Emergency rescue vehicle with goods and drugs in and out of the warehouse safety management system

Abstract

The application discloses an emergency rescue vehicle article and medicine warehouse-in and warehouse-out safety management system, belongs to the technical field of medicine warehouse-in and warehouse-out, and comprises an intelligent hardware layer, a software control layer and a cloud data layer. The intelligent hardware layer is arranged in the rescue vehicle and is used for executing high-precision data collection, local real-time intelligent decision and stable physical operation under the conditions of rescue vehicle movement, use and various cabin environments. Through multi-modal sensor fusion and edge calculation, the traditional rescue vehicle is upgraded to an intelligent body with clinical perception decision execution capability. The software control layer receives data from the intelligent hardware layer, is responsible for the whole life cycle digital management of the rescue vehicle internal article and medicine and the clinical first-aid process assistance, and converts the data collected by the hardware layer into executable nursing and first-aid tasks. On the basis of realizing the article and medicine warehouse-in and warehouse-out management, the application realizes the paradigm leap from tool intelligence to system intelligence, and realizes the global optimization of resources.

Description

Technical Field

[0001] This invention relates to the field of pharmaceutical warehousing technology, and more specifically, to a safety management system for the warehousing and release of pharmaceutical supplies for emergency medical vehicles. Background Technology

[0002] Ambulances are essential emergency equipment in hospital emergency departments, intensive care units, and wards. The medicines, instruments, and consumables stored inside are vital for emergency rescue.

[0003] Currently, the management of emergency vehicle supplies and medications generally adopts a manual, static model, relying mainly on institutional constraints and the sense of responsibility of medical staff, which has many limitations: First, the management process is cumbersome and inefficient. Daily handover and inventory checks are required, using seals or combination locks for management. After opening, a comprehensive check and resealing are necessary, consuming a significant amount of nursing time. Second, the real-time monitoring of status is poor. Medication expiration dates and inventory quantities rely on periodic manual checks, making proactive early warning impossible. This increases the risk of medications nearing their expiration date not being replaced in time or shortages of supplies during emergency situations. Third, the problem of information silos is serious. Emergency vehicle usage records and supplies consumption data are independent of the hospital information system (HIS), unable to be linked with patient electronic medical records and doctor's orders to form a closed management loop. It also makes it difficult to provide data support for departmental quality control and resource allocation. Furthermore, although some technological attempts have been made, they are mostly focused on the digitization of single links, failing to achieve end-to-end intelligent management from item perception and intelligent replenishment to multi-vehicle collaborative dispatch at the system level. Moreover, it cannot be deeply integrated with vital sign monitoring and remote guidance in pre-hospital emergency scenarios. Summary of the Invention

[0004] In view of the problems existing in the prior art, the purpose of this invention is to provide a safety management system for the entry and exit of emergency medical supplies and medicines. This invention not only realizes the management of the entry and exit of supplies and medicines, but also realizes a paradigm leap from tool intelligence to system intelligence, and achieves global resource optimization.

[0005] To solve the above problems, the present invention adopts the following technical solution: An emergency medical vehicle supplies and medicines storage security management system includes: an intelligent hardware layer, a software control layer, and a cloud data layer; The intelligent hardware layer is deployed inside the emergency vehicle to perform high-precision data acquisition, local real-time intelligent decision-making, and stable physical operation under various conditions of vehicle movement, use, and internal environment. Through multimodal sensor fusion and edge computing, the traditional emergency vehicle is upgraded into an intelligent entity with clinical perception, decision-making, and execution capabilities. The software control layer receives data from the intelligent hardware layer and is responsible for the full lifecycle digital management of items and medicines inside the emergency vehicle and assisting in the clinical emergency process. It transforms the data collected by the hardware layer into executable nursing and emergency tasks. The cloud data layer stores all emergency vehicle owner data, operation logs, and historical records for the entire hospital, providing data synchronization and persistence services for the software control layer. It also supports multi-terminal access and data analysis, and by aggregating data assets, it elevates the management dimension from individual vehicles to the entire hospital network.

[0006] Compared with the prior art, the advantages of this invention are: (1) This invention continuously scans the drug expiration date label through the visual recognition module of the intelligent hardware layer, and runs the early warning algorithm in real time through the near-expiration date and equipment health warning module of the software layer. When the remaining expiration date of the drug enters the 90-day window level one warning, the system automatically marks it in the electronic inventory list. When it enters the 30-day window level two warning, it actively sends sound and light and pop-up alarms to the vehicle terminal and nurse workstation. The level three disposal automatically generates a near-expiration drug allocation suggestion form and links with the intelligent replenishment trigger unit. In the replenishment request, it prioritizes the allocation of similar near-expiration drugs from the pharmacy to high-frequency use departments, rather than receiving new batches of drugs, so as to minimize the overall scrap loss.

[0007] (2) This invention utilizes a voice command parsing module, allowing medical staff to simply verbally describe the procedure during a rescue. The system automatically drives the electric lock to open the corresponding drawer and controls the indicator light to flash, positioning the device to its designated location. The retrieval action is captured in real-time by a weight sensor and a visual recognition module, automatically recording the retrieval time, operator, drug batch number, and quantity. No verbal repetition or handwritten registration is required, eliminating the risk of omissions or errors. This seamless operation perfectly aligns with the existing habits of medical staff, eliminating the need for additional RFID tags or scanning, thus avoiding resistance caused by changing habits. Furthermore, it achieves a refined transformation of management, streamlining the previously cumbersome management processes that relied on manual labor, systems, and self-discipline into precise and efficient procedures. The automated, traceable data flow transforms time-consuming and labor-intensive shift handovers, easily overlooked expiration date management, and difficult-to-assess inventory accuracy into modules that can be automatically executed, provide real-time alerts, and be quantitatively evaluated. This not only completely liberates medical staff from heavy non-medical tasks, truly returning their time to patients, but also makes management transparent, objective, and efficient. The nursing department can conduct precise quality control and performance management based on real system data, thereby driving the management of emergency vehicles in all departments of the hospital towards standardization and homogenization. Ultimately, this solution improves the success rate of treatment on the front lines of life-or-death emergency care through technological means, and enhances operational efficiency and safety in the daily operation and maintenance backend of management, achieving a dual leap in medical quality and operational management level. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of a module of a safety management system for the entry and exit of supplies and medicines in an emergency rescue vehicle according to the present invention; Figure 2 This is a schematic diagram of the intelligent hardware layer in the emergency medical supplies and medicines entry and exit safety management system of the present invention. Detailed Implementation

[0009] 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 a part of the embodiments of the present invention, and not all of them. 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.

[0010] Example: Please see Figure 1-2 An emergency medical supplies and medicines storage security management system includes: a smart hardware layer, a software control layer and a cloud data layer; The intelligent hardware layer is deployed inside the emergency vehicle to perform high-precision data acquisition, local real-time intelligent decision-making, and stable physical operation under various conditions of the emergency vehicle's movement, use, and internal environment. Through multimodal sensor fusion and edge computing, the traditional emergency vehicle is upgraded into an intelligent agent with clinical perception, decision-making, and execution capabilities. The software control layer receives data from the smart hardware layer and is responsible for the full lifecycle digital management of items and medicines inside the ambulance and assisting in the clinical emergency process. It transforms the data collected by the hardware layer into executable nursing and emergency tasks. The cloud-based data layer stores all emergency vehicle owner data, operation logs, and historical records for the entire hospital, providing data synchronization and persistence services for the software control layer. It also supports multi-terminal access and data analysis, and by aggregating data assets, it elevates the management dimension from individual vehicles to the entire hospital network.

[0011] In a specific embodiment of the present invention, the intelligent hardware layer collects information such as the location, quantity, and expiration date of medicines and items in real time through multimodal sensors, and completes local real-time decision-making at the edge computing gateway. The software control layer runs on the vehicle-mounted intelligent terminal and is responsible for converting hardware data into standardized nursing and emergency tasks, such as automatically recording retrieval actions and generating replenishment lists. The cloud data layer aggregates data from all emergency vehicles in the hospital, providing data synchronization, persistent storage, multi-terminal access, and big data analysis services, thereby elevating the management dimension from a single vehicle to the entire hospital network.

[0012] Specifically, the intelligent hardware layer includes an object recognition and recording module, an environmental perception module, a core processing and decision-making module, a human-computer interaction module, a device control and execution module, and a communication and collaboration module; The item recognition and recording module integrates a weight sensor array and a vision recognition module to perceive the use and replenishment of medicines and items in real time, and automatically record the use time, operator, medicine / item identification, batch number and quantity. It also integrates a microphone array to collect voice commands from medical staff and extract semantic logic flow. The environmental perception module includes a rescue vehicle motion state perception unit and an operating area optical environment perception unit. The rescue vehicle motion state perception unit is used to collect the state data of the rescue vehicle in real time when it is pushed or collided. By analyzing its motion characteristics, it is quantified into a stability disturbance index, which reflects the potential threat to the stability of precision operations or the placement of items inside the vehicle. The operating area optical environment perception unit integrates an ambient light sensor and an infrared supplementary light module to monitor the light intensity and uniformity of the operating area of ​​the rescue vehicle in real time. When the ambient light is lower than the threshold required for visual recognition or there is a severe shadow, the infrared supplementary light is automatically activated to ensure that the visual recognition module can work reliably under any lighting conditions. At the same time, the lighting conditions are quantified into a light intensity influence coefficient to correct the confidence of the visual recognition algorithm or trigger a backup recognition scheme. The core processing and decision-making module is deployed on the vehicle edge computing gateway. It is used to receive pick-up and put-down event data and voice command streams from the item recognition and recording module, execute data synchronization and calibration algorithms, generate item entry and exit records and inventory update instructions. When the picking of medicine is detected, it automatically generates an operation log containing time, operator, medicine identification, batch number and quantity. When the replenishment of medicine is detected, it automatically identifies and records the replenishment information, realizing seamless data collection throughout the entire process of picking up and replenishing. The equipment control execution module is used to control the switching of electric locks and indicator lights according to the instructions of the core processing and decision-making module, and works with the vision recognition module to complete the automatic storage, retrieval and verification of medicines and supplies. The communication and collaboration module uploads anonymized operation logs, inventory data, and expiration information to the nurse station or department command center in real time via the hospital intranet or WiFi using the MQTT over TLS security protocol. It also receives allocation instructions from department experts and uploads emergency allocation requests to the cloud data layer. The core processing and decision-making module is also based on anti-interference control loop parameters and operation path instructions, and on robust parameters. With real-time feedback data from six-axis sensors, the PID control parameters of the servo motor are dynamically adjusted to form an adaptive control system. This ensures the accuracy and stability of automatic storage, retrieval, and verification of medicines and instruments inside the vehicle even when the vehicle is moved or slightly disturbed.

[0013] In a specific embodiment of the invention, the item identification and recording module senses the retrieval and replenishment events of medicines and items in real time, and automatically records the retrieval time, operator, medicine / item identification, batch number, and quantity. Simultaneously, it collects voice recordings from medical staff in the ward or emergency room environment. The motion state sensing unit in the environmental perception module integrates a high-precision six-axis IMU and vibration sensors to collect real-time state data when the emergency cart is pushed or collided with, quantifying it as a stability interference index. The optical environmental perception unit monitors the light intensity of the operating area and automatically activates infrared supplementary lighting when the light is insufficient. The core processing and decision-making module is deployed on the vehicle-mounted edge computing gateway, receiving multi-source data and performing synchronous calibration to ensure accurate execution of access operations even when the emergency cart is moving or slightly disturbed. The communication and collaboration module uploads anonymized data to the nurse station or department command center via the hospital intranet or WiFi.

[0014] Specifically, the object recognition and recording module and the environmental perception module send the raw data stream to the edge computing gateway of the core processing and decision-making module through an internal sensor bus based on the I2C or SPI protocol; After the core processing and decision-making module completes data processing locally, it pushes the generated drug and item retrieval records, replenishment confirmation records, and expiration date warning information to the display unit of the human-computer interaction module and the control unit of the equipment control execution module through a low-latency ZeroMQ-based real-time message queue, and simultaneously synchronizes them to the real-time inventory maintenance unit of the dynamic inventory intelligent replenishment module. The human-computer interaction module simultaneously receives AI voice prompts from the core processing and decision-making module and device status feedback from the device control and execution module. After multimodal fusion, it provides rescuers with a collaborative operation interface that overlays visual guidance and voice prompts through augmented reality (AR) glasses or bone conduction headphones. The visual recognition unit in the item identification and recording module and the expiration date management algorithm in the core processing and decision-making module form a specific correlation closed loop. The visual recognition unit not only identifies drug information, but also analyzes the expiration date label on the drug packaging in real time. It automatically extracts the expiration date and enters it into the system through OCR technology, realizing seamless expiration date collection. The bump index provided by the environmental perception module Instead of simple accelerometer data, it extracts specific spectral energy related to the device's resonant frequency through Fourier transform. This specific parameter is directly used as a robustness parameter in the third-level algorithm. The correction achieves tight coupling between the hardware vibration environment and the software control algorithm.

[0015] In a specific embodiment of the present invention, a tightly coupled real-time data bus and message queue system is constructed to ensure that the entire link from sensor data acquisition and edge computing inference to command issuance and multimodal human-machine interaction has extremely low latency and deterministic timing. This ensures that in emergency rescue scenarios where every second counts, the system's response is real-time, coherent, and reliable. In particular, the direct closed loop of drug identification and dispensing records and expiration date management algorithms, and the direct coupling of specific spectral turbulence index and equipment control, avoid the common problems of lengthy data flow and weak correlation in general systems. This achieves efficient collaboration of proprietary data driving proprietary functions, and ensures the stable implementation of intelligent functions from an engineering perspective.

[0016] Specifically, the software control layer includes a voice command parsing and task generation module, a dynamic inventory intelligent replenishment module, a multi-mode handover and status management module, a near-expiration date and equipment health warning module, and a clinical emergency assistance and departmental collaboration module. The voice command parsing and task generation module includes a command attribute parsing unit and a task sequence generation unit. The command attribute parsing unit receives voice signals from the microphone array of the smart hardware layer, converts them into text through a localized speech recognition engine, and parses out the operation type, target item / medicine identifier and quantity parameters of the command based on a preset emergency operation keyword library and natural language processing technology. The task sequence generation unit automatically generates a standardized operation task sequence based on the parsed operation type and target identifier. The dynamic inventory intelligent replenishment module includes a real-time inventory maintenance unit and an intelligent replenishment triggering unit. The real-time inventory maintenance unit maintains a real-time data connection with the intelligent hardware layer and receives pick-up and put-down events reported by its weight sensor array and visual recognition module. This unit maintains a full vehicle electronic inventory list. Each record in the list includes the unique identifier of the item, the current quantity, the unit location, the batch number, and the expiration date. The intelligent replenishment triggering unit is used to derive P based on the status of the emergency vehicle equipment and the inventory early warning quantification algorithm. When the P value exceeds the preset threshold, it automatically generates a replenishment list containing specific items, quantities, and suggestions for prioritizing the allocation of near-expiration medicines, and pushes it to the pharmacy or materials management department through the hospital's internal communication system. The multi-mode handover and status management module includes a handover mode decision unit and a process execution engine. The handover mode decision unit relies directly on the item outbound logs provided by the dynamic inventory intelligent replenishment module during the handover period through decision logic. If the logs are empty, the system automatically enters the intelligent no-handover mode; otherwise, it enters the forced handover mode. In the forced handover mode, the process execution engine guides the incoming staff to use a handheld scanning terminal or vehicle terminal to scan the vehicle's QR code, triggering a quick inventory process. Subsequently, it compares the theoretical list in the dynamic inventory maintenance unit with the inventory results fed back in real time by the intelligent hardware layer. Differences are highlighted and require confirmation. In the intelligent no-handover mode, the incoming staff only needs to scan the code or press a button to confirm the current unused status. The system automatically generates an electronic handover record and synchronizes it to the cloud. The near-expiration date and equipment health early warning module includes a three-level expiration date management unit and an equipment health assessment unit. The three-level expiration date management unit is used to continuously scan all drugs in the inventory, with the first-level warning occurring when the remaining expiration date of a drug approaches [a certain date]. At the 90-day window, the system is marked in the inventory list. A level-two warning is triggered when the remaining shelf life is shorter than the more urgent 30-day window, proactively sending audible, visual, and pop-up alerts to the vehicle terminal and nurse workstation. A level-three response automatically generates a near-expiry drug allocation suggestion form and attempts to link with the intelligent replenishment trigger unit, prioritizing the allocation of similar near-expiry drugs from the pharmacy in replenishment requests, rather than requesting drugs with new batch numbers, thereby reducing overall waste losses. The equipment health assessment unit periodically obtains communication power and data storage remaining capacity information from the vehicle intelligent terminal and sensor devices from the intelligent hardware layer, combining historical data and accumulated pressure to calculate the system health decay factor. And provide it to the early warning algorithm of the intelligent replenishment trigger unit, when When the value is too high, an independent equipment maintenance alarm will be issued; The clinical emergency care assistance and departmental collaboration module includes an emergency task material matching unit and a one-click ward information alert unit. The emergency task material matching unit, upon receiving emergency task information from the departmental dispatch system, performs a matching calculation based on the standard material requirement package corresponding to the task type and the real-time list in the dynamic inventory maintenance unit. The matching algorithm is as follows: The system displays the matching degree M and the list of missing items on the vehicle terminal in real time, and issues a prompt to medical staff when M is below the set standard of 95%. The one-click early warning unit for ward information is used to support medical staff to quickly enter or automatically generate a summary of the patient's condition through the voice command parsing module in the emergency room or ward. At the same time, the information is bound to the department location and sent to the emergency department or related departments with one click through the communication module, realizing early warning and feedback of information.

[0017] In a specific embodiment of the invention, the voice command parsing and task generation module uses a localized voice recognition engine and an emergency operation keyword library to parse the verbal orders of medical staff into standardized operation task sequences. The dynamic inventory intelligent replenishment module is connected in real time to the weight sensor array and visual recognition module of the intelligent hardware layer to maintain the electronic inventory list of the entire vehicle and calculate the comprehensive risk index P through an early warning quantification algorithm. When the P value exceeds the threshold, a replenishment list is automatically generated. The multi-mode handover and status management module automatically selects the intelligent non-handover mode or the forced handover mode based on the outbound log within the handover cycle. If the log is empty, only scanning the code is required to complete the handover; otherwise, the handheld terminal is guided to scan the QR code of the entire vehicle. The system automatically compares the theoretical list with the inventory results, and highlights the differences. The near-expiration date and equipment health warning module adopts a three-level expiration date management system: level one warning, level two warning, and level three treatment. When the clinical emergency assistance and departmental collaboration module receives an emergency task from the departmental dispatch system, it matches the standard material requirement package according to the task type, calculates the matching degree M, and displays the missing list in real time, supporting one-click warning of ward information to relevant departments.

[0018] Specifically, the quantitative algorithm formula for the equipment status and inventory early warning of the emergency vehicle is as follows: Where P is the intelligent early warning value, is the dimensionless comprehensive risk index, and n is the total number of items monitored inside the vehicle. This represents the preset safety stock threshold for item type i, expressed in units of one piece per item. This represents the current real-time inventory of item type i, expressed in units of one piece. The clinical urgency weighting coefficient for item i is dimensionless and determined by historical usage frequency and the importance of emergency treatment; m represents the quantity of the drug within the warning window. This is the early warning time window for the expiration date, in days. This represents the remaining shelf life of the j-th drug, in days. The failure risk weighting coefficient for the j-th drug is dimensionless. The system health decay factor is dimensionless and is provided by the near-expiration and equipment health early warning module, reflecting the status of equipment power and storage capacity.

[0019] In a specific embodiment of the present invention, a multi-factor fusion risk assessment model is established, which not only considers the inventory quantity, but also integrates the clinical urgency of the drug, the expiration date risk, and the system's own health. This algorithm upgrades inventory warning from a simple low-quantity alarm to an intelligent comprehensive risk warning. The system can automatically identify which drugs are critical and consumable, and which drugs are nearing their expiration date but can be allocated, thereby generating more scientific and economical replenishment suggestions. This solves the waste problem caused by traditional replenishment relying on manual experience and easily ignoring the correlation between expiration dates, and realizes lean inventory management.

[0020] Specifically, the real-time inventory maintenance unit of the dynamic inventory intelligent replenishment module is the core data source of the system. It provides log basis for the decision-making of the multi-mode handover and status management module, provides inventory snapshot for the matching calculation of the clinical emergency assistance and departmental collaboration module, and provides expiration date scanning data source for the near-expiration and equipment health warning module. The expiration date warning information and equipment health factors output by the near-expiration date and equipment health warning module As a key parameter, it is input into the early warning algorithm of the dynamic inventory intelligent replenishment module, affecting the replenishment logic. At the same time, its expiration date disposal suggestions can be directly called by the replenishment trigger unit. If the clinical emergency assistance and departmental collaboration module triggers drug retrieval during task execution, it will generate an event log and synchronize it to the dynamic inventory intelligent replenishment module, thereby updating the inventory and potentially triggering subsequent replenishment processes. At the same time, the generated operation records, early warning events, and handover logs are all synchronized to the cloud data layer in real time through a unified data bus interface, forming a complete and traceable management closed loop.

[0021] In a specific embodiment of the present invention, the voice command parsing module enables medical staff to achieve second-level location through verbal commands during rescue, eliminating the need for manual searching. The early warning algorithm of the dynamic inventory intelligent replenishment module comprehensively considers the urgency of items, inventory quantity, remaining expiration days, and equipment health, avoiding false alarms caused by a single factor. The intelligent no-handover mode of the multi-mode handover module is an industry first. When the emergency cart is not used within the handover period, the system automatically skips the mandatory inventory process, requiring only the incoming staff to scan the code for confirmation, which will reduce handover time and significantly reduce the burden on nurses.

[0022] Specifically, the global inventory and deployment optimization module includes a hospital-wide emergency vehicle resource map unit and a non-uniform deployment optimization unit based on ward spatial big data. The hospital-wide emergency vehicle resource map unit is used to dynamically store and visualize the real-time status of all emergency vehicles. The real-time status includes the electronic inventory list of each vehicle, equipment health, and current task status. The task status includes idle, in use, under handover, and under maintenance. The non-uniform deployment optimization unit based on ward spatial big data is used to apply the principle of emergency vehicle usage heat domain analysis. At the same time, it continuously receives the usage trigger locations of each emergency vehicle and generates an emergency vehicle usage heat map of the hospital's internal and surrounding areas through the hospital's GIS geographic information system. Based on this non-uniform distribution feature, it supports the coordinated allocation of emergency vehicle supplies across all wards. When a certain medicine enters the near-expiration warning window on a certain emergency vehicle, the system automatically matches it with the inventory of similar medicines in other emergency vehicles throughout the hospital and prioritizes the allocation of low-expiration medicines to the wards with the highest usage frequency to avoid waste due to expiration. The model training and OTA management module receives historical early warning data from the dynamic inventory intelligent replenishment module of all emergency vehicle software control layers. This data is used to continuously train and optimize key weight parameters in the early warning algorithm. By analyzing historical data, it automatically replenishes frequently used emergency medications. The weight of clinical urgency is increased while the weight of low-frequency special drugs is decreased, making the early warning more accurate. After the model is optimized and iterated, the updated algorithm parameters or logic are packaged into an upgrade package and pushed to the software control layer of the relevant emergency vehicle through the 5G wireless communication network. The vehicle automatically receives and updates the package at night, realizing the synchronization and optimization of the hospital's emergency vehicle management strategy and forming a self-evolutionary closed loop of data collection, cloud training, algorithm update and vehicle execution. The quality control and traceability audit module includes a full-process digital traceability unit and a multi-dimensional quality control analysis unit. The full-process digital traceability unit receives and permanently stores all key operation logs uploaded by the emergency cart software control layer, including the operator, time, drug identification, operator, expiration date, batch number, handover information, and near-expiration disposal records for each drug / item retrieval. These data are linked in a timeline to form an immutable electronic audit chain. The multi-dimensional quality control analysis unit runs a dynamic calculation algorithm for the department's emergency cart management efficiency KPIs. It regularly extracts data from the digital traceability unit each month to automatically generate visual reports and rankings, providing data-driven decision-making basis for the nursing department to conduct departmental performance evaluations and allocate hospital-wide resources.

[0023] In a specific embodiment of the invention, the global inventory and deployment optimization module constructs a hospital-wide emergency vehicle resource map, displaying the electronic inventory list, equipment health, and task status of each vehicle in real time. The non-uniform deployment optimization unit, based on ward spatial big data, generates an emergency vehicle usage heatmap through the hospital's GIS geographic information system, analyzes the usage frequency distribution in each ward, and supports the coordinated allocation of emergency vehicle supplies across all wards. When a certain medication enters its near-expiration warning window on a particular emergency vehicle, the system automatically matches it with the inventory of similar medications in other emergency vehicles throughout the hospital, prioritizing the allocation of near-expiration medications to the wards with the highest usage frequency to avoid delays due to expiration. To prevent waste and transform dormant medications into mobile resources, the model training and OTA management module receives historical warning data from all vehicle terminals, continuously trains and optimizes the weight parameters in the warning algorithm, automatically increasing the weight of high-frequency medications and automatically decreasing the weight of low-frequency medications, and pushes the updated algorithm parameters to each vehicle terminal via wireless network OTA. It automatically updates during non-use periods. The quality control and traceability audit module permanently stores all operation logs and automatically calculates multi-dimensional quality control KPIs. When the emergency department receives a rescue call, the clinical emergency collaboration module automatically selects the nearest emergency vehicle with sufficient inventory to the department where the incident occurred and pre-pushes the patient information to the doctor's workstation in the target department.

[0024] Specifically, the clinical emergency collaboration module includes an emergency task information synchronization and preprocessing unit and an interdepartmental information seamless connection unit. The emergency task information synchronization and preprocessing unit is used to push task information and preliminary patient conditions to the vehicle's software control layer in advance when the hospital's emergency department receives a rescue call or dispatch information, so that the vehicle is ready before the information arrives. The interdepartmental information seamless connection unit is used to receive structured patient data sent by the ward information one-click early warning unit from the vehicle's software control layer, pre-connect the data with the hospital information system (HIS), automatically generate a pre-admission record in the target department, and display the patient's vital signs trend chart and used medication list in advance on the receiving department's doctor's workstation. At the same time, it supports the receiving department's doctor to send further guidance and suggestions to the vehicle through this channel.

[0025] Specifically, the multi-dimensional quality control analysis unit calculates multiple quality control indicators as follows: Emergency medication qualification rate = (Number of qualified emergency medications / Total number of emergency medications) × 100%; First aid equipment pass rate = (Number of qualified first aid equipment / Total number of first aid equipment) × 100%; Proactive disposal rate of near-expiry drugs = (Quantity of near-expiry drugs allocated and used through system recommendations / Total number of near-expiry drugs generated) × 100%; The global inventory and deployment optimization module is the cloud-based data aggregation center, providing training data sources for the model training and OTA management module, vehicle resource query services for the clinical emergency collaboration module, and real-time status snapshots for the quality control and traceability audit module. The model training and OTA management module obtains data from other modules for optimization, and its output directly affects the decision-making logic of all vehicle-side software control layers through the OTA distribution unit, forming a core optimization closed loop. The data foundation of the quality control and traceability audit module relies on log uploads from the vehicle-side software control layer and various cloud modules. The resulting analysis results can be used as parameters to input into the deployment algorithm of the global inventory and deployment optimization module.

[0026] In a specific embodiment of the present invention, by defining quantifiable KPIs, vague management requirements are transformed into precise data indicators, and an optimization closed loop is formed within the cloud through data flow between modules. This provides hospital management with an objective, accurate, and measurable management tool. The nursing department can conduct performance evaluations based on reports and rankings automatically generated by the system, and the analysis results can be fed back to deploy optimization algorithms to dynamically adjust resource allocation. This forms a continuous improvement closed loop of data collection, analysis, decision-making, and optimization.

[0027] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and its improved concept, should be covered within the scope of protection of the present invention.

Claims

1. A safety management system for the entry and exit of supplies and medicines for an emergency medical vehicle, characterized in that, It includes a smart hardware layer, a software control layer, and a cloud data layer; The intelligent hardware layer is deployed inside the emergency vehicle to perform high-precision data acquisition, local real-time intelligent decision-making, and stable physical operation under various conditions of vehicle movement, use, and internal environment. Through multimodal sensor fusion and edge computing, the traditional emergency vehicle is upgraded into an intelligent entity with clinical perception, decision-making, and execution capabilities. The software control layer receives data from the intelligent hardware layer and is responsible for the full lifecycle digital management of items and medicines inside the emergency vehicle and assisting in the clinical emergency process. It transforms the data collected by the hardware layer into executable nursing and emergency tasks. The cloud data layer stores all emergency vehicle owner data, operation logs, and historical records for the entire hospital, providing data synchronization and persistence services for the software control layer. It also supports multi-terminal access and data analysis, and by aggregating data assets, it elevates the management dimension from individual vehicles to the entire hospital network.

2. The emergency medical supplies and medicines entry and exit safety management system according to claim 1, characterized in that, The intelligent hardware layer includes an object recognition and recording module, an environmental perception module, a core processing and decision-making module, a human-computer interaction module, a device control and execution module, and a communication and collaboration module. The item identification and recording module integrates a weight sensor array and a visual recognition module to perceive the use and replenishment events of medicines and items in real time, and automatically record the use time, operator, medicine and item identification, batch number and quantity. It also integrates a microphone array to collect voice commands from medical staff and extract semantic logic flow. The environmental perception module includes a rescue vehicle motion state perception unit and an operating area optical environment perception unit. The rescue vehicle motion state perception unit is used to collect real-time state data of the rescue vehicle when it is pushed or collided with, and analyze its motion characteristics. The interference is quantified into a stability disturbance index, which reflects the potential threat to the stability of precision operations or the placement of items inside the vehicle. The optical environment perception unit of the operation area integrates an ambient light sensor and an infrared supplementary light module to monitor the light intensity and uniformity of the operation area of ​​the rescue vehicle in real time. When the ambient light is lower than the threshold required for visual recognition or there is a severe shadow, the infrared supplementary light is automatically activated to ensure that the visual recognition module can work reliably under any lighting conditions. At the same time, the lighting conditions are quantified into a light intensity influence coefficient to correct the confidence of the visual recognition algorithm or trigger a backup recognition scheme. The core processing and decision-making module is deployed on the vehicle-mounted edge computing gateway. It is used to receive pick-up and put-down event data and voice command streams from the item recognition and recording module, execute data synchronization and calibration algorithms, generate item entry and exit records and inventory update instructions. When the picking of medicine is detected, an operation log containing time, operator, medicine identification, batch number and quantity is automatically generated. When the replenishment of medicine is detected, the replenishment information is automatically identified and recorded, realizing seamless data collection throughout the entire process of picking up and replenishing.

3. The emergency medical supplies and medicines entry and exit safety management system according to claim 2, characterized in that, The equipment control execution module is used to control the switching of the electric lock and indicator lights according to the instructions of the core processing and decision module, and to complete the automatic storage, retrieval and verification of medicines and items in conjunction with the visual recognition module; The communication and collaboration module uploads the anonymized operation logs, inventory data, and expiration information to the nurse station or department command center in real time via the hospital intranet or WiFi using the MQTT over TLS security protocol, and receives allocation instructions from department experts, and uploads emergency allocation requests to the cloud data layer. The core processing and decision-making module is also based on anti-interference control loop parameters and operation path instructions, and on robust parameters. With real-time feedback data from six-axis sensors, the PID control parameters of the servo motor are dynamically adjusted to form an adaptive control system. This ensures the accuracy and stability of automatic storage, retrieval, and verification of medicines and instruments inside the vehicle even when the vehicle is moved or slightly disturbed.

4. The emergency medical supplies and medicines entry and exit safety management system according to claim 2, characterized in that, The object identification and recording module and the environmental perception module send the raw data stream to the edge computing gateway of the core processing and decision-making module through an internal sensor bus based on the I2C or SPI protocol. After completing data processing locally, the core processing and decision-making module will push the generated drug and item retrieval records, replenishment confirmation records, and expiration date warning information to the display unit of the human-computer interaction module and the control unit of the device control execution module through a low-latency ZeroMQ-based real-time message queue, and simultaneously synchronize them to the real-time inventory maintenance unit of the dynamic inventory intelligent replenishment module. The human-computer interaction module simultaneously receives AI voice prompts from the core processing and decision-making module and device status feedback from the device control and execution module. After multimodal fusion, it provides the rescuer with a collaborative operation interface that overlays visual guidance and voice prompts through augmented reality (AR) glasses or bone conduction headphones. The visual recognition unit in the item identification and recording module and the expiration date management algorithm in the core processing and decision-making module form a specific correlation closed loop. The visual recognition unit not only identifies drug information, but also analyzes the expiration date label on the drug packaging in real time. It automatically extracts the expiration date and enters it into the system through OCR technology, realizing seamless expiration date collection. The environmental perception module provides a bump index. Instead of simple accelerometer data, it extracts specific spectral energy related to the device's resonant frequency through Fourier transform. This specific parameter is directly used as a robustness parameter in the third-level algorithm. The correction achieves tight coupling between the hardware vibration environment and the software control algorithm.

5. The emergency medical supplies and medicines entry and exit safety management system according to claim 4, characterized in that, The software control layer includes a voice command parsing and task generation module, a dynamic inventory intelligent replenishment module, a multi-mode handover and status management module, a near-expiration date and equipment health warning module, and a clinical emergency assistance and departmental collaboration module. The voice command parsing and task generation module includes a command attribute parsing unit and a task sequence generation unit. The command attribute parsing unit is used to receive voice signals from the microphone array of the smart hardware layer, convert them into text through a localized speech recognition engine, and parse out the operation type, target item / medicine identifier and quantity parameters of the command based on a preset emergency operation keyword library and natural language processing technology. The task sequence generation unit is used to automatically generate a standardized operation task sequence based on the parsed operation type and target identifier. The dynamic inventory intelligent replenishment module includes a real-time inventory maintenance unit and an intelligent replenishment triggering unit. The real-time inventory maintenance unit maintains a real-time data connection with the intelligent hardware layer and receives pick-up and put-down events reported by its weight sensor array and visual recognition module. This unit maintains a full vehicle electronic inventory list. Each record in the list includes the unique identifier of the item, the current quantity, the unit location, the batch number, and the expiration date. The intelligent replenishment triggering unit is used to derive P based on the status of the emergency vehicle equipment and the inventory early warning quantification algorithm. When the P value exceeds a preset threshold, it automatically generates a replenishment list containing specific items, quantities, and suggestions for prioritizing the allocation of near-expiration drugs, and pushes it to the pharmacy or materials management department through the hospital's internal communication system. The multi-mode handover and status management module includes a handover mode decision unit and a process execution engine. The handover mode decision unit relies directly on the item outbound logs provided by the dynamic inventory intelligent replenishment module during the handover period through decision logic. If the logs are empty, the system automatically enters the intelligent no-handover mode; otherwise, it enters the forced handover mode. The process execution engine is used to guide the incoming staff to scan the vehicle's QR code using a handheld scanning terminal or vehicle terminal in the forced handover mode, triggering a quick inventory process. Subsequently, the theoretical list in the dynamic inventory maintenance unit is compared with the inventory results fed back in real time by the intelligent hardware layer. Differences are highlighted and confirmation is required. In the intelligent no-handover mode, the incoming staff only needs to scan the code or press the button to confirm the current unused status. The system automatically generates an electronic handover record and synchronizes it to the cloud. The near-expiration date and equipment health early warning module includes a three-level expiration date management unit and an equipment health assessment unit. The three-level expiration date management unit is used to continuously scan all drugs in the inventory, with the first-level warning occurring when the remaining expiration date of a drug approaches [a certain date]. At the 90-day window, the item is marked in the inventory list. A level-two warning is triggered when the remaining shelf life is shorter than the more urgent 30-day window, proactively sending audible, visual, and pop-up alerts to the vehicle terminal and nurse workstation. A level-three response automatically generates a near-expiry drug allocation suggestion form and attempts to link with the intelligent replenishment trigger unit, prioritizing the allocation of similar near-expiry drugs from the pharmacy in replenishment requests, rather than obtaining new batches of drugs, thereby reducing overall waste losses. The device health assessment unit periodically obtains communication power and data storage remaining capacity information from the vehicle intelligent terminal and sensor devices from the intelligent hardware layer, and calculates the system health decay factor by combining historical data and accumulated pressure. And provide it to the early warning algorithm of the intelligent replenishment trigger unit, when When the value is too high, an independent equipment maintenance alarm will be issued; The clinical emergency care assistance and departmental collaboration module includes an emergency task material matching unit and a ward information one-click early warning unit. The emergency task material matching unit, upon receiving emergency task information from the departmental dispatch system, performs a matching calculation based on the standard material requirement package corresponding to the task type and the real-time list in the dynamic inventory maintenance unit. The matching degree algorithm is as follows: The system displays the matching degree M and the list of missing items on the vehicle terminal in real time, and issues a prompt to medical staff when M is below the set standard of 95%. The one-click early warning unit for ward information is used to support medical staff to quickly enter or automatically generate a summary of the patient's condition through the voice command parsing module in the emergency room or ward. At the same time, the information is bound to the department location and sent to the emergency department or related departments with one click through the communication module, so as to realize the early warning and feedback of information.

6. The emergency medical supplies and medicines entry and exit safety management system according to claim 5, characterized in that, The quantitative algorithm formula for the equipment status and inventory early warning of the rescue vehicle is as follows: Where P is the intelligent early warning value, is the dimensionless comprehensive risk index, and n is the total number of items monitored inside the vehicle. This represents the preset safety stock threshold for item type i, expressed in units of one piece per item. This represents the current real-time inventory of item type i, expressed in units of one piece. The clinical urgency weighting coefficient for item i is dimensionless and determined by historical usage frequency and the importance of emergency treatment; m represents the quantity of the drug within the warning window. This is the early warning time window for the expiration date, in days. This represents the remaining shelf life of the j-th drug, in days. The failure risk weighting coefficient for the j-th drug is dimensionless. The system health decay factor is dimensionless and is provided by the near-expiration and equipment health early warning module, reflecting the status of equipment power and storage capacity.

7. The emergency medical supplies and medicines entry and exit safety management system according to claim 5, characterized in that, The voice command parsing and task generation module is the starting point for task input, and the task sequence it generates drives the dynamic inventory intelligent replenishment module and the intelligent hardware layer to perform specific operations. The real-time inventory maintenance unit of the dynamic inventory intelligent replenishment module is the core data source of the system. It provides log basis for the decision-making of the multi-mode handover and status management module, provides inventory snapshot for the matching calculation of the clinical emergency assistance and departmental collaboration module, and provides expiration date scanning data source for the near-expiration and equipment health warning module. The expiration date warning information and equipment health factors output by the near-expiration date and equipment health warning module As a key parameter, it is input into the early warning algorithm of the dynamic inventory intelligent replenishment module, affecting the replenishment logic. At the same time, its expiration date disposal suggestions can be directly called by the replenishment trigger unit. If the clinical emergency assistance and departmental collaboration module triggers drug dispensing during task execution, it will generate an event log and synchronize it to the dynamic inventory intelligent replenishment module, thereby updating the inventory and potentially triggering subsequent replenishment processes. At the same time, the generated operation records, early warning events, and handover logs are all synchronized to the cloud data layer in real time through a unified data bus interface, forming a complete and traceable management closed loop.

8. The emergency medical supplies and medicines entry and exit safety management system according to claim 1, characterized in that, The cloud data layer includes a global inventory and deployment optimization module, a model training OTA over-the-air download technology management module, a quality control and traceability audit module, and a clinical emergency collaboration module. The global inventory and deployment optimization module includes a hospital-wide emergency vehicle resource map unit and a non-uniform deployment optimization unit based on ward spatial big data. The hospital-wide emergency vehicle resource map unit is used to dynamically store and visualize the real-time status of all emergency vehicles. The real-time status includes the electronic inventory list of each vehicle, equipment health, and current task status. The task status includes idle, in use, under handover, and under maintenance. The non-uniform deployment optimization unit based on ward spatial big data applies the principle of emergency vehicle usage heat domain analysis and continuously receives the usage trigger locations of each emergency vehicle. It generates a heat map of emergency vehicle usage distribution in the hospital and surrounding areas through the hospital's GIS geographic information system. Based on this non-uniform distribution feature, it supports the coordinated allocation of emergency vehicle supplies across all wards. When a certain drug enters the near-expiration warning window on a certain emergency vehicle, the system automatically matches it with the inventory of similar drugs in other emergency vehicles throughout the hospital and prioritizes the allocation of low-expiration drugs to the wards with the highest usage frequency to avoid waste due to expiration. The model training and OTA management module receives historical early warning data from the dynamic inventory intelligent replenishment module of all emergency vehicle software control layers. It uses this data to continuously train and optimize key weight parameters in the early warning algorithm, and automatically replenishes frequently used emergency medications by analyzing historical data. The weight of clinical urgency is increased while the weight of low-frequency special drugs is decreased, making the early warning more accurate. After the model is optimized and iterated, the updated algorithm parameters or logic are packaged into an upgrade package and pushed to the software control layer of the relevant emergency vehicle through the 5G wireless communication network. The vehicle automatically receives and updates the package at night, realizing the synchronization and optimization of the hospital's emergency vehicle management strategy and forming a self-evolutionary closed loop of data collection, cloud training, algorithm update and vehicle execution. The quality control and traceability audit module includes a full-process digital traceability unit and a multi-dimensional quality control analysis unit. The full-process digital traceability unit receives and permanently stores all key operation logs uploaded by the emergency cart software control layer, including the operator, time, drug identification, operator, expiration date, batch number, handover information, and near-expiration disposal records for each drug / item retrieval. These data are linked in a timeline to form an immutable electronic audit chain. The multi-dimensional quality control analysis unit runs a dynamic calculation algorithm for the department's emergency cart management efficiency KPIs. It regularly extracts data from the digital traceability unit each month to automatically generate visual reports and rankings, providing data-driven decision-making basis for the nursing department to conduct departmental performance evaluations and allocate hospital-wide resources.

9. A safety management system for the entry and exit of supplies and medicines for an emergency medical vehicle according to claim 8, characterized in that, The clinical emergency collaboration module includes an emergency task information synchronization and preprocessing unit and an inter-departmental information seamless docking unit. The emergency task information synchronization and preprocessing unit is used to push task information and preliminary patient conditions to the vehicle's software control layer in advance when the hospital's emergency department receives a rescue call or dispatch information, so that the vehicle is ready before the information arrives. The inter-departmental information seamless docking unit is used to receive structured patient data sent from the ward information one-click early warning unit from the vehicle's software control layer, pre-dock the data with the hospital information system (HIS), automatically generate a pre-admission record in the target department, and display the patient's condition summary and list of used medications in advance on the receiving department's doctor's workstation. At the same time, it supports the receiving department's doctor to send further guidance and suggestions to the vehicle through this channel.

10. A safety management system for the entry and exit of supplies and medicines for an emergency medical vehicle according to claim 8, characterized in that, The multi-dimensional quality control analysis unit calculates multiple quality control indicators as follows: Emergency medication qualification rate = (Number of qualified emergency medications / Total number of emergency medications) × 100%; First aid equipment pass rate = (Number of qualified first aid equipment / Total number of first aid equipment) × 100%; Proactive disposal rate of near-expiry drugs = (Quantity of near-expiry drugs allocated and used through system recommendations / Total number of near-expiry drugs generated) × 100%; The global inventory and deployment optimization module is a cloud-based data aggregation center that provides training data sources for the model training and OTA management module, vehicle resource query services for the clinical emergency collaboration module, and real-time status snapshots for the quality control and traceability audit module. The model training and OTA management module obtains data from other modules for optimization, and its output directly affects the decision-making logic of all vehicle-side software control layers through the OTA distribution unit, forming a core optimization closed loop. The data foundation of the quality control and traceability audit module relies on the log uploads from the vehicle-side software control layer and various cloud modules. The analysis results generated can be used as parameters to input into the deployment algorithm of the global inventory and deployment optimization module.

Images