First-aid trolley article checking system based on intelligent identification

Abstract

The invention relates to the technical field of medical equipment management, and particularly discloses a first-aid trolley article checking system based on intelligent identification, an intelligent identification module is integrated with an image identification unit and a bar code / two-dimensional code identification unit, and the image identification unit establishes a medical article image database through a deep learning algorithm; identification of types, specifications and quantity of articles in the storage grids is realized; the latter scans and acquires information such as production date and validity period of the article. And the system processing module integrates the multi-source data, judges the quantity difference and the validity period state according to a preset rule, generates a visual inventory report and synchronizes the visual inventory report to a hospital information management system. And the alarm prompting module adopts an acousto-optic and wireless dual alarm mechanism to ensure that abnormal conditions are handled in time. The user interaction module supports touch screen operation and voice instructions, and the power management module provides dual-power switching and electric quantity monitoring. The checking efficiency and accuracy of the first-aid appliances are remarkably improved, the manual error rate is reduced, and intelligent management of medical resources is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment management, and in particular to an inventory system for emergency trolley supplies based on intelligent identification. Background Art

[0002] In the modern medical system, emergency carts are key equipment for emergency medical rescue. The completeness, accuracy and timeliness of the medical supplies stored in them are directly related to the patient's life safety and treatment effect.

[0003] The traditional method of inventorying supplies on emergency trolleys relies primarily on manual counting. This method has many drawbacks. On the one hand, manual inventorying consumes a significant amount of time and manpower. In a busy medical work environment, medical staff need to devote extra energy to inventorying, which undoubtedly increases their workload. On the other hand, manual operations are prone to omissions, such as incorrect identification of the type and specifications of medical supplies, and deviations in quantity counting, which can lead to missing or incorrect use of supplies during emergency treatment. Even more seriously, it is difficult for manual labor to accurately control the expiration dates of medical supplies. For example, the incident in Xi'an Yanta Renji Hospital, where expired drugs were found in emergency trolleys, highlighted the shortcomings of manual inventory in expiration date supervision, which posed a great threat to patients' lives.

[0004] The rapid advancement of technology, particularly the gradual penetration of artificial intelligence and the Internet of Things (IoT) in the medical field, has created new opportunities for solving the problem of inventorying emergency cart supplies. Intelligent recognition technologies such as image recognition and barcode / QR code recognition have achieved significant success in other fields, making their application in emergency cart inventory management a natural progression. For example, in medical image analysis, image recognition technology powered by deep learning algorithms can help doctors quickly detect lesions, demonstrating the potential of intelligent recognition technology in medical settings. Furthermore, with the continuous advancement of hospital information technology and the widespread use of medical information management systems, the need for data exchange between emergency cart inventory systems and hospital medical information management systems is becoming increasingly urgent to achieve efficient integration and management of medical resources.

[0005] Therefore, an inventory system for emergency trolley supplies based on intelligent identification has become an urgent problem to be solved. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide an inventory system for emergency trolley supplies based on intelligent identification. Through intelligent identification technology, it can realize rapid and accurate inventory of emergency trolley supplies, monitor the status of supplies in real time, improve the intelligence level and work efficiency of medical resource management, and ensure the smooth implementation of emergency work.

[0007] To solve the above technical problems, the present invention provides a technical solution: an emergency trolley supplies inventory system based on intelligent identification, comprising:

[0008] An intelligent recognition module, comprising an image recognition unit and a barcode / QR code recognition unit. The image recognition unit is used to capture and identify images of medical supplies in the storage compartment of the emergency cart, and the barcode / QR code recognition unit is used to scan barcodes or QR codes on the medical supplies.

[0009] The system processing module is connected to the intelligent recognition module, and is used to receive and process the information transmitted by the intelligent recognition module and generate inventory results according to preset rules;

[0010] A data storage module, connected to the system processing module, for storing a medical supplies image database, basic information of medical supplies, and inventory result data;

[0011] An alarm prompt module, connected to the system processing module, is used to send an alarm signal when the system processing module determines that the medical supplies have an abnormal situation;

[0012] The user interaction module is connected to the system processing module and is used by medical staff to operate and view inventory results;

[0013] The power management module provides power supply for the entire system.

[0014] Furthermore, the image recognition unit adopts a deep learning algorithm to establish a medical supply image database by training medical supply images.

[0015] Furthermore, during inventory, the camera captures images of the storage compartments and transmits the images to the system processing module, which compares the images with standard images in the database to identify the types, specifications, and quantities of medical supplies.

[0016] Furthermore, the barcode / QR code recognition unit scans the barcode or QR code to obtain relevant information of the medical supplies, including name, specification, production date, and expiration date, and transmits it to the system processing module.

[0017] Furthermore, the system processing module determines whether the quantity of medical supplies is accurate and whether any of them are expired according to preset inventory rules, and generates inventory results.

[0018] Furthermore, the system processing module exchanges data with the hospital's medical information management system and uploads the inventory results to the hospital information management system.

[0019] Furthermore, the alarm prompt module includes an audible and visual alarm fixed on the emergency cart and a wireless alarm device connected to the medical staff's mobile phone or terminal equipment.

[0020] Furthermore, the user interaction module includes a touch screen fixed on the emergency cart, which is used by medical staff to start, pause, and query historical inventory records for inventory operations, and to display inventory results and status information of emergency cart supplies in real time.

[0021] Furthermore, the user interaction module is also provided with a voice recognition unit for medical staff to perform relevant operations through voice commands.

[0022] Furthermore, the power management module is powered by a rechargeable lithium battery and an external power supply, monitors the battery power and power status in real time, and displays relevant information on the user interaction module.

[0023] The advantages of the present invention compared with the prior art are:

[0024] The present invention realizes the rapid and automatic inventory of emergency trolley supplies through image recognition and barcode / QR code recognition technology, avoids the tediousness and errors of manual inventory, and greatly improves the efficiency and accuracy of inventory.

[0025] The present invention can monitor the quantity, expiration date and other status information of medical supplies in real time. Once an abnormal situation occurs, an alarm will be issued immediately, so that medical staff can replenish and replace supplies in time to ensure the smooth progress of emergency work.

[0026] The present invention interacts with the hospital's medical information management system to achieve unified management and sharing of medical supplies information, facilitates query and statistics by medical staff, and improves the level of informatization of medical resource management.

[0027] The touch screen and voice interaction functions of the user interaction module make operations more convenient for medical staff and adapt to different usage scenarios and needs.

[0028] The dual power supply mode and power monitoring function of the power management module ensure that the system can run stably in various situations without being restricted by power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The present invention is a system block diagram of an emergency trolley supplies inventory system based on intelligent identification. DETAILED DESCRIPTION

[0030] Various exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention.

[0031] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0032] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0033] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0034] The following is a further detailed description of an emergency trolley supplies inventory system based on intelligent identification according to the present invention with reference to the accompanying drawings.

[0035] Combined with attachment Figure 1 , the present invention is introduced in detail.

[0036] An inventory system for emergency trolley supplies based on intelligent recognition, including the following parts:

[0037] Intelligent recognition module:

[0038] Image Recognition Unit: A high-definition camera is installed above each storage compartment of the emergency cart to capture images of the medical supplies within. This camera features wide-angle photography, autofocus, and low-light compensation, enabling clear images of supplies in a variety of environments. The image recognition unit uses a deep learning algorithm to build a medical supply image database through training on a large number of medical supply images. The specific method is as follows:

[0039] (1) Data collection: Collect images of various medical supplies under different angles, lighting conditions, and placement methods to form an initial image dataset.

[0040] (2) Data preprocessing: Cleaning, enhancing, and labeling the collected images. The cleaning process includes removing blurry, duplicate, or poor-quality images; the enhancement process expands the dataset through rotation, flipping, scaling, brightness adjustment, etc.; and the labeling process labels the category, specifications, and other information of the medical supplies in each image.

[0041] (3) Model construction: A convolutional neural network (CNN) architecture is used to build a deep learning model. The model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer, which are used to extract image features and perform classification and recognition.

[0042] (4) Model training: The preprocessed image dataset is divided into a training set, a validation set, and a test set. The model is trained using the training set, and the model parameters are continuously adjusted through the back-propagation algorithm to minimize the error between the prediction results and the annotation information.

[0043] (5) Model optimization: Optimize model performance and improve recognition accuracy by adjusting hyperparameters, increasing training data, and applying regularization techniques (such as L2 regularization and Dropout).

[0044] (6) Model deployment: The trained model is deployed to the system processing module for real-time recognition of medical supply images. During inventory, the camera captures images of the storage compartments and transmits them to the system processing module. The system processing module compares the images with standard images in the database to identify the type, specifications, and quantity of medical supplies.

[0045] Barcode / QR code recognition unit: For medical supplies with barcodes or QR codes, a scanning device is installed on the side of the emergency cart. When medical supplies are placed in the cart or taken out for use, the scanning device automatically scans the barcode or QR code, obtains relevant information about the medical supplies, such as name, specifications, production date, expiration date, etc., and transmits this information to the system processing module.

[0046] System processing module: As the core of the entire system, it uses a high-performance computer processor to receive information transmitted from the intelligent recognition module, analyze and process the data obtained by image recognition and barcode / QR code recognition. According to the preset inventory rules, it determines whether the quantity of medical supplies is accurate, whether there are any expired items, etc., and generates inventory results. The specific method is as follows:

[0047] (1) Data integration: The type, specifications, and quantity of medical supplies identified by the image recognition unit are integrated with the name, specifications, production date, expiration date, and other information of the medical supplies obtained by the barcode / QR code recognition unit to form a complete record of medical supply information. For the same medical supply, the accurate information obtained by the barcode / QR code recognition unit is preferred, and the image recognition results are used as a supplement and verification.

[0048] (2) Quantity judgment: Compare the quantity of medical supplies currently obtained by inventory with the standard quantity preset in the data storage module to determine whether there is a quantity difference. Suppose the quantity of a certain medical supply currently obtained by inventory is Q current , the standard quantity is Q standard , the allowed error range is ∈, then the quantity judgment rule is:

[0049]

[0050] Among them, ∈ is set according to the characteristics and management requirements of medical supplies. For some commonly used and easily consumed medicines, ∈ can be set to a larger value; for expensive or critical emergency medicines, ∈ should be set to a smaller value to ensure accurate management.

[0051] (3) Validity period judgment: Based on the production date and validity period information obtained by the barcode / QR code recognition unit, combined with the current date, calculate the remaining validity period of the medical supplies D remaining and the preset validity threshold D threshold Compare and determine whether it has expired or is about to expire. Let the current date be D current The expiration date of medical supplies is D expiry ,but:

[0052] D remaining =D expiry -D current ;

[0053] The validity period determination rules are as follows:

[0054]

[0055] Among them, D threshold It is set according to the characteristics and management requirements of medical supplies, and can generally be 30 days, 60 days or 90 days.

[0056] (4) Abnormal marking and priority classification: Based on the results of quantity judgment and expiration date judgment, medical supplies with abnormalities are marked and prioritized according to the severity of the abnormality. For example, expired medicines have the highest priority, followed by emergency medicines with severe shortages, and medicines about to expire and medicines with slight shortages have lower priority.

[0057] (5) Generate an inventory report: The inventory information of all medical supplies (including type, specification, quantity, status, expiration date, etc.) is organized into an inventory report. The report content includes a list of normal supplies, a list of abnormal supplies (sorted by priority), quantity difference statistics, expiration date distribution statistics, etc. At the same time, generate visual charts (such as bar charts, pie charts, etc.) to intuitively display the inventory results, so that medical staff can quickly understand the overall status of the emergency trolley supplies.

[0058] (6) Data update and synchronization: Update the inventory results to the data storage module and synchronize the data with the hospital's medical information management system to ensure the timeliness and accuracy of the emergency cart supplies information in the hospital information system.

[0059] Data Storage Module: This module stores a database of medical supply images, basic information about medical supplies (such as name, specifications, production date, expiration date, storage location, etc.), and inventory results. The data storage module utilizes distributed storage technology to ensure data security and reliability while supporting rapid data retrieval and query.

[0060] Alarm Module: When the system processing module detects an abnormality in medical supplies, such as insufficient quantity, expiration, or misplacement, the alarm module immediately issues an audible and visual alarm. The alarm module includes an audible and visual alarm installed on the emergency cart and a wireless alarm device connected to the medical staff's mobile phone or terminal device, allowing them to promptly detect and address abnormal situations.

[0061] User Interaction Module: A touchscreen installed on the emergency cart allows medical staff to start and pause inventory operations, query historical inventory records, and perform other operations. The touchscreen also displays real-time inventory results and the status of the emergency cart's supplies, allowing medical staff to intuitively understand the cart's condition. The user interaction module also supports voice interaction, allowing medical staff to perform related operations through voice commands.

[0062] The Power Management Module provides a stable power supply for the entire system, utilizing both a rechargeable lithium battery and an external power supply. The rechargeable lithium battery provides power when the emergency cart is in motion. When the battery is low, it automatically switches to the external power supply, simultaneously charging the lithium battery. The Power Management Module also features a power monitoring function, monitoring the battery level and power status in real time and displaying this information on the User Interaction Module.

[0063] The specific implementation process of the intelligent identification-based emergency trolley supplies inventory system of the present invention is as follows:

[0064] The system hardware deployment is as follows:

[0065] The structure of the traditional emergency trolley has been optimized, with a customized multi-angle HD camera (model: Hikvision DS-2CD2T47G0-L, supporting 1080p, 120° wide-angle, autofocus and infrared fill light) installed above each storage compartment, and a fixed barcode / QR code scanner (model: Honeywell Xenon 1902, supporting automatic sensing triggering) integrated on the side.

[0066] The top of the cart is equipped with an industrial-grade computer (model: Advantech UNO-2172G, Intel Core i7 processor, 8GB RAM, 128GB SSD) as the system processing module, and the side is equipped with a 10.1-inch capacitive touch screen (model: Dell P1019H, supporting multi-touch) as the user interface.

[0067] Network and power supply configuration:

[0068] The system communicates in real time with the medical information management system (HIS / LIS) through the hospital intranet, and supports Wi-Fi 6 (802.11ax) and 4G / 5G dual network backup.

[0069] The power management module uses a 24V / 10Ah rechargeable lithium battery (model: Desay Battery DB-LFP2410) and automatically switches with an external power supply. The battery life is up to 8 hours and the charging time is ≤4 hours.

[0070] The working process of the present invention is as follows:

[0071] 1. Intelligent recognition module

[0072] Image recognition unit:

[0073] Dataset construction: We collected over 100,000 images of more than 200 commonly used medical supplies in hospitals (including syringes, infusion sets, first aid medicines, and instruments) under different lighting conditions (50-1000 lux), angles (0-360°), and placement methods (stacked and scattered).

[0074] Model training: We built an improved ResNet-50 convolutional neural network based on the PyTorch framework and introduced a brain-focused attention mechanism (CBAM) to improve small object recognition accuracy. Training parameters: batch size = 32, epochs = 100, learning rate 0.001, and the Adam optimizer.

[0075] Recognition accuracy: 98.7% classification accuracy was achieved on the test set, and single image recognition took ≤ 0.5 seconds.

[0076] Barcode / QR code recognition unit:

[0077] The scanner supports mainstream symbologies such as EAN-13 and QR Code, with a recognition range of 5-15cm and a speed of 0.3 seconds per scan. For irregularly packaged pharmaceuticals, a multi-angle scanning algorithm is used, ensuring 100% recognition rate through three scans at different angles.

[0078] 2. System processing module

[0079] Real-time inventory trigger:

[0080] Medical staff trigger the inventory process using a touchscreen or voice command (e.g., "Start inventory"). The system first scans the storage compartments, activating the corresponding cameras and scanners in the order of "upper layer, middle layer, and lower layer."

[0081] Data integration and judgment:

[0082] Quantity judgment: Taking epinephrine injection (specification 1mg / vial) as an example, the standard quantity Q standard = 10 pieces, and the allowable error ∈ = 1 piece. If the current inventory quantity Q current =8, it is judged as "shortage" and triggers a yellow warning.

[0083] Validity period judgment: A batch of nitroglycerin tablets is valid until December 31, XXXX. The current date is November 15, XXXX. The remaining validity period is D remaining = 46 days. If the preset threshold D threshold = 30 days, the status is "normal"; if D threshold =60 days, it will be marked as "about to expire" (orange warning).

[0084] Exception handling mechanism:

[0085] The system classifies exceptions into three priority levels:

[0086] Level 1 (red): Expired drugs and shortages of emergency drugs (such as epinephrine and atropine);

[0087] Level 2 (orange): Drugs about to expire, shortages of major consumables (such as syringes and infusion sets);

[0088] Level 3 (yellow): There is a shortage of common drugs and the quantity is slightly abnormal.

[0089] 3. User interaction and alarm implementation

[0090] Touchscreen interface:

[0091] The Tailwind CSS framework was used to develop a responsive interface. The main interface is divided into three modules: "Real-time Inventory," "History," and "System Settings." Abnormal items are distinguished by color (red / orange / yellow).

[0092] Voice interaction:

[0093] Integrates iFlytek speech recognition SDK, supports natural language commands (such as "query adrenaline status" and "print inventory report"), and has a recognition accuracy rate of ≥95%.

[0094] Alarm system:

[0095] When a level one abnormality is detected, the sound and light alarm (model: Omron EE-SX672-WR) on the top of the cart emits a continuous beep (85dB) and a red flash, and at the same time sends a vibration reminder and pop-up notification to the responsible nurse's mobile terminal (pushed via the hospital APP).

[0096] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. An inventory system for emergency trolley supplies based on intelligent identification, characterized in that: include: An intelligent recognition module, comprising an image recognition unit and a barcode / QR code recognition unit. The image recognition unit is used to capture and identify images of medical supplies in the storage compartment of the emergency cart, and the barcode / QR code recognition unit is used to scan barcodes or QR codes on the medical supplies. The system processing module is used to receive and process the information transmitted by the intelligent recognition module and generate inventory results according to preset rules; A data storage module for storing a medical supplies image database, basic information of medical supplies, and inventory result data; An alarm prompt module is used to send an alarm signal when the system processing module determines that there is an abnormality in the medical supplies; User interaction module, used by medical staff to perform operations and view inventory results; The power management module provides power supply for the entire system.

2. The intelligent identification-based inventory system for emergency trolley supplies according to claim 1, characterized in that: The image recognition unit uses a deep learning algorithm to establish a medical supply image database by training medical supply images.

3. The intelligent identification-based inventory system for emergency trolley supplies according to claim 2, characterized in that: During inventory, the camera captures images of the storage compartments and transmits the images to the system processing module, which compares the images with standard images in the database to identify the types, specifications, and quantities of medical supplies.

4. The intelligent identification-based inventory system for emergency trolley supplies according to claim 3, characterized in that: The barcode / QR code recognition unit scans the barcode or QR code to obtain relevant information of the medical supplies, including name, specifications, production date, and expiration date, and transmits the information to the system processing module.

5. The intelligent identification-based inventory system for emergency trolley supplies according to claim 4, characterized in that: The system processing module determines whether the quantity of medical supplies is accurate and whether any of them are expired according to preset inventory rules, and generates inventory results.

6. The intelligent identification-based inventory system for emergency trolley supplies according to claim 5, characterized in that: The system processing module exchanges data with the hospital's medical information management system and uploads the inventory results to the hospital information management system.

7. The intelligent identification-based inventory system for emergency trolley supplies according to claim 6, characterized in that: The alarm prompt module includes an audible and visual alarm fixed on the emergency cart and a wireless alarm device connected to the medical staff's mobile phone or terminal equipment.

8. The intelligent identification-based inventory system for emergency trolley supplies according to claim 7, characterized in that: The user interaction module includes a touch screen fixed on the emergency cart, which is used by medical staff to start, pause, and query historical inventory records for inventory operations, and to display inventory results and status information of emergency cart supplies in real time.

9. The intelligent identification-based inventory system for emergency trolley supplies according to claim 8, characterized in that: The user interaction module is also equipped with a voice recognition unit for medical staff to perform related operations through voice commands.

10. The intelligent identification-based inventory system for emergency trolley supplies according to claim 9, characterized in that: The power management module is powered by a rechargeable lithium battery and an external power supply, monitors the battery power and power status in real time, and displays relevant information on the user interaction module.

Images