Intelligent injury and sign real-time monitoring and transferring stretcher system

Abstract

The present invention discloses an intelligent triage and real-time vital sign monitoring transfer stretcher system, belonging to the technical field of intelligent medical devices. The present invention combines the advantages of hardware adaptability, intelligent triage and monitoring, and cloud scheduling integration, greatly improving the overall efficiency and safety of casualty transfer and first aid. The stretcher hardware adopts lightweight composite materials and is equipped with a multi-functional design, adapting to different casualties and scenarios. Multiple sensors achieve high-precision continuous collection of vital sign and environmental data, dynamically adjust the body temperature of the casualty, and ensure transfer safety at the basic level. Through the trained injury triage model, four-level accurate classification of injuries is achieved, and the monitoring threshold can be dynamically adjusted according to the injury condition. Early warnings are given for abnormal vital signs in a timely manner. Relying on cloud-based multi-mode wireless transmission, data interconnection in the whole scenario is realized, docking with the hospital information system, completing multi-terminal linkage scheduling for the entire first aid process, optimizing transfer specifications through data correlation analysis, rationally allocating first aid resources, and achieving seamless connection between pre-hospital and in-hospital first aid.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical equipment technology, and in particular to an intelligent triage and real-time monitoring system for transporting medical personnel using a transport stretcher. Background Technology

[0002] In pre-hospital emergency care, disaster relief, and in-hospital patient transport scenarios, traditional transport stretchers only have basic load-bearing functions and lack vital sign monitoring and intelligent triage capabilities. Injury classification relies entirely on manual judgment by medical staff, which is easily influenced by experience and environmental factors, leading to inaccurate classification and delays in treatment. Chinese patent CN223667934U discloses a stretcher-based patient transport life monitoring and support system. A fixing device is detachably mounted on the stretcher equipment, and each monitoring device and life support device is detachably fixed to the fixing device. The stretcher transports the injured person; each monitoring device monitors the injured person's vital signs, acquires vital sign data, and displays the data. Electronic devices receive the vital sign data transmitted from each monitoring device and identify the data, outputting alerts when abnormal vital sign data is detected. This solves the problem of the lack of effective life monitoring and support during stretcher patient transport. It eliminates the need for medical staff to carry equipment for life monitoring and support, thus saving medical personnel resources. Furthermore, medical staff are not required to view the vital signs data monitored by the monitoring equipment in real time, which further saves medical staff's human resources.

[0003] However, while the aforementioned patents have achieved the basic function of monitoring vital signs during the transfer of wounded soldiers, the following problems still exist: 1. Existing technologies can only perform simple identification and abnormal alerts for vital signs data. They do not completely free up manpower and are prone to classification deviations due to experience and interference from the on-site environment. They are difficult to adapt to the differentiated treatment needs of patients with different injuries in various scenarios. Especially in scenarios such as disaster relief, where there are many injured people and their injuries are complex, it is easy to delay the treatment of critically injured patients. 2. Existing technology does not have a dynamic adjustment mechanism for normal medical thresholds based on injury severity. It uses a uniform and fixed threshold standard to monitor all injured persons, which cannot adapt to the personalized monitoring needs of injured persons with different injuries and body types. This can easily lead to situations where the monitoring threshold for critically injured persons is too high, resulting in missed reports, or the threshold for minor injured persons is too low, resulting in false reports. It cannot guarantee the accuracy and relevance of monitoring. 3. Without a distributed cloud-based dispatch platform, seamless data integration between pre-hospital transport and in-hospital treatment is impossible. The emergency command center struggles to monitor transport progress in real time and allocate resources effectively. Medical staff are also unable to obtain patient information in advance and prepare for treatment, impacting the efficiency of the entire emergency process. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent triage and real-time vital sign monitoring transport stretcher system, which is adapted to different wounded patients and scenarios. Multiple sensors are used to achieve high-precision continuous collection of vital sign and environmental data. Through a trained triage model, four-level accurate classification of injuries is realized, and the monitoring threshold can be dynamically adjusted according to the injury condition. Abnormal vital signs are promptly warned. Relying on multi-mode wireless transmission in the cloud, data interconnection in the whole scenario is achieved, and the hospital information system is docked to complete multi-terminal linkage scheduling in the whole first-aid process, so as to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: An intelligent triage and real-time vital sign monitoring transport stretcher system, comprising: A transport perception module, which is used to deploy vital sign collection sensors and environmental collection sensors at each collection node of the stretcher main body, and based on the vital sign collection sensors and environmental collection sensors, collect the vital sign parameters of the wounded patient and the transport environmental parameters in real time, and preprocess the collected real-time parameters to generate standardized collection data; An intelligent triage module, which is used to extract the target feature parameters required for triage and monitoring from the standardized collection data, input the target feature parameters into the trained triage model for injury analysis, obtain the injury classification result of the wounded patient and perform classification annotation; A real-time monitoring module, which is used to preset the normal medical thresholds of each vital sign parameter, dynamically adjust the normal medical thresholds in combination with the injury classification result, compare the monitored vital sign parameters with the adjusted normal medical thresholds in real time, and when the vital sign parameters exceed the adjusted normal medical thresholds, immediately trigger a local audible and visual alarm, and display the abnormal parameters, warning level and corresponding treatment suggestions; A cloud scheduling module, which is used to build a cloud data processing and scheduling platform by adopting a distributed cloud server architecture, and perform data interaction, synchronization and scheduling management among the stretcher main body, the emergency command center terminal, the medical staff's handheld terminal, the hospital emergency department terminal and the hospital emergency department information system.

[0006] Further, the vital sign collection sensors specifically include: A heart rate / oxygen saturation sensor, which is deployed inside the restraint band, fitting the wrist / chest of the wounded patient, and is used to obtain the heart rate and oxygen saturation data of the wounded patient; A body temperature sensor, which is deployed in the center of the bed surface, fitting the back of the wounded patient, and is used to monitor the body temperature change data of the wounded patient in real time; A blood pressure sensor, which is deployed at the wrist restraint band, and is used to obtain the systolic blood pressure and diastolic blood pressure data of the wounded patient; A respiratory rate sensor, which is deployed at the chest restraint band, and is used for non-invasive collection based on perceiving the chest起伏变化 of the wounded patient to obtain the respiratory rate data of the wounded patient and judge whether there is an abnormal situation with the wounded patient; A consciousness status detection sensor is deployed at the main handle of the stretcher. It is used to obtain the patient's response based on the manual operation of medical staff through sound stimulation and light stimulation, combine the patient's response with the manually input consciousness status information, and associate the input consciousness status information with vital signs data.

[0007] Furthermore, the environmental acquisition sensor includes: Temperature and humidity sensors are deployed on the side of the stretcher to monitor the temperature and humidity of the transport environment in real time, and to help determine whether the patient's abnormal body temperature is caused by environmental factors. Vibration sensors are deployed at the bottom of the stretcher body to monitor the amplitude of bumps during transport in real time. When the amplitude of bumps exceeds the preset safety threshold, an early warning signal is triggered immediately to prompt medical staff to slow down. Positioning sensors are used to collect real-time data on the stretcher's transport trajectory and current location coordinates, and synchronize this data to the cloud-based dispatch module. An air quality sensor, deployed on the side of the stretcher, is used to monitor the air quality of the transport environment in real time. When harmful gases are detected and their concentration exceeds the standard, an audible and visual alarm is triggered to remind medical staff to take protective measures.

[0008] Furthermore, the damage detection model specifically includes: The model sample acquisition unit is used to acquire historical emergency case data of patients with different injuries and body types in multiple scenarios, and to build training datasets, validation datasets and test datasets. The hierarchical construction unit is used to construct a hierarchical reasoning structure based on the clinical priority of the target feature parameters, including the vital signs critical judgment layer, the severity judgment layer, and the graded output layer; The model training unit is used to input the training dataset into the constructed hierarchical inference structure for model training, optimize the model parameters using the validation dataset, and perform performance testing using the test dataset to obtain the trained fault detection model. The injury analysis unit receives target feature parameters from standardized collected data, inputs each target feature parameter into the trained injury assessment model, and performs weighted calculations on the degree of abnormality and level of consciousness of each vital sign parameter based on the hierarchical reasoning structure to obtain the injury assessment score. The injury classification unit is used to match the injury assessment score with the preset assessment score range, determine the corresponding injury classification of the injured person, output the classification result, and simultaneously output the classification classification basis.

[0009] Furthermore, the reasoning hierarchy includes: The critical assessment layer for vital signs is used to analyze critical abnormal data of heart rate, blood oxygen saturation, and respiratory rate, and to prioritize the assessment of whether the vital signs meet the criteria for black marks. The severity assessment layer is used to assign weights to the target feature parameters of non-blacklisted samples; The graded output layer integrates the reasoning results of the vital signs criticality judgment layer and the criticality judgment layer, and outputs the graded results of red label, sample table, green label and black label.

[0010] Furthermore, the intelligent flaw detection module also includes: The model optimization unit is used to acquire historical emergency data and real-time transport and treatment data, and optimize the triage model based on the historical emergency data and real-time transport and treatment data, adjusting the parameter weights of the triage model. The data correlation analysis unit is used to correlate and analyze environmental data, bump data and injury change data during the transfer process, obtain the impact of environmental factors and transfer operations on the injury of the wounded, and output an analysis report. The report generation unit is used to generate statistical reports based on historical injury classification data and transfer data, and synchronize the statistical reports to the emergency command center terminal and hospital management terminal.

[0011] Furthermore, the real-time monitoring module dynamically adjusts the normal medical threshold based on the injury severity classification results, specifically including: The standard normal medical threshold ranges for heart rate, blood oxygen saturation, body temperature, blood pressure, and respiratory rate in healthy adults are used as the monitoring benchmark thresholds. At the same time, the baseline correction coefficients corresponding to the four injury levels of red, yellow, green and black are pre-stored, as well as preset amplitude ratio thresholds, preset data number thresholds and preset threshold lower limits, which serve as the basis for determining the dynamic adjustment of medical thresholds. Based on the injury severity classification results, the monitoring baseline threshold is initially differentiated and corrected using the corresponding baseline correction coefficient to obtain the initial correction threshold corresponding to each injury severity classification. The deviation values ​​of each vital sign parameter and the corresponding initial correction threshold are extracted during real-time monitoring, and the proportion of the deviation value to the initial correction threshold is calculated as the vital sign deviation amplitude proportion. Compare the calculated deviation percentages of each vital sign with the preset percentage thresholds. Extract target vital sign parameters whose deviation ratio exceeds a preset amplitude ratio threshold, and generate a set of parameter deviation amplitude ratios by combining them with the corresponding injury severity classification. Based on each level of injury severity, a threshold for the number of data points adapted to the urgency of the injury is preset. The number of vital sign parameters whose deviation ratio exceeds the standard in the parameter deviation ratio set corresponding to each level of injury severity is compared with the preset threshold for the number of data points for that level of injury. When the number of parameters exceeding the preset data number threshold indicates that the current initial correction threshold is not well matched with the actual physical condition of the injured person, the lower limit of the initial correction threshold range for the corresponding injury level is dynamically adjusted using the proportion of the deviation of physical signs contained in the set.

[0012] Furthermore, after dynamically adjusting the lower limit of the initial correction threshold range for the corresponding injury severity level, it also includes: When the lower limit of the initial correction threshold range corresponding to a certain injury level is adjusted, the parameter deviation amplitude ratio set corresponding to that injury level is immediately cleared, and the vital signs parameters and deviation ratios of the corresponding injured persons collected by the vital signs acquisition sensors are re-acquired and recorded. Real-time monitoring of the lower limit of the initial correction interval of the threshold after each injury severity level correction, and real-time comparison of the lower limit of the initial correction interval of the threshold after correction with the preset lower limit of the threshold; When the lower limit of the initial correction range of the corrected threshold is lower than the preset lower limit, an alarm is triggered for abnormal threshold monitoring on the equipment, and medical staff are simultaneously prompted to check the threshold adjustment parameters and the status of the patient's vital signs collection.

[0013] Furthermore, the cloud scheduling module includes: The wireless transmission unit is used for data transmission between the stretcher body and the cloud data processing and scheduling platform, as well as various associated terminals and systems, based on the multi-mode wireless transmission component. The cloud-based data management unit is used to receive and store data transmitted by the stretcher body based on the cloud-based data processing and scheduling platform, including standardized collection data, injury classification results, real-time monitoring records, alarm information, transfer trajectory, patient identity information, and hospital treatment feedback information. The multi-terminal linkage unit is used to synchronize the real-time vital sign monitoring data, injury classification results, alarm information, and transfer trajectory of the stretcher body to the emergency command center terminal, medical staff handheld terminals, and hospital emergency department terminals. The permission hierarchical management unit is used to assign corresponding operation permissions using a multi-role permission hierarchical mechanism and record the operation logs of each role.

[0014] Furthermore, the cloud scheduling module also includes: The system monitors the operating status of each monitoring terminal on the main body of the stretcher in real time. When a fault is detected in any monitoring terminal, the system promptly sends a fault notification to the equipment maintenance personnel, indicating the fault type and location. Based on standardized API interfaces, it connects with the hospital's existing HIS system, EMR system, and emergency department information system. Through the connection channel, it synchronizes the stretcher data with the hospital's information system data, and at the same time receives the patient's treatment information from the hospital.

[0015] Furthermore, the cloud-based scheduling module includes: Networking unit, used for: Based on the cloud data processing and scheduling platform, multiple stretcher entities within the target activity range are obtained, and each stretcher entity is used as a network node to construct a dynamic self-organizing network topology among the multiple stretcher entities within the target activity range; In a dynamic self-organizing network topology, each network node broadcasts a status message for the corresponding stretcher body at fixed time intervals. The status message includes the current injury classification result, current vital signs parameters, and geographic coordinates. Analysis unit, used for: Based on the preset aggregation node in the dynamic self-organizing network topology, the status messages broadcast by each network node are summarized, and the injury classification results in each status message are accumulated and counted according to red, yellow, green and black labels based on the summary results. Meanwhile, based on the status message, the time series data of vital signs parameters of each network node within a preset time period are extracted, and the coefficients of variation of heart rate, respiratory rate and blood oxygen saturation are determined based on the time series data of vital signs parameters. The coefficients of variation of heart rate, respiratory rate and blood oxygen saturation are weighted and summed to obtain the vital signs stability index of each network node. Based on preset standards, the coverage area of ​​the dynamic self-organizing network topology is divided into non-overlapping rectangular grid cells, and each network node is mapped to the corresponding grid cell based on geographical coordinates. Based on the cumulative count of injury severity classification results, the total number of red-marked wounded and wounded nodes corresponding to all network nodes in each grid unit is counted. The proportion of red-marked wounded is obtained based on the total number of red-marked wounded and wounded nodes. At the same time, the frequency of vital sign changes in network nodes in each grid unit is counted based on the change in vital sign stability index under different time periods. Based on the geographical coordinates of each network node, the average Euclidean distance between each network node in each grid cell and the preset aggregation node is determined, and the transit distance factor of each grid cell is obtained. The severity score of injuries in each grid cell is obtained by weighting and fusing the proportion of red-marked wounded, the frequency of abnormal vital signs, and the transport distance factor based on preset dynamic weights. The grid cells are traversed, and the severity score of the injury is linearly mapped to the corresponding color value in the preset color scale array based on the traversal results. The corresponding grid area is filled in the same geographic coordinate system to generate a regional injury aggregation heat map composed of grids of different color levels. The scheduling management unit is used for: The regional injury heat map is synchronized to the emergency command center terminal, and the emergency command center terminal performs conditional judgment on the regional injury heat map to determine whether there are stretcher bodies with serious abnormalities in vital signs or failure of the main sensor of the equipment. When present, the shortest resource scheduling path is determined based on the transfer distance factor in the regional injury aggregation heatmap, and resource scheduling response is carried out for the stretcher body that has experienced severe changes in vital signs or failure of the main sensor of the equipment based on the shortest resource scheduling path.

[0016] Furthermore, using the proportion of vital sign deviations included in this set, the lower limit of the initial correction threshold range for the corresponding injury severity level is dynamically adjusted, including: The proportion determination unit is used to obtain the lower limit of the initial correction threshold range corresponding to the current injury severity level. And obtain the set of proportions of the deviation of vital signs corresponding to the injury severity level; Computational unit, used for: Obtain the preset threshold for the amplitude ratio of vital signs, and calculate the dynamic adjustment coefficient based on the set of amplitude ratios of deviation of vital sign parameters; The adjusted lower limit of the threshold is calculated based on the dynamic adjustment coefficient; The adjustment unit is used to replace the initial lower limit of the corrected threshold interval with the adjusted lower limit, and to use the adjusted lower limit as the initial lower limit of the corrected threshold interval after the corresponding injury level update, and to set the deviation ratio of vital signs parameters. Clear the cache.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. The main body of the stretcher of the present invention is made of high-strength lightweight composite material, and is equipped with a detachable medical bed surface and adjustable restraint straps to adapt to patients of different body types; multiple types of sensors are distributed to achieve non-invasive, continuous and high-precision acquisition of vital signs, while also collecting data on the transport environment and bumps; when the body temperature is abnormal, it can be dynamically adjusted through heating pads, thus avoiding patient displacement, secondary injury and abnormal body temperature injury from a hardware perspective.

[0018] 2. This invention uses a hierarchical reasoning-based triage model to accurately classify injuries into four levels: red, yellow, green, and black. The model can be continuously optimized using big data. The monitoring threshold can be dynamically adjusted according to the injury level to meet the individual needs of different patients. Abnormal vital signs trigger local audible and visual alarms and provide handling suggestions, significantly reducing the probability of missed or false alarms and improving the scientific rigor and timeliness of triage and monitoring.

[0019] 3. This invention adopts multi-mode wireless transmission to adapt to the data transmission needs of all scenarios. The cloud platform realizes real-time data communication with the emergency command center, medical handheld terminals, and hospital information systems, and supports multi-role permission hierarchical management and data security protection. It can correlate and analyze the relationship between the environment, transfer operation and injury, optimize transfer procedures, and assist managers in allocating emergency resources, improve the efficiency of the entire emergency process, and reduce the incidence of secondary injury.

[0020] 4. This invention achieves autonomous aggregation and real-time updating of casualty information within a target area by constructing a dynamic self-organizing network and status message broadcasting mechanism among multiple stretcher units. Simultaneously, it performs gridded spatial aggregation of injury severity levels and generates a severity score based on the degree of vital sign fluctuations and spatial distance, which is then transformed into a visualized regional aggregation heat map. This allows for a direct presentation of the on-site injury distribution. When severe abnormalities in vital signs or equipment failure are detected, the shortest scheduling path is dynamically calculated based on the distance factor in the heat map, triggering resource allocation instructions. This represents a leap from single-point monitoring to cluster collaboration, solving the technical challenges of dispersed stretchers, fragmented information, and delayed scheduling in large-scale casualty events. It significantly improves the response efficiency and accuracy of mass casualty triage and dynamic emergency resource allocation.

[0021] 5. This invention introduces a dynamic adjustment coefficient to achieve closed-loop adaptive correction between the lower limit of the monitoring threshold and the real-time vital signs of the injured. Based on the deviation ratio of multiple abnormal vital signs parameters under the current injury classification and their clinical criticality weights, combined with a preset amplitude ratio threshold and a classification step size factor, the initial correction threshold lower limit is quantitatively scaled, ensuring that the monitoring standard closely matches the individual physiological fluctuations of the injured. When the number of abnormal parameters triggers the adjustment condition, the system automatically lowers the threshold and clears the deviation set, avoiding the risk of missed reports caused by threshold rigidity. At the same time, the differentiated step size for classification ensures that red-labeled injured receive a more sensitive monitoring response, while green-labeled injured maintain an appropriate alarm accuracy. This technology significantly improves the targeting and dynamic adaptation capabilities of vital sign monitoring under multiple injury conditions and scenarios, providing accurate and reliable early warning support for pre-hospital emergency transport. Attached Figure Description

[0022] Figure 1 This is a block diagram of the intelligent triage and real-time vital sign monitoring transport stretcher system of the present invention; Figure 2 This is a flowchart of the intelligent triage and real-time monitoring of vital signs of the injured persons according to the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] To address the technical issues in existing technologies, such as the lack of intelligent triage and injury classification functions, insufficient adaptability, poor cloud scheduling capabilities, and inefficient information exchange, please refer to [link to relevant documentation]. Figures 1-2 The present invention provides the following technical solutions: A smart triage and real-time vital sign monitoring transport stretcher system includes: The transport sensing module is used to deploy vital sign acquisition sensors and environmental acquisition sensors at each acquisition node of the stretcher body. Based on the vital sign acquisition sensors and environmental acquisition sensors, it collects the vital sign parameters of the wounded and the transport environment parameters in real time, and preprocesses the collected real-time parameters to generate standardized acquisition data, including vital sign parameters, transport environment parameters and information on the wounded's consciousness status manually entered by medical staff. The intelligent triage module is used to extract target feature parameters required for triage and monitoring from standardized collected data, including heart rate, blood oxygen saturation, body temperature, blood pressure, respiratory rate and patient consciousness status. The target feature parameters are input into the trained triage model for injury analysis, and the injury grading results of the patients are obtained and graded. The real-time monitoring module is used to preset the normal medical thresholds for various vital signs parameters. It dynamically adjusts the normal medical thresholds based on the injury grading results to adapt to the personalized monitoring needs of patients with different injuries and body types. It compares the monitored vital signs parameters with the adjusted normal medical thresholds in real time. When the vital signs parameters exceed the adjusted normal medical thresholds, it immediately triggers a local audible and visual alarm and displays the abnormal parameters, warning level, and corresponding handling suggestions. The cloud-based dispatch module is used to build a cloud-based data processing and dispatch platform using a distributed cloud server architecture, enabling data interaction, synchronization, and dispatch management between the stretcher and the emergency command center terminal, medical staff handheld terminals, hospital emergency department terminals, and the hospital emergency department information system. In this embodiment, the cloud scheduling module further includes: The system monitors the operating status of each monitoring terminal on the stretcher in real time, including battery level, sensor status, and signal strength. When a fault is detected in any monitoring terminal, the system promptly sends a fault alert to the equipment maintenance personnel, indicating the fault type and location, so that the equipment maintenance personnel can handle the situation quickly. Based on a standardized API interface and adopting the HL7 international medical information exchange standard, it connects with the hospital's existing HIS system, EMR system, and emergency department information system without affecting the normal operation of the hospital's systems. Through the connection channel, the stretcher data is synchronized with the data in the hospital's information system, while receiving the patient's treatment information from the hospital, thereby improving emergency response efficiency and treatment quality.

[0025] In this embodiment, the stretcher body is integrally molded from high-strength, lightweight composite materials. The main structure includes pre-installed heating pads that are linked to the temperature monitoring unit in the vital signs acquisition module. The stretcher bed surface is made of breathable, waterproof, medical-grade fabric and is detachable for easy high-temperature sterilization or alcohol disinfection. Adjustable medical restraint straps are installed on both sides of the bed surface to accommodate different body types of patients and prevent displacement during transport. When an abnormal body temperature is detected, the heating pads adjust the bed surface temperature to maintain stable body temperature. In this embodiment, by real-time collection, standardized processing, precise injury analysis, and dynamic monitoring of the patient's vital signs and transport environment parameters, medical staff are not required to manually complete multiple steps, reducing the inconvenience caused by human intervention. At the same time, it ensures that the injury analysis and monitoring process conforms to actual clinical needs and adapts to the personalized needs of patients with different injuries and body types, providing comprehensive vital sign protection for the patient transport process and reducing the probability of various risks during transport. Simultaneously, by linking with body temperature monitoring, it achieves stable maintenance of the patient's body temperature, improving the patient's comfort and safety during transport and avoiding additional impacts on the patient due to transport operations or environmental factors. The cloud-based dispatch module's architecture and interface adaptation enable efficient data interaction and synchronization across multiple terminals and systems, breaking down information barriers between emergency transport and hospital treatment. This ensures efficient collaboration across all stages, including emergency command, medical care, and hospital reception. Real-time monitoring of equipment operation status facilitates timely handling of equipment malfunctions, ensuring continuous and stable system operation. This provides equipment support for the smooth operation of emergency transport, adapts to the actual needs of clinical emergency transport, improves the overall service quality and efficiency of emergency transport, and enhances the standardization and safety of the emergency process.

[0026] In this embodiment, the vital sign acquisition sensor specifically includes: The heart rate / blood oxygen sensor, deployed inside the restraint strap and fitted to the wrist / chest of the injured person, uses reflective photoelectric sensing technology to acquire the injured person's heart rate and blood oxygen saturation data. The measurement range is heart rate 40-200 beats / minute and blood oxygen saturation 80%-100%, with a measurement accuracy of ±1 beat / minute (heart rate) and ±2% (blood oxygen). The response time is ≤1 second, and it can continuously collect data without missed or erroneous data collection. It achieves non-invasive and continuous measurement through the photoelectric reflection principle, has a high degree of integration, and can filter out noise through algorithms to improve measurement accuracy. The body temperature sensor, deployed in the center of the bed and attached to the back of the injured person, uses a medical-grade NTC thermistor. It has a measurement range of 32℃-42℃, an accuracy of ±0.1℃, and a response time of ≤3 seconds. It is used to monitor the body temperature changes of the injured person in real time to avoid harm to the injured person due to low or high temperatures. The blood pressure sensor, deployed at the wrist restraint, uses electronic blood pressure measurement technology to acquire the systolic and diastolic blood pressure data of the injured person. The measurement range is systolic blood pressure 60-200 mmHg and diastolic blood pressure 40-140 mmHg, with an accuracy of ±3 mmHg. It can be manually triggered for rapid measurement or set to timed automatic measurement with an adjustable interval of 1-10 minutes, making it suitable for triage and monitoring scenarios. The respiratory rate sensor, deployed at the chest restraint strap, uses strain gauge sensing technology to non-invasively acquire respiratory rate data based on the perception of changes in the rise and fall of the patient's chest. The measurement range is 8-40 breaths / minute, with an accuracy of ±1 breath / minute and a response time of ≤2 seconds. The non-invasive acquisition does not affect the patient's breathing and can continuously monitor changes in respiratory rhythm to determine whether the patient has respiratory arrest, rapid breathing, or other abnormal conditions. A consciousness status detection sensor is deployed at the main handle of the stretcher. Based on the manual operation of medical staff through sound and light stimulation to obtain the patient's response, and combined with the patient's response to manually input consciousness status information: awake, drowsy, stuporous, comatose, the sensor is linked with vital signs data to assist in the classification of injury.

[0027] In this embodiment, two sets of heart rate and blood oxygen sensors are deployed, respectively attached to the wrist and chest. If one set of sensors fails, the other set automatically switches to ensure continuous data collection. All sensors have self-test functions, automatically calibrating upon power-on and providing timely alerts in case of malfunction. Simultaneously, heating pads are installed within the support fabric, upper and lower straps. When the temperature sensor detects an abnormal body temperature in the injured person, the heating pads can maintain the injured person's temperature, improving comfort and safety during transport.

[0028] In this embodiment, the environmental acquisition sensor includes: Temperature and humidity sensors are deployed on the side of the stretcher to monitor the temperature and humidity of the transport environment in real time. The measurement range is -20℃ to 60℃ and humidity is 10% to 90%RH, with an accuracy of ±0.5℃ and ±5%RH. This helps to determine whether the patient's abnormal body temperature is caused by environmental factors. It is low in cost and highly adaptable. Vibration sensors are deployed at the bottom of the stretcher body to monitor the amplitude of bumps during transport in real time. The measurement range is 0.1-10g. When the amplitude of bumps exceeds the preset safety threshold, an early warning signal is triggered in time to prompt medical staff to slow down and avoid secondary injury to the patient caused by the bumps. The positioning sensor integrates GPS and Beidou dual-mode positioning units, with a positioning accuracy of ±1m. It is used to collect stretcher transfer trajectory and current location coordinate data in real time and synchronize them to the cloud dispatch module, so that the emergency command center can keep track of the transfer progress and optimize the transfer route. The single point positioning accuracy is controlled within 1.51m, and the lateral and longitudinal deviations are both less than 0.5m, ensuring positioning accuracy. An air quality sensor, deployed on the side of the stretcher, uses an MQ-135 sensor that is sensitive to harmful gases such as CO2, NH3, and benzene. The analog output signal is connected to the core processing module to monitor the air quality of the transport environment in real time. When harmful gases are detected and the concentration exceeds the standard, an audible and visual alarm is triggered to remind medical staff to take protective measures and ensure the safety of personnel during the transport process.

[0029] In this embodiment, the damage detection model specifically includes: The model sample acquisition unit is used to acquire historical emergency case data of patients with different injuries and body types in multiple scenarios. The historical emergency case data includes the patient's vital signs parameters, state of consciousness, actual injury classification and treatment records. The target feature parameters in the historical case data are labeled and classified based on the injury classification results, and the association between the labeled labels and the target feature parameters is established to construct training datasets, validation datasets and test datasets. The hierarchical construction unit is used to construct a hierarchical reasoning structure based on the clinical priority of the target feature parameters, including the vital signs critical judgment layer, the severity judgment layer, and the graded output layer; The model training unit is used to input the training dataset into the constructed hierarchical inference structure for model training. Hierarchical accuracy and hierarchical recall are used as evaluation metrics for model training. The model parameters are optimized in combination with the validation dataset, and the performance is tested using the test dataset to obtain the trained fault detection model. The injury analysis unit receives target feature parameters from standardized collected data, inputs each target feature parameter into the trained injury assessment model, and performs weighted calculations on the degree of abnormality and level of consciousness of each vital sign parameter based on the hierarchical reasoning structure to obtain the injury assessment score. The grading and determination unit is used to preset the assessment score range corresponding to the four levels of injury. It matches the injury assessment score with the preset assessment score range to determine the injury level of the injured person and outputs a four-level grading result: red, yellow, green, and black. Among them, red corresponds to critical injury, yellow corresponds to severe injury, green corresponds to mild injury, and black corresponds to death. The grading and determination criteria are output simultaneously, and the key target feature parameters, weight ratios, and score matching status that affect the grading results are marked.

[0030] In this embodiment, the training dataset is used for model training, the validation dataset is used for parameter adjustment during the training process, and the test dataset is used for final model performance verification, ensuring that the training dataset is consistent with the standardized collected data and target feature parameter terminology described above and that the data format is compatible. In this embodiment, the reasoning hierarchy includes: Level 1: Critical vital signs assessment layer, used to analyze critical abnormal data of heart rate, blood oxygen saturation, and respiratory rate, and to prioritize whether it meets the criteria for black mark vital signs. When heart rate, blood oxygen saturation, and respiratory rate all reach the clinical death threshold, and the consciousness state is comatose and unresponsive, it is initially judged as black mark. The second level is the severity assessment layer, which is used to assign weights to the target feature parameters of non-blacklisted samples. Among them, heart rate, blood oxygen saturation, and respiratory rate are assigned higher weights, body temperature and blood pressure are assigned medium weights, and consciousness status is assigned auxiliary weights. The weight assignment is based on the clinical injury grading guidelines and can be dynamically adjusted according to the validation results of the validation dataset. The third level is the graded output layer, which integrates the reasoning results of the vital signs critical judgment layer and the critical severity judgment layer. It combines the specific values ​​of blood pressure, body temperature and consciousness status to output the graded results of red label, sample form, green label and black label.

[0031] In this embodiment, by utilizing historical emergency case data from multiple scenarios, the trained model can better adapt to the triage needs of patients with different injuries and body types, reducing grading bias caused by limited data. The construction of a hierarchical inference structure enables the triage model to combine clinical priorities to accurately analyze and comprehensively judge the patient's vital signs and level of consciousness, achieving reasonable injury grading. At the same time, clear judgment criteria are output simultaneously during the grading process, making it easy for medical staff to intuitively understand the grading logic, further ensuring the reliability and stability of the model's grading results, and ensuring that the model can continuously and stably output accurate injury grading results.

[0032] In this embodiment, the intelligent flaw detection module further includes: The model optimization unit is used to acquire historical emergency data and real-time transport and treatment data. Based on the historical emergency data and real-time transport and treatment data, the model is optimized through big data analysis algorithms to adjust the parameter weights of the model and improve the accuracy of injury classification. The data correlation analysis unit is used to correlate and analyze environmental data, bump data, and injury change data during the transfer process. The environmental data comes from the environmental acquisition sensor, the bump data comes from the vibration sensor, and the injury change data comes from the injury analysis results of the injury assessment model. The correlation analysis obtains the impact of environmental factors and transfer operations on the patient's injury, outputs an analysis report, optimizes the transfer route and operating procedures, and reduces the incidence of secondary injury. The report generation unit is used to generate statistical reports such as transport efficiency, injury distribution, and equipment failure rate based on historical injury classification data and transport data. The statistical reports are then synchronized to the emergency command center terminal and hospital management terminal to assist medical staff and managers in optimizing emergency procedures and rationally allocating emergency resources.

[0033] In this embodiment, the real-time monitoring module dynamically adjusts the normal medical threshold based on the injury severity classification results, specifically including: The standard normal medical threshold ranges for heart rate, blood oxygen saturation, body temperature, blood pressure, and respiratory rate in healthy adults are used as the monitoring benchmark thresholds. Simultaneously, pre-stored baseline correction coefficients corresponding one-to-one with the four injury levels of red, yellow, green, and black, as well as preset amplitude ratio thresholds, preset data number thresholds, and preset threshold lower limits, serve as the basis for determining the dynamic adjustment of medical thresholds; among them, red corresponds to critical injuries, yellow corresponds to severe injuries, green corresponds to mild injuries, and black corresponds to death, which is consistent with the injury grading results output by the intelligent triage module, and the vital sign parameters are consistent with the vital sign parameters in the standardized collection data generated after collection and preprocessing by the vital sign collection module; Based on the injury severity classification results, the monitoring baseline threshold is initially differentiated and corrected using the corresponding baseline correction coefficient to obtain the initial correction threshold corresponding to each injury severity classification. The deviation values ​​of each vital sign parameter and the corresponding initial correction threshold are extracted during real-time monitoring, and the proportion of the deviation value to the initial correction threshold is calculated as the vital sign deviation amplitude ratio. The vital sign parameters are the target feature parameters in the standardized collection data acquired in real-time by the real-time monitoring module. The calculated deviation ratios of each vital sign are compared with preset deviation ratio thresholds. Different injury grades correspond to different preset deviation ratio thresholds. The deviation ratio threshold for red-label critical injuries is higher than that for yellow and green labels, which is suitable for scenarios where the vital signs of critically injured patients fluctuate greatly. Target vital sign parameters with deviation ratios exceeding preset deviation ratio thresholds are extracted and combined with the corresponding injury grades to generate a set of parameter deviation ratios. Based on each level of injury severity, a data number threshold is preset to match the urgency of the injury. The data number threshold is set differently based on the number of vital sign sensors deployed and the response efficiency requirements of the real-time monitoring module for real-time comparative monitoring. The number of vital sign parameters with excessive deviation ratios in the parameter deviation range set corresponding to each level of injury severity is compared with the preset data number threshold for that level of injury. When the number of parameters exceeding the preset data number threshold indicates that the current initial correction threshold is not well matched with the actual vital signs of the injured person, the lower limit of the initial correction threshold range for the corresponding injury level is dynamically adjusted using the proportion of the deviation of vital signs contained in the set. Among them, red and yellow marks focus on lowering the lower limit to avoid underreporting, green marks focus on minor adjustments to ensure monitoring accuracy, and black marks do not require threshold adjustment. When the lower limit of the initial correction threshold range corresponding to a certain injury level is adjusted, the parameter deviation amplitude ratio set corresponding to that injury level is immediately cleared, and the vital signs parameters and deviation ratios of the corresponding injured persons collected by the vital signs acquisition sensors are re-collected and recorded to ensure the accuracy of subsequent threshold adjustment data. Real-time monitoring of the lower limit of the initial correction interval of the threshold after each injury severity level correction, and real-time comparison of the lower limit of the initial correction interval of the threshold after correction with the preset lower limit of the threshold; When the lower limit of the initial correction range of the corrected threshold is lower than the lower limit of the preset threshold, it indicates that the threshold has been over-adjusted, which may lead to monitoring failure or frequent false alarms. In this case, an alarm will be issued for abnormal monitoring threshold of the equipment, and medical staff will be prompted to check the threshold adjustment parameters and the status of the patient's vital signs collection. If necessary, manual intervention will be performed to correct the threshold range. In this embodiment, the preset lower limit threshold is based on the fault judgment standard of the vital sign acquisition sensor and is set differently according to the monitoring accuracy requirements of different injury levels. It corresponds to the function of timely prompting when the sensor fails. The alarm method adopts local sound and light alarm and sensor fault prompt. At the same time, the alarm information is pushed to the cloud scheduling module and the handheld terminal of medical staff according to the urgency of the injury, reminding medical staff to check the operating status of the vital sign acquisition sensor in time, ensuring the accuracy of vital sign parameter collection, and thus ensuring the reliability of dynamic adjustment of normal medical thresholds corresponding to each injury level. In this embodiment, the cloud scheduling module includes: The wireless transmission unit is used to integrate wide-area wireless transmission components, local wireless transmission components, and short-range wireless transmission components based on multi-mode wireless transmission components to transmit data between the stretcher body and the cloud data processing and scheduling platform, as well as various associated terminals and systems. The cloud-based data management unit is used to receive and store data transmitted by the stretcher body based on the cloud-based data processing and scheduling platform. This includes standardized collection data, injury grading results, real-time monitoring records, alarm information, transfer trajectory, patient identity information, and hospital treatment feedback information. It supports data retrieval by keywords such as stretcher number, time range, patient information, and injury grading. It also supports data export, printing, backup, and recovery. At the same time, it classifies and organizes the stored data and checks for anomalies to ensure data integrity and accuracy. The multi-terminal linkage unit is used to synchronize the real-time vital sign monitoring data, injury classification results, alarm information, and transfer trajectory of the stretcher body to the emergency command center terminal, medical staff handheld terminals, and hospital emergency department terminals, so as to realize real-time synchronization of information across multiple terminals. The permission-based hierarchical management unit is used to assign corresponding operation permissions to super administrators, emergency command personnel, medical staff, equipment maintenance personnel, and hospital liaison personnel using a multi-role permission-based hierarchical mechanism. It clarifies the data viewing and operation scope of each role and records the operation logs of each role, including data retrieval, parameter adjustment, alarm cancellation, and other operations, to ensure the security of the cloud platform and protect the privacy of the injured and the security of medical data.

[0034] In this embodiment, based on the needs of outdoor no-LAN scenarios, an industrial-grade full-network compatible design is adopted to build a wide-area wireless transmission component, which supports multiple operator networks, sets the transmission rate to ≥10Mbps, and is suitable for remote areas, disaster sites and other scenarios. In this embodiment, based on the requirements of local area network coverage scenarios such as hospital and ambulance, the local wireless transmission component is set to have a transmission rate of ≥100Mbps and a latency of ≤50ms to ensure high-speed and low-latency data transmission. In this embodiment, based on the requirements of scenarios without public network signals such as underground garages and tunnels, the transmission distance of the short-range wireless transmission component is set to ≥10m to achieve short-range data transmission; In this embodiment, a high-strength encryption algorithm is used to encrypt all transmitted data throughout the process to protect against data leakage and tampering. At the same time, dynamic fragmentation optimization technology is used to improve the data transmission success rate and reduce the packet loss rate. In this embodiment, real-time voice communication between terminals is realized based on the wireless transmission unit, which facilitates remote dispatching by the emergency command center, collaborative treatment by medical staff, and hospital's advance prediction of treatment needs, and advance preparation for surgery. In the above embodiments, it is applicable to pre-hospital emergency care in urban roads, old residential areas, and remote areas; disaster relief such as earthquakes, floods, and fires; intra-hospital transfer between emergency rooms, operating rooms, and ICUs; and field rescue operations by the military and police. It can meet the transfer and monitoring needs of different groups of adults and children, and is especially suitable for the full-process monitoring and transfer of critically injured patients.

[0035] This embodiment provides an intelligent triage and real-time vital sign monitoring transport stretcher system, including a cloud-based dispatch module: Networking unit, used for: Based on the cloud data processing and scheduling platform, multiple stretcher entities within the target activity range are obtained, and each stretcher entity is used as a network node to construct a dynamic self-organizing network topology among the multiple stretcher entities within the target activity range; In a dynamic self-organizing network topology, each network node broadcasts a status message for the corresponding stretcher body at fixed time intervals. The status message includes the current injury classification result, current vital signs parameters, and geographic coordinates. Analysis unit, used for: Based on the preset aggregation node in the dynamic self-organizing network topology, the status messages broadcast by each network node are summarized, and the injury classification results in each status message are accumulated and counted according to red, yellow, green and black labels based on the summary results. Meanwhile, based on the status message, the time series data of vital signs parameters of each network node within a preset time period are extracted, and the coefficients of variation of heart rate, respiratory rate and blood oxygen saturation are determined based on the time series data of vital signs parameters. The coefficients of variation of heart rate, respiratory rate and blood oxygen saturation are weighted and summed to obtain the vital signs stability index of each network node. Based on preset standards, the coverage area of ​​the dynamic self-organizing network topology is divided into non-overlapping rectangular grid cells, and each network node is mapped to the corresponding grid cell based on geographical coordinates. Based on the cumulative count of injury severity classification results, the total number of red-marked wounded and wounded nodes corresponding to all network nodes in each grid unit is counted. The proportion of red-marked wounded is obtained based on the total number of red-marked wounded and wounded nodes. At the same time, the frequency of vital sign changes in network nodes in each grid unit is counted based on the change in vital sign stability index under different time periods. Based on the geographical coordinates of each network node, the average Euclidean distance between each network node in each grid cell and the preset aggregation node is determined, and the transit distance factor of each grid cell is obtained. The severity score of injuries in each grid cell is obtained by weighting and fusing the proportion of red-marked wounded, the frequency of abnormal vital signs, and the transport distance factor based on preset dynamic weights. The grid cells are traversed, and the severity score of the injury is linearly mapped to the corresponding color value in the preset color scale array based on the traversal results. The corresponding grid area is filled in the same geographic coordinate system to generate a regional injury aggregation heat map composed of grids of different color levels. The scheduling management unit is used for: The regional injury heat map is synchronized to the emergency command center terminal, and the emergency command center terminal performs conditional judgment on the regional injury heat map to determine whether there are stretcher bodies with serious abnormalities in vital signs or failure of the main sensor of the equipment. When present, the shortest resource scheduling path is determined based on the transfer distance factor in the regional injury aggregation heatmap, and resource scheduling response is carried out for the stretcher body that has experienced severe changes in vital signs or failure of the main sensor of the equipment based on the shortest resource scheduling path.

[0036] In this embodiment, the frequency of vital sign changes is the cumulative number of times the vital sign parameters of a node exceed the dynamic monitoring threshold per unit time.

[0037] In this embodiment, the dynamic self-organizing network topology refers to a network architecture in which multiple stretcher entities are interconnected via wireless transmission within the target activity range, autonomously constructing a network architecture that does not require centralized infrastructure.

[0038] In this embodiment, the status message refers to the data packet broadcast by each stretcher body as a network node at fixed time intervals, which includes information such as the current injury classification result, vital signs parameters and geographical coordinates.

[0039] In this embodiment, the preset aggregation node refers to the stretcher body or computing unit pre-designated in the dynamic self-organizing network topology for summarizing and processing the status messages of each node.

[0040] In this embodiment, the vital signs stability index refers to a quantitative indicator obtained by weighted summation of the coefficients of variation of heart rate, respiratory rate and blood oxygen saturation, which is used to characterize the severity of fluctuations in the vital signs of the injured.

[0041] In this embodiment, a rectangular grid cell refers to a non-overlapping regular rectangular area into which the coverage area of ​​the dynamic self-organizing network topology is divided according to a preset standard, and is used for spatial aggregation analysis.

[0042] In this embodiment, the percentage of injured persons with red-marked injuries refers to the ratio of the number of injured persons with red-marked injuries in each grid cell to the total number of injured persons in that cell.

[0043] In this embodiment, the frequency of abnormal vital signs refers to the frequency of abnormal fluctuations in vital signs per unit time, which is obtained by statistically analyzing the changes in the vital sign stability index at different time periods.

[0044] In this embodiment, the transit distance factor refers to the spatial distance quantification value calculated based on the average Euclidean distance between each network node and the preset aggregation node.

[0045] In this embodiment, the injury severity score refers to a comprehensive score obtained by weighting and fusing the proportion of red-marked wounded, the frequency of abnormal vital signs, and the transport distance factor according to preset dynamic weights.

[0046] In this embodiment, the preset color gradient array refers to a predefined color gradient sequence used to linearly map the severity score of injury to the corresponding color value.

[0047] In this embodiment, the regional injury aggregation heatmap refers to a visual injury distribution map generated by splicing together grids of different color levels after filling the corresponding grid area color values ​​under the same geographic coordinate system.

[0048] In this embodiment, the shortest resource scheduling path refers to the optimal path determined based on the transfer distance factor in the regional injury aggregation heatmap for resource allocation to stretcher bodies experiencing severe changes in vital signs or equipment failure.

[0049] In this embodiment, resource scheduling response refers to the allocation of rescue personnel, medical equipment, or transfer resources to the target stretcher body according to the shortest resource scheduling path.

[0050] The working principle and beneficial effects of the above technical solution are as follows: By constructing a dynamic self-organizing network and status message broadcasting mechanism among multiple stretcher units, the autonomous aggregation and real-time updating of casualty information within the target area are realized. At the same time, the severity of injuries is spatially aggregated in a grid, and a severity score is generated by combining the degree of fluctuation of vital signs and spatial distance. This score is then transformed into a visualized regional aggregation heat map, which allows the distribution of injuries on site to be presented intuitively. When a severe change in vital signs or equipment failure is detected, the shortest scheduling path is dynamically calculated based on the distance factor in the heat map, and a resource allocation command is triggered. This achieves a leap from single-point monitoring to cluster collaboration, solving the technical problems of dispersed stretchers, fragmented information, and delayed scheduling in large-scale casualty events. It significantly improves the response efficiency and accuracy of mass casualty triage and emergency resource dynamic allocation.

[0051] This embodiment provides an intelligent injury assessment and real-time vital sign monitoring stretcher system. Utilizing the proportion of vital sign deviations included in this set, the system dynamically adjusts the lower limit of the initial correction threshold range for the corresponding injury severity level, including: The proportion determination unit is used to obtain the lower limit of the initial correction threshold range corresponding to the current injury severity level. And obtain the set of proportions of the deviation of vital signs corresponding to the injury severity level. ,in, This represents the percentage of the deviation of the first physical characteristic parameter; This represents the percentage of the deviation of the second physical characteristic parameter; This represents the percentage of the deviation range of the nth physical characteristic parameter; n represents the total number of percentage deviations of the physical characteristic parameters. Computational unit, used for: Obtain the preset threshold for the amplitude ratio of vital signs, and calculate the dynamic adjustment coefficient based on the set of amplitude ratios of deviation of vital sign parameters; ; in, The dynamic adjustment coefficient is calculated based on the set of deviations in vital signs parameters. The index value represents the percentage of deviation in vital signs parameters; Indicates the first Individual characteristic parameter deviation ratio; Indicates the first The weight of the severity of injury in the injury assessment model corresponding to the deviation ratio of individual trait parameters; This indicates the preset threshold for the proportion of vital signs. This indicates the threshold adjustment step size corresponding to the injury severity level, and when marked in red, The value range is (0.15, 0.25), when marked with a yellow label. The value range is (0.08, 0.12), when marked green. The value range is (0.03, 0.05); The adjusted lower limit of the threshold is calculated based on the dynamic adjustment coefficient. ; ; The adjustment unit is used to replace the initial lower limit of the corrected threshold interval with the adjusted lower limit, and to use the adjusted lower limit as the initial lower limit of the corrected threshold interval after the corresponding injury level update, and to set the deviation ratio of vital signs parameters. Clear the cache.

[0052] In this embodiment, the lower limit of the initial correction threshold range refers to the lower limit of the normal range of each vital sign parameter obtained after the initial correction of the monitoring benchmark threshold for healthy people based on the injury grading results.

[0053] In this embodiment, the set of deviation amplitude proportions of vital signs refers to the set of values ​​formed by the deviation amplitude proportions of the measured values ​​of multiple vital sign parameters under the same injury grade relative to the initial correction threshold of that grade during real-time monitoring.

[0054] In this embodiment, the dynamic adjustment coefficient refers to the dimensionless coefficient calculated by multiplying the weighted average deviation of the abnormal vital signs parameters by a preset amplitude ratio threshold and then by the graded step size factor. It is used to quantify the degree of deviation between the current threshold and the actual condition of the injured person and to guide the downward adjustment of the threshold.

[0055] In this embodiment, the preset vital sign amplitude ratio threshold refers to the critical ratio value preset by the system to determine whether the deviation of a single vital sign parameter exceeds the standard. When the deviation amplitude ratio of a certain vital sign parameter exceeds θ, the parameter is included in the deviation amplitude ratio set.

[0056] In this embodiment, the threshold adjustment step size refers to a preset adjustment amplitude factor that corresponds one-to-one with the red, yellow, and green injury levels, and is used to control the rate of single downward adjustment when the threshold is dynamically adjusted for patients with different levels of urgency. In this embodiment, the severity weight refers to the weighting coefficient assigned to each vital sign parameter by the severity judgment layer in the injury assessment model. It is used to characterize the clinical contribution of different vital sign parameters such as heart rate, blood oxygen saturation, respiratory rate, body temperature, and blood pressure in the assessment of the severity of injury. The higher the weight value, the higher the priority of the parameter in determining critically injured patients.

[0057] The working principle and beneficial effects of the above technical solution are as follows: By introducing a dynamic adjustment coefficient, a closed-loop adaptive correction of the lower limit of the monitoring threshold and the real-time vital signs of the injured person is achieved; based on the deviation ratio of multiple abnormal vital signs parameters under the current injury classification and their clinical criticality weight, combined with the preset amplitude ratio threshold and classification step size factor, the initial correction threshold lower limit is quantitatively scaled, so that the monitoring standard closely matches the individual physiological fluctuations of the injured person; when the number of abnormal parameters triggers the adjustment condition, the system automatically completes the threshold reduction and clears the deviation set, avoiding the risk of missed reports caused by threshold rigidity. At the same time, the differentiated step size of the classification ensures that red-labeled injured persons receive a more sensitive monitoring response, and green-labeled injured persons maintain an appropriate alarm accuracy. This technical approach significantly improves the targeting and dynamic adaptation capability of vital sign monitoring under multiple injury conditions and scenarios, providing accurate and reliable early warning support for pre-hospital emergency transport.

[0058] 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 inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A smart triage and real-time vital sign monitoring transport stretcher system, characterized in that, include: The transport sensing module is used to deploy vital sign sensors and environmental sensors at each collection node of the stretcher body. Based on the vital sign sensors and environmental sensors, it collects the vital sign parameters of the wounded and the transport environment parameters in real time, and preprocesses the collected real-time parameters to generate standardized collection data. The intelligent triage module is used to extract the target feature parameters required for triage and monitoring from the standardized collected data, input the target feature parameters into the trained triage model for injury analysis, obtain the injury grading results of the injured persons and perform grading labeling; The real-time monitoring module is used to preset the normal medical thresholds for various vital signs parameters, dynamically adjust the normal medical thresholds based on the injury grading results, and compare the monitored vital signs parameters with the adjusted normal medical thresholds in real time. When the vital signs parameters exceed the adjusted normal medical thresholds, a local audible and visual alarm is immediately triggered, and the abnormal parameters, warning level, and corresponding handling suggestions are displayed. The cloud-based dispatch module is used to build a cloud-based data processing and dispatch platform using a distributed cloud server architecture. It enables data interaction, synchronization, and dispatch management between the stretcher and the emergency command center terminal, medical staff handheld terminals, hospital emergency department terminals, and the hospital emergency department information system.

2. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 1, characterized in that, The vital sign acquisition sensor specifically includes: Heart rate / blood oxygen sensor, deployed inside the restraint strap, fits against the wrist / chest of the injured person, and is used to obtain heart rate and blood oxygen saturation data of the injured person; A body temperature sensor is deployed in the center of the bed, attached to the back of the injured person, to monitor changes in the injured person's body temperature in real time; A blood pressure sensor, deployed at the wrist restraint, is used to acquire the patient's systolic and diastolic blood pressure data; A respiratory rate sensor, deployed at the chest restraint strap, is used to non-invasively collect respiratory rate data based on changes in the rise and fall of the wounded's chest, and to determine whether there are any abnormalities in the wounded. A consciousness status detection sensor is deployed at the main handle of the stretcher. It is used to obtain the patient's response based on the manual operation of medical staff through sound stimulation and light stimulation, combine the patient's response with the manually input consciousness status information, and associate the input consciousness status information with vital signs data.

3. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 1, characterized in that, The environmental acquisition sensor includes: Temperature and humidity sensors are deployed on the side of the stretcher to monitor the temperature and humidity of the transport environment in real time, and to help determine whether the patient's abnormal body temperature is caused by environmental factors. Vibration sensors are deployed at the bottom of the stretcher body to monitor the amplitude of bumps during transport in real time. When the amplitude of bumps exceeds the preset safety threshold, an early warning signal is triggered immediately to prompt medical staff to slow down. Positioning sensors are used to collect real-time data on the stretcher's transport trajectory and current location coordinates, and synchronize this data to the cloud-based dispatch module. An air quality sensor, deployed on the side of the stretcher, is used to monitor the air quality of the transport environment in real time. When harmful gases are detected and their concentration exceeds the standard, an audible and visual alarm is triggered to remind medical staff to take protective measures.

4. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 1, characterized in that, The damage assessment model specifically includes: The model sample acquisition unit is used to acquire historical emergency case data of patients with different injuries and body types in multiple scenarios, and to build training datasets, validation datasets and test datasets. The hierarchical construction unit is used to construct a hierarchical reasoning structure based on the clinical priority of the target feature parameters, including the vital signs critical judgment layer, the severity judgment layer, and the graded output layer; The model training unit is used to input the training dataset into the constructed hierarchical inference structure for model training, optimize the model parameters using the validation dataset, and perform performance testing using the test dataset to obtain the trained fault detection model. The injury analysis unit receives target feature parameters from standardized collected data, inputs each target feature parameter into the trained injury assessment model, and performs weighted calculations on the degree of abnormality and level of consciousness of each vital sign parameter based on the hierarchical reasoning structure to obtain the injury assessment score. The injury classification unit is used to match the injury assessment score with the preset assessment score range, determine the injury classification of the injured person, output the classification result, and simultaneously output the classification basis. The reasoning hierarchy includes: The critical assessment layer for vital signs is used to analyze critical abnormal data of heart rate, blood oxygen saturation, and respiratory rate, and to prioritize the assessment of whether the vital signs meet the criteria for black marks. The severity assessment layer is used to assign weights to the target feature parameters of non-blacklisted samples; The graded output layer integrates the reasoning results of the vital signs criticality judgment layer and the criticality judgment layer, and outputs the graded results of red label, sample table, green label and black label.

5. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 1, characterized in that, The intelligent defect detection module also includes: The model optimization unit is used to acquire historical emergency data and real-time transport and treatment data, and optimize the triage model based on the historical emergency data and real-time transport and treatment data, adjusting the parameter weights of the triage model. The data correlation analysis unit is used to correlate and analyze environmental data, bump data and injury change data during the transfer process, obtain the impact of environmental factors and transfer operations on the injury of the wounded, and output an analysis report. The report generation unit is used to generate statistical reports based on historical injury classification data and transfer data, and synchronize the statistical reports to the emergency command center terminal and hospital management terminal.

6. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 1, characterized in that, The real-time monitoring module dynamically adjusts the normal medical threshold based on the injury severity classification results, specifically including: The standard normal medical threshold ranges for heart rate, blood oxygen saturation, body temperature, blood pressure, and respiratory rate in healthy adults are used as the monitoring benchmark thresholds. At the same time, the baseline correction coefficients corresponding to the four injury levels of red, yellow, green and black are pre-stored, as well as preset amplitude ratio thresholds, preset data number thresholds and preset threshold lower limits, which serve as the basis for determining the dynamic adjustment of medical thresholds. Based on the injury severity classification results, the monitoring baseline thresholds are initially differentiated and corrected using the corresponding baseline correction coefficients to obtain the initial correction thresholds for each injury severity classification. Extract the deviation values ​​of each vital sign parameter from the corresponding initial correction threshold during real-time monitoring, and calculate the proportion of the deviation value to the initial correction threshold as the vital sign deviation amplitude proportion; Compare the calculated deviation percentages of each vital sign with the preset percentage thresholds. Extract target vital sign parameters whose deviation ratio exceeds a preset amplitude ratio threshold, and generate a set of parameter deviation amplitude ratios by combining them with the corresponding injury severity classification. Based on each level of injury severity, a threshold for the number of data points adapted to the urgency of the injury is preset. The number of vital sign parameters whose deviation ratio exceeds the standard in the parameter deviation ratio set corresponding to each level of injury severity is compared with the preset threshold for the number of data points for that level of injury. When the number of parameters exceeding the preset data number threshold indicates that the current initial correction threshold is not well matched with the actual physical condition of the injured person, the lower limit of the initial correction threshold range for the corresponding injury level is dynamically adjusted using the proportion of the deviation of physical signs contained in the set.

7. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 6, characterized in that, After dynamically adjusting the lower limit of the initial correction threshold range for the corresponding injury severity level, it also includes: When the lower limit of the initial correction threshold range corresponding to a certain injury level is adjusted, the parameter deviation amplitude ratio set corresponding to that injury level is immediately cleared, and the vital signs parameters and deviation ratios of the corresponding injured persons collected by the vital signs acquisition sensors are re-acquired and recorded. Real-time monitoring of the lower limit of the initial correction interval of the threshold after each injury severity level correction, and real-time comparison of the lower limit of the initial correction interval of the threshold after correction with the preset lower limit of the threshold; When the lower limit of the initial correction range of the corrected threshold is lower than the preset lower limit, an alarm is triggered for abnormal threshold monitoring on the equipment, and medical staff are simultaneously prompted to check the threshold adjustment parameters and the status of the patient's vital signs collection.

8. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 1, characterized in that, The cloud-based scheduling module includes: The wireless transmission unit is used for data transmission between the stretcher body and the cloud data processing and scheduling platform, as well as various associated terminals and systems, based on the multi-mode wireless transmission component. The cloud-based data management unit is used to receive and store data transmitted by the stretcher body based on the cloud-based data processing and scheduling platform, including standardized collection data, injury classification results, real-time monitoring records, alarm information, transfer trajectory, patient identity information, and hospital treatment feedback information. The multi-terminal linkage unit is used to synchronize the real-time vital sign monitoring data, injury classification results, alarm information, and transfer trajectory of the stretcher body to the emergency command center terminal, medical staff handheld terminals, and hospital emergency department terminals. The permission hierarchical management unit is used to allocate corresponding operation permissions using a multi-role permission hierarchical mechanism and record the operation logs of each role. The cloud-based scheduling module also includes: The system monitors the operating status of each monitoring terminal on the main body of the stretcher in real time. When a fault is detected in any monitoring terminal, the system promptly sends a fault notification to the equipment maintenance personnel, indicating the fault type and location. Based on standardized API interfaces, it connects with the hospital's existing HIS system, EMR system, and emergency department information system. Through the connection channel, it synchronizes the stretcher data with the hospital's information system data, and at the same time receives the patient's treatment information from the hospital.

9. The intelligent triage and real-time vital sign monitoring transport stretcher system according to claim 1, characterized in that, The cloud-based scheduling module includes: Networking unit, used for: Based on the cloud data processing and scheduling platform, multiple stretcher entities within the target activity range are obtained, and each stretcher entity is used as a network node to construct a dynamic self-organizing network topology among the multiple stretcher entities within the target activity range; In a dynamic self-organizing network topology, each network node broadcasts a status message for the corresponding stretcher body at fixed time intervals. The status message includes the current injury classification result, current vital signs parameters, and geographic coordinates. Analysis unit, used for: Based on the preset aggregation node in the dynamic self-organizing network topology, the status messages broadcast by each network node are summarized, and the injury classification results in each status message are accumulated and counted according to red, yellow, green and black labels based on the summary results. Meanwhile, based on the status message, the time series data of vital signs parameters of each network node within a preset time period are extracted, and the coefficients of variation of heart rate, respiratory rate and blood oxygen saturation are determined based on the time series data of vital signs parameters. The coefficients of variation of heart rate, respiratory rate and blood oxygen saturation are weighted and summed to obtain the vital signs stability index of each network node. Based on preset standards, the coverage area of ​​the dynamic self-organizing network topology is divided into non-overlapping rectangular grid cells, and each network node is mapped to the corresponding grid cell based on geographical coordinates. Based on the cumulative count of injury severity classification results, the total number of red-marked wounded and wounded nodes corresponding to all network nodes in each grid unit is counted. The proportion of red-marked wounded is obtained based on the total number of red-marked wounded and wounded nodes. At the same time, the frequency of vital sign changes in network nodes in each grid unit is counted based on the change in vital sign stability index under different time periods. Based on the geographical coordinates of each network node, the average Euclidean distance between each network node in each grid cell and the preset aggregation node is determined, and the transit distance factor of each grid cell is obtained. The severity score of injuries in each grid cell is obtained by weighting and fusing the proportion of red-marked wounded, the frequency of abnormal vital signs, and the transport distance factor based on preset dynamic weights. The grid cells are traversed, and the severity score of the injury is linearly mapped to the corresponding color value in the preset color scale array based on the traversal results. The corresponding grid area is filled in the same geographic coordinate system to generate a regional injury aggregation heat map composed of grids of different color levels. The scheduling management unit is used for: The regional injury heat map is synchronized to the emergency command center terminal, and the emergency command center terminal performs conditional judgment on the regional injury heat map to determine whether there are stretcher bodies with serious abnormalities in vital signs or failure of the main sensor of the equipment. When present, the shortest resource scheduling path is determined based on the transfer distance factor in the regional injury aggregation heatmap, and resource scheduling response is carried out for the stretcher body that has experienced severe changes in vital signs or failure of the main sensor of the equipment based on the shortest resource scheduling path.

10. The intelligent triage and real-time vital sign monitoring transport stretcher system as described in claim 6, characterized in that, Using the proportion of vital sign deviations included in this set, the lower limit of the initial correction threshold range for the corresponding injury grade is dynamically adjusted, including: The proportion determination unit is used to obtain the lower limit of the initial correction threshold range corresponding to the current injury severity level. And obtain the set of proportions of the deviation of vital signs corresponding to the injury severity level; Computational unit, used for: Obtain the preset threshold for the amplitude ratio of vital signs, and calculate the dynamic adjustment coefficient based on the set of amplitude ratios of deviation of vital sign parameters; The adjusted lower limit of the threshold is calculated based on the dynamic adjustment coefficient; The adjustment unit is used to replace the initial lower limit of the corrected threshold interval with the adjusted lower limit, and to use the adjusted lower limit as the initial lower limit of the corrected threshold interval after the corresponding injury level update, and to set the deviation ratio of vital signs parameters. Clear the cache.

Images