records, and scrapping information. The system ensures the integrity and accuracy of all data and regularly generates compliance reports to show the usage and compliance status of consumables. These reports will be used for internal audits and external regulatory inspections to ensure that hospitals comply with relevant laws, regulations and industry standards in the management of high-value consumables. In addition, the compliance supervision module can also provide real-time alarm functions. When it is found that consumables are used improperly or storage conditions do not meet the requirements, the system will immediately notify relevant managers to handle it.

[0099] The beneficial effects of the present invention are as follows:

[0100] (1) The intelligent sensing module of the present invention uses technologies such as RFID tags, QR codes and sensors to achieve real-time identification and tracking of high-value consumables, ensuring that the storage and circulation information of consumables is updated in a timely manner, reducing errors and omissions in manual records; through automated readers and image recognition technology, the system can collect the location information, status information and usage of consumables in real time, reducing manual intervention and improving management efficiency.

[0101] (2) The artificial intelligence analysis module of the present invention uses a deep learning algorithm (such as a CNN-LSTM combined network model) to analyze the frequency of consumables use, the correlation between surgical types and consumables demand, and can accurately predict future consumables demand to avoid insufficient or excessive inventory. By analyzing historical data, the system can provide data-driven decision support for hospital management and optimize procurement and inventory management strategies.

[0102] (3) The inventory management module of the present invention dynamically adjusts the inventory of high-value consumables based on real-time usage and future demand forecasts to ensure that the inventory level matches actual demand and reduce inventory costs. Through accurate demand forecasting and inventory management, the waste of high-value consumables due to expiration or surplus is reduced, thereby improving resource utilization.

[0103] (4) The surgical assistance module of the present invention automatically generates a list of consumables based on the preset surgical type and patient condition, reminding medical staff to prepare consumables, thereby reducing pre-operative preparation time and improving surgical efficiency. The present invention can provide personalized consumables recommendations based on the doctor's operating habits and the patient's specific conditions to ensure the smooth progress of the operation.

[0104] (5) The compliance supervision module of the present invention records the full life cycle data of high-value consumables, including procurement, storage, use and scrapping, to ensure that the compliance requirements of medical supervision are met; the present invention can generate traceability reports on the use of consumables, support auditing and supervision, and improve the management transparency and compliance of hospitals.

[0105] (6) The image recognition submodule of the present invention can improve the recognition accuracy of consumable status information through image denoising, enhancement and feature extraction technologies, ensuring that the system can accurately judge the status of consumables (such as whether it is opened, whether it is expired, etc.); combined with computer vision technology, the system can monitor the status of consumables in real time, detect abnormal situations in time, and ensure surgical safety.

[0106] (7) The present invention can continuously optimize and update the prediction model according to actual usage and deviations from the prediction results, thereby improving the accuracy and reliability of the prediction; and can be adjusted and optimized according to the actual conditions of different hospitals or departments, and has good adaptability and scalability.

[0107] (8) The present invention can provide an intuitive visual interface, which facilitates medical staff and management personnel to query consumables information in real time, thereby improving user experience; and the present invention can automatically remind medical staff to replenish or prepare consumables based on the usage and inventory status of consumables, thereby reducing human omissions.

[0108] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0109] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. An artificial intelligence-based high-value consumables management system for operating rooms, characterized in that: include: Intelligent sensing module, used to identify and track high-value consumables, and collect the location information, status information and usage of the high-value consumables in real time; An artificial intelligence analysis module, for analyzing the usage frequency of the high-value consumables, the correlation between the type of surgery and the consumables demand based on the deep learning algorithm and the location information, the status information and the usage status, so as to predict the future consumables demand; An inventory management module, used to dynamically adjust the inventory of the high-value consumables according to the usage and the future consumables demand; The surgical assistance module is used to automatically generate a list of consumables based on the preset surgical type and patient condition to remind medical staff to prepare consumables; The compliance supervision module is used to record the full life cycle data of the high-value consumables and provide a traceability report on the use of the high-value consumables.

2. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 1 is characterized in that: The intelligent sensing module comprises: An identification submodule, used to attach a unique radio frequency identification tag to each of the high-value consumables to identify the unique identity information of the high-value consumables; A two-dimensional code submodule, used for printing a spare code scanning mark on each of the high-value consumables to identify the unique identity information of the high-value consumables; the spare code scanning mark includes a two-dimensional code and a barcode; The reader / writer submodule is used to install the reader / writer at the entrance of the consumables warehouse and the door of the operating room, so as to monitor the entry and exit and circulation of the corresponding high-value consumables through the unique identity information, and obtain the usage and location information; The image recognition submodule is used to install a camera in the operating room or consumables storage area to combine computer vision technology to determine the status information through the image information of the high-value consumables.

3. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 1 is characterized in that: The intelligent sensing module also includes: The sensor submodule is used to embed a sensor device on the packaging of each of the high-value consumables, and collect environmental information of the high-value consumables in real time through the sensor device.

4. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 2 is characterized in that: The image recognition submodule comprises: An image denoising unit, used to perform image denoising on the image information to obtain a denoised image; An image enhancement unit, used for performing image enhancement on the denoised image to obtain an enhanced image; A feature extraction unit, used to extract features from the enhanced image to obtain image features; The state recognition unit is used to input the image features into a trained state recognition model to obtain the state information.

5. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 4 is characterized in that: The image enhancement unit comprises: A window construction subunit, used for constructing a moving image enhancement window according to the scale of the denoised image; The pixel difference calculation subunit is used to calculate the pixel difference according to the formula F ij =|x ij -y ij |Get the pixel difference of each pixel under each image enhancement window; where F ij represents the pixel difference of the denoised image at position (i, j), x ij Represents the pixel value of the image information at position (i, j), y ij Represents the pixel value of the denoised image at position (i, j); The correction difference calculation subunit is used to calculate the difference according to the formula Perform brightness uniformity correction on the pixel difference to obtain a corrected difference; wherein, D ij is the correction difference of the pixel point, D dark is the dark signal value of the pixel point, K(i) is the correction coefficient; A feature point recording subunit, used to take the pixel points whose correction difference is greater than the feature threshold as feature points and record the number; A proportion value determination subunit, used to take the proportion of the feature points to all the pixels under the image enhancement window as the proportion value; A denoising subunit, configured to perform median denoising on the denoised image in the corresponding moving image enhancement window to obtain median denoising data when the proportion value in the moving image enhancement window is greater than a preset threshold; The enhancement subunit is used to slide the moving image enhancement window until the entire median denoising data is traversed to obtain the enhanced image.

6. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 4 is characterized in that: The feature extraction unit comprises: A normalization subunit, used for performing image scaling and image normalization processing on the enhanced image to obtain a normalized image; The extraction subunit is used to perform feature extraction on the normalized image using a preset VGG16 network to obtain the image features.

7. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 4 is characterized in that: The artificial intelligence analysis module includes: The sample data acquisition submodule is used to acquire a sample location information data set, a sample status information data set, a sample usage data set, and operation-related information; the operation-related information includes the operation type, operation duration, patient medical record information, and doctor's operation habits; A labeling submodule is used to label the sample location information data set, the sample status information data set and the sample usage data set to obtain a training data set; the labeling content includes: the usage frequency of high-value consumables, the correlation between high-value consumables and surgical types, and the demand for high-value consumables; A training submodule, used for inputting the training data set into the initial neural network for training to obtain a trained future demand forecasting model; The prediction submodule is used to input the location information, the state information and the usage status into the future demand prediction model to obtain the future consumables demand.

8. The artificial intelligence-based high-value consumables management system for operating rooms according to claim 7 is characterized in that: The initial neural network is a CNN-LSTM combined network model.

Images