AED sharing management method and system for OHCA first aid based on Internet of Things

Abstract

The invention discloses an AED sharing management method and system for OHCA first aid based on the Internet of Things, and particularly relates to the technical field of data management. Scheduling data and task completion rate data of AED equipment and patient treatment and survival prognosis data are extracted from an Internet of Things system, the equipment utilization rate and response time distribution of different areas are analyzed in combination with a geographic information technology, and resource allocation blind spots and deviations are marked; the correlation between the dispatching success rate and the patient survival rate is further analyzed, the influence of the dispatching precision on the patient prognosis is evaluated, and the accuracy of distributing the limited AED resources by the algorithm is calculated by integrating the resource deviation and the dispatching precision evaluation result; and if the allocation is inaccurate, the AED resource sharing strategies of different areas are optimized by predicting the severity of errors and dynamically adjusting algorithm parameters, so that the accuracy and the high efficiency of the first-aid resource allocation are realized, and the overall performance and the first-aid efficiency of the internet of things first-aid system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an OHCA emergency AED sharing management method and system based on the Internet of Things. Background Art

[0002] Cardiac arrest (CA) is a serious life-threatening emergency characterized by the sudden cessation of the heart's ejection function, leading to interruption of blood circulation, respiratory arrest, and loss of consciousness. The onset of CA is extremely rapid, and the patient may lose consciousness in about 10 seconds. If emergency treatment is not performed within 5 minutes, it is very likely to cause irreversible brain damage or even death. For every minute of delay in rescue, the patient's survival rate will drop by 7%-10%. Due to untimely emergency treatment and other reasons, the overall survival rate of patients with out-of-hospital cardiac arrest (OHCA) in my country is currently less than 1%. It is worth noting that about 80% of OHCA cases are caused by ventricular fibrillation, and early defibrillation of OHCA patients can significantly improve the survival rate.

[0003] An automated external defibrillator (AED) is an emergency device that can quickly and accurately diagnose defibrillation rhythm and implement defibrillation in a timely manner. AED is known as a "foolproof tool" because of its advantages such as simple operation and convenient use. Even non-professionals who have received simple training can use it correctly. Studies have shown that cardiopulmonary resuscitation (CPR) within 1 minute and AED defibrillation within 3-5 minutes can increase the survival rate of OHCA patients to 50%-70%. Therefore, the popularization of AED equipment in public places is of great significance for the first aid of OHCA. Based on the popularization and development of Internet of Things technology, providing OHCA patients with more intelligent and efficient first aid services has become an important technical direction for solving the problem of first aid in private places. By utilizing the interconnected characteristics of the Internet of Things, the AED equipment is organically combined with the first aid management system, which can provide effective help to OHCA patients within the golden first aid time.

[0004] The prior art has the following deficiencies:

[0005] Intelligent algorithms in IoT emergency rescue systems are often used to assess patient needs and dispatch AED devices, such as calculating the optimal dispatch path based on the patient's location. However, when the algorithm misjudges due to insufficient training data or logical design flaws, serious consequences may occur. In multi-patient scenarios, the algorithm may not be able to reasonably allocate limited AED resources, such as prioritizing the dispatch of equipment to areas with lower demand and ignoring truly urgent patients, thereby reducing overall emergency rescue efficiency. At the same time, in high-demand areas, incorrect scheduling may cause some AED devices to be concentratedly occupied, while equipment in other areas is idle for a long time or cannot meet sudden demand. This imbalance in resource distribution may lead to regional rescue blind spots and serious delays in emergency rescue. Summary of the invention

[0006] The purpose of the present invention is to provide an OHCA emergency AED sharing management method and system based on the Internet of Things to solve the shortcomings of the background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things, comprising the following steps:

[0008] S1: Extract data related to the utilization rate of emergency resources from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; collect data related to the improvement rate of OHCA patient survival prognosis from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment;

[0009] S2: Based on geographic information technology, analyze the utilization rate and response time distribution of AED equipment in different regions, mark the blind spots and deviations of resource allocation, and determine the degree of deviation in the utilization rate of emergency resources;

[0010] S3: If the degree of deviation is high, analyze the correlation between dispatch success rate and patient survival rate to determine the degree of influence of dispatch accuracy on patient prognosis;

[0011] S4: Evaluate the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in the utilization of emergency resources and the impact of scheduling accuracy on patient prognosis;

[0012] S5: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

[0013] Preferably, in S2, after analyzing the AED device utilization deviation in the area with high demand but delayed response time in the demand deviation graph, an AED device utilization deviation index is generated, and the method for obtaining the AED device utilization deviation index is:

[0014] The utilization rate of AED equipment in the i-th area is marked as ; Mark the response time of the i-th region as ; Construct the decision matrix: ; S represents the value of the i-th region on the j-th indicator; the weighted matrix S is obtained by weighted standardization : ; represents the weighted and standardized utilization rate of AED equipment, Represents the weighted normalized response time; determines the ideal solution and negative ideal solution; ideal solution: ; Negative ideal solution: ; Calculate the distance to the ideal solution and the negative ideal solution. The distance to the ideal solution is: ; is the Euclidean distance between the ith region and the ideal solution; the distance from the negative ideal solution: : is the Euclidean distance between the ith region and the negative ideal solution, m represents the number of indicators, represents the weighted normalized value of the i-th sample on the j-th index, represents the ideal solution of the jth index; is the negative ideal solution of the jth index; the AED equipment utilization deviation index is calculated, and the expression is: ; HK is the AED equipment utilization deviation index.

[0015] Preferably, in S3, the dispatch survival correlation index is generated after analyzing the correlation between the dispatch success rate and the patient survival rate. The method for obtaining the dispatch survival correlation index is:

[0016] The dispatch success rate of each region is marked as , and the patient survival rate in each region is marked as , marking the emergency response time as , mark the first aid start time as ; Represent the data of each region as a multidimensional vector and construct a data matrix : ; n is the number of regions, m is the number of variables, ; After standardizing each column of data, calculate the covariance matrix, the formula is: ; In the formula, is the standardized data matrix, Covariance matrix, each element in the covariance matrix Represents the covariance of the j-th and k-th variables; perform eigenvalue decomposition on the covariance matrix Σ: ; is the eigenvalue diagonal matrix, sorted from large to small, P is the eigenvector matrix, each column corresponds to the direction of a principal component, and the variance contribution rate of each eigenvalue is calculated , the expression is: ; Variance contribution rate indicates the ability of the i-th principal component to explain the change of the total variable; Determine the first principal component Corresponding to the maximum eigenvalue , calculate the score of the first principal component: : is the first principal component score of the i-th region, T is the matrix transpose, and the dispatch survival correlation index is calculated as follows: ; In the formula, GH is the dispatch survival correlation index, is the sum of the total eigenvalues.

[0017] Preferably, in S4, the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources is evaluated according to the degree of deviation of the emergency resource utilization rate and the degree of influence of the scheduling accuracy on the patient's prognosis, specifically:

[0018] The AED equipment utilization deviation index and the dispatch survival correlation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the accuracy value label of the limited AED resources allocated by the intelligent algorithm in the Internet of Things emergency rescue system as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the limited AED resources allocated by the intelligent algorithms in all Internet of Things emergency rescue systems as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy value of the limited AED resources allocated by the intelligent algorithm in the Internet of Things emergency rescue system is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0019] Preferably, the obtained accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system is compared with a gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system is compared with the first standard threshold and the second standard threshold respectively;

[0020] If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is greater than the second standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is high, and an accuracy signal is generated at this time without adjustment;

[0021] If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is average, and an incomplete accuracy signal is generated at this time; regular monitoring is required;

[0022] If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is less than the first standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is low. At this time, an inaccuracy signal is generated and optimization is required.

[0023] Preferably, in S5, if the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas, specifically:

[0024] The error severity is defined as the difference between the actual allocation accuracy value and the ideal allocation accuracy value, expressed as: : represents the severity of the error in the ith region, is the actual allocation accuracy value of the i-th region, is the ideal allocation accuracy value for the i-th region;

[0025] The errors in all regions are combined into a global error to measure the overall deviation of the system: ; In the formula, The global error severity, n is the total number of regions; the formula for setting the prediction error is: ; The predicted error of the ith region, f is the functional form of the prediction model, is the AED equipment utilization deviation index in the i-th region, is the dispatch survival correlation index in the i-th region, is the current algorithm parameter; the error prediction model is trained through historical data so that it can be based on the comprehensive feature vector Accurate prediction error.

[0026] Preferably, the algorithm parameters are dynamically adjusted according to the prediction error , so that the global error of the prediction is minimized: ; Use the gradient descent method to adjust the algorithm parameters so that the prediction error is gradually reduced: ; The parameter value of the i-th region in the t-th iteration, is the learning rate, which controls the amplitude of parameter updates. is the gradient of the error to the parameter; the adjusted parameter Used to optimize the AED resource sharing strategy in different areas: high error area: improve resource allocation priority; low error area: maintain the current strategy and save resources; based on the optimized parameters , dynamically adjust the AED resource sharing strategy in different areas; resource allocation priority is calculated based on error and algorithm parameters: ; α, β are weight factors, which dynamically adjust the ratio of regional shared resources according to the resource allocation priority: ; In the formula, The proportion of resources allocated to the i-th region.

[0027] The present invention also provides an AED sharing management system for OHCA emergency treatment based on the Internet of Things, including a data extraction module, a deviation detection module, a correlation analysis and scheduling accuracy evaluation module, an allocation accuracy evaluation module and a dynamic optimization module;

[0028] Data extraction module: extracts data related to the utilization rate of emergency resources from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; collects data related to the improvement rate of OHCA patient survival prognosis from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment;

[0029] Deviation detection module: Based on geographic information technology, it analyzes the utilization rate and response time distribution of AED equipment in different regions, marks the blind spots and deviations of resource allocation, and determines the degree of deviation in the utilization rate of emergency resources;

[0030] Correlation analysis and dispatch accuracy assessment module: If the degree of deviation is high, the correlation between the dispatch success rate and the patient survival rate is analyzed to determine the impact of dispatch accuracy on patient prognosis;

[0031] Allocation accuracy evaluation module: evaluates the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in the utilization rate of emergency resources and the impact of scheduling accuracy on patient prognosis;

[0032] Dynamic optimization module: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in allocating limited AED resources by the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

[0033] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0034] 1. The present invention provides a systematic optimization solution for the problem of misjudgment of resource allocation caused by logical defects of intelligent algorithms and insufficient training data in the prior art by constructing an AED sharing management method for OHCA emergency treatment based on the Internet of Things. By extracting emergency resource utilization data, patient treatment data and survival prognosis data, combining geographic information technology to analyze equipment utilization and response time distribution, marking resource blind spots and deviations, and then generating an AED equipment utilization deviation index and a dispatch survival-related index, the configuration rationality and scheduling accuracy of emergency resources are comprehensively evaluated. A machine learning model is used to predict the allocation accuracy of the intelligent algorithm, and an accuracy signal is generated through threshold comparison, providing a dynamic adjustment basis for resource scheduling optimization, effectively solving the problems of concentrated resource occupation and rescue blind spots.

[0035] 2. The implementation of the present invention significantly improves the allocation efficiency of limited AED resources in the IoT emergency rescue system. By predicting the severity of errors, the algorithm parameters are dynamically adjusted to optimize the resource sharing strategy in different areas, taking into account the priority allocation in high-error areas and resource conservation in low-error areas. The effect is to achieve precise allocation of emergency resources and real-time dynamic adjustment, shorten response time, reduce emergency delays, and improve the survival rate of OHCA patients, providing a practical solution for the intelligent development of the public emergency rescue system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0037] Figure 1 The present invention is a flow chart of the method.

[0038] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Example 1, please refer to Figure 1 and Figure 2As shown, the OHCA emergency AED sharing management method based on the Internet of Things described in this embodiment includes the following steps:

[0041] S1: Extract data related to the utilization rate of emergency resources from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; collect data related to the improvement rate of OHCA patient survival prognosis from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment;

[0042] S2: Based on geographic information technology, analyze the utilization rate and response time distribution of AED equipment in different regions, mark the blind spots and deviations of resource allocation, and determine the degree of deviation in the utilization rate of emergency resources;

[0043] S3: If the degree of deviation is high, analyze the correlation between dispatch success rate and patient survival rate to determine the degree of influence of dispatch accuracy on patient prognosis;

[0044] S4: Evaluate the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in the utilization of emergency resources and the impact of scheduling accuracy on patient prognosis;

[0045] S5: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

[0046] In S1, data related to the utilization rate of emergency resources is extracted from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; data related to the improvement rate of OHCA patient survival prognosis is collected from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment, specifically:

[0047] The emergency resource utilization rate data mainly reflects the utilization efficiency, dispatch effect and equipment status of AED equipment, including the following data:

[0048] AED device dispatch data: Records of each AED device dispatch, including: Dispatch time (the time from when the alarm is triggered to when the device starts moving). Dispatch location (the coordinates of the device’s initial storage location and destination). Dispatch route (the actual dispatched driving or transportation route). Dispatch time (the time it takes for the device to get from the storage location to the patient). Dispatch failure records (situations where the device fails to arrive on time or is dispatched incorrectly). Data source: The device tracking module in the IoT platform records the movement trajectory and status of the device in real time based on positioning technologies such as GPS, LoRa, Wi-Fi, or Bluetooth. Dispatch log database of the emergency system. Analyze the dispatch efficiency of AED devices and identify possible delays or incorrect dispatch problems in dispatch.

[0049] Task completion rate data of AED equipment: the ratio of the total number of equipment that completed the task to the total number of dispatched equipment. Task status breakdown: Successfully completed the task (AED equipment was delivered in time and successfully used on patients). Partially completed the task (AED was delivered but failed to be used normally, such as an operating error). Uncompleted task (equipment was not delivered or was not used). Data source: Equipment usage status recorded by the Internet of Things system (AED shock records, equipment usage time, patch status, etc.). Equipment local logs and user feedback are transmitted back to the management platform via wireless transmission. Measure the actual utilization efficiency of AED equipment, discover unfinished tasks caused by scheduling or technical problems, and provide a basis for optimizing scheduling algorithms.

[0050] The data on the improvement rate of patient survival prognosis are mainly used to evaluate the impact of AED use on patient treatment, including the following data:

[0051] Patient treatment data: Basic information: patient age, gender, place of onset, time of first aid initiation. First aid process: time of discovery and alarm (the time interval from the patient falling to the ground to the time when the witness triggered the alarm). CPR implementation (whether cardiopulmonary resuscitation was performed, start time and duration). AED use (first defibrillation time, number of defibrillations, whether autonomous heart rhythm was restored). Result data: survival to hospital (whether alive when arriving at the hospital). Discharge survival rate (survival status of the patient when discharged). Neurological function recovery rate (measured by CPC score to measure the recovery of the patient's nervous system function). Data source: medical records of the first aid team, including records of the first aid process provided by EMS (emergency medical services). Hospital emergency and intensive care records, combined with data collected from subsequent patient treatment. Analyze the impact of the timeliness and accuracy of first aid equipment (such as AED) on patient survival, and evaluate the correlation between scheduling strategy and treatment effect.

[0052] Actual rescue data of AED equipment: Specific usage data of AED equipment: Heart rhythm analysis report (whether the device correctly detects abnormal heart rhythm). Electric shock effect record (whether the electric shock successfully terminates ventricular fibrillation). User feedback (evaluation of the device usage process by witnesses or first aid personnel). Equipment usage environment: whether it is a high-demand area (such as public places, transportation hubs). Whether there are interference factors (such as complex terrain, poor network signal). Data source: AED device's own usage log (uploaded through the built-in storage module or IoT transmission technology). Usage report from the first aid team or witnesses, submitted through the application or feedback system. Evaluate the actual use effect of the AED device, analyze the impact of equipment performance on the rescue success rate, and discover potential problems with the device under specific conditions.

[0053] To ensure that the above data can be used comprehensively to evaluate the IoT emergency rescue system, the data needs to be processed as follows: Unify data from different sources (such as IoT device data and emergency rescue records) into a comparable structured format. Associate dispatch data with patient treatment data, and ensure that the equipment dispatch records correspond to the actual treatment results through timestamp and geographic location matching. Filter outliers (such as missing data or input errors) to ensure the accuracy and reliability of the analysis.

[0054] In this application, by extracting data related to the utilization rate of emergency resources (dispatching data and task completion rate) from the IoT system and collecting data related to the improvement rate of patient survival prognosis (patient treatment data and AED actual rescue data) from the emergency team and hospital, the resource utilization efficiency and rescue effect of the emergency system can be fully reflected. These data provide a complete basis for analyzing the dispatch accuracy of intelligent algorithms in the IoT emergency system, and can provide important support for optimizing dispatch strategies and equipment management.

[0055] S2: Based on geographic information technology, analyze the AED equipment utilization and response time distribution in different regions, mark the blind spots and deviations in resource allocation, and determine the degree of deviation in emergency resource utilization.

[0056] Based on geographic information technology, the utilization rate and response time distribution of AED equipment in different regions are analyzed, including: AED equipment information: geographical location of the equipment (latitude and longitude). Equipment usage record (number of dispatches, usage). Equipment status data (battery level, maintenance frequency). Emergency incident data: OHCA occurrence location (latitude and longitude). Alarm time and equipment dispatch time. Equipment arrival time and emergency response time (total time from alarm to equipment delivery). Regional attribute data: regional population density and activity characteristics (such as public places, residential areas). Road network data (path length, traffic flow). Environmental factors (such as positioning interference that may be caused by high-rise buildings and underground areas).

[0057] Equipment utilization calculation: Formula: ; Actual usage times: the number of times the equipment is scheduled and actually used within a specific period of time; Available days: the number of days the equipment is in normal status within the period of time.

[0058] Use GIS technology to match AED equipment usage records with geographic locations to generate a spatial distribution map of equipment utilization. Mark high-utilization areas (areas where equipment is frequently used) and low-utilization areas (areas where equipment is idle for a long time). Use GIS heat maps to display equipment utilization in different areas, and use color gradients to mark high and low utilization distributions. Mark possible resource shortage points in high-utilization areas, and mark equipment redundancy points in low-utilization areas.

[0059] Response time calculation: Formula: ;Use GIS path analysis function, combined with regional road network and traffic flow, simulate the optimal path from the storage point to the patient location, and calculate the theoretical response time. Compare the actual response time with the theoretical response time, and mark the delayed areas (areas where the actual response time is significantly higher than the theoretical value).

[0060] Use GIS vector maps to display the spatial distribution of device response time, and use contour lines to mark areas with different response times. Mark potential traffic or scheduling problem points in high-latency areas.

[0061] Blind spots refer to areas where AED equipment is insufficiently covered or not deployed, usually in places where emergency incidents occur frequently but equipment utilization is extremely low. Combine the equipment distribution map and the emergency incident heat map to mark high-demand areas that are not covered by the equipment. Use the Voronoi diagram (Thiessen polygon) to analyze the service range of each AED, and mark the service gap areas based on the event distribution. Deviation refers to the phenomenon that the distribution of equipment does not match the emergency demand, which may be manifested as insufficient resources in areas with high equipment utilization or waste of resources in areas where equipment is idle. Calculate the demand-equipment ratio (number of emergency incidents / number of equipment) for each area and generate a demand deviation map. Overlay the demand deviation map with the response time distribution map to mark areas with high demand but delayed response time.

[0062] After analyzing the AED equipment utilization deviation in the area with high demand but delayed response time in the demand deviation graph, an AED equipment utilization deviation index is generated. The method for obtaining the AED equipment utilization deviation index is as follows:

[0063] The utilization rate of AED equipment in the i-th area (such as the percentage of usage frequency) is marked as ; The response time of the i-th region (the average time from device scheduling to delivery) is marked as ; Construct the decision matrix: ; S represents the value of the i-th region on the j-th indicator; the weighted matrix S is obtained by weighted standardization : ; represents the weighted and standardized utilization rate of AED equipment, Represents the weighted normalized response time; determines the ideal solution and negative ideal solution; ideal solution (optimal case): ; The maximum value of the maximization index (such as utilization) is taken, and the minimum value of the minimization index (such as response time) is taken. Negative ideal solution (worst case): ; The maximization index takes the minimum value, and the minimization index takes the maximum value.

[0064] Calculate the distance to the ideal solution and the negative ideal solution. The distance to the ideal solution is: ; is the Euclidean distance between the ith region and the ideal solution; the distance from the negative ideal solution: : is the Euclidean distance between the ith region and the negative ideal solution, m represents the number of indicators, represents the weighted normalized value of the i-th sample on the j-th index, represents the ideal solution of the jth index; is the negative ideal solution of the jth index; the AED equipment utilization deviation index is calculated, and the expression is: ; HK is the AED equipment utilization deviation index.

[0065] The closer the AED equipment utilization deviation index is to 1, the closer the allocation of emergency resources in the region is to the ideal state, the higher the utilization rate of AED equipment, the shorter the response time, and the better the resource utilization efficiency. This shows that the distribution of emergency resources in the region is relatively balanced, the equipment scheduling is reasonable, and it can meet the emergency needs in a relatively short time, thereby effectively supporting the rescue needs of OHCA patients. Regions with large deviation indexes are usually the target state of resource allocation optimization and can serve as a reference model for improvement in other regions.

[0066] The closer the AED equipment utilization deviation index is to 0, the further the emergency resource allocation in the area deviates from the ideal state, the lower the AED equipment utilization rate, the longer the response time, and the poor resource utilization efficiency. This usually reflects that there are problems such as uneven distribution of emergency resources, excessive idle equipment, or untimely scheduling in the area, which may result in patients not being able to receive effective rescue within the golden rescue time. Areas with small deviation indexes need to be paid special attention to, and blind spots and delays in resource allocation can be reduced by increasing equipment deployment or optimizing scheduling strategies.

[0067] S3: If the degree of deviation is high, analyze the correlation between the dispatch success rate and the patient survival rate to determine the extent to which dispatch accuracy affects patient prognosis.

[0068] The dispatch success rate represents the dispatch accuracy and is defined as: ; Successful dispatch includes equipment arriving on time, equipment functioning properly, and being used successfully. Patient survival rate represents the quality of the emergency outcome and is defined as: ; The survival rate data comes from the status of the emergency patient at the time of arrival or the follow-up record after discharge. The dispatch accuracy reflects the performance of the dispatch task in terms of timeliness and matching.

[0069] Extract dispatch records of AED devices from areas with high deviation, including dispatch success rate and delay data. Patient treatment records, including patient survival rate, emergency timeline (alarm time, equipment arrival time, etc.). Remove abnormal data (such as extreme failures caused by equipment failure). Standardize the data (such as normalizing to a range of 0-1) to eliminate dimensional differences between indicators. According to the deviation index, divide the areas into different groups (high deviation group, medium deviation group, low deviation group), focusing on the data of the high deviation group.

[0070] After analyzing the correlation between the dispatch success rate and the patient survival rate, a dispatch survival correlation index is generated. The method for obtaining the dispatch survival correlation index is as follows:

[0071] The dispatch success rate of each region is marked as , and the patient survival rate in each region is marked as , marking the emergency response time as , mark the first aid start time as ; Represent the data of each region as a multidimensional vector and construct a data matrix : ; n is the number of regions, m is the number of variables, ; Columns represent variables (such as dispatch success rate, survival rate, etc.), and rows represent data from different regions. After standardizing each column of data (zero mean, unit variance) to eliminate dimensional differences, the covariance matrix is ​​calculated using the formula: ; In the formula, is the standardized data matrix, The covariance matrix describes the correlation between variables. Each element in the covariance matrix Represents the covariance of the j-th and k-th variables; perform eigenvalue decomposition on the covariance matrix Σ: ; is the eigenvalue diagonal matrix, sorted from large to small, P is the eigenvector matrix, each column corresponds to the direction of a principal component, and the variance contribution rate of each eigenvalue is calculated , the expression is: ; The variance contribution rate indicates the ability of the i-th principal component to explain the change of the total variable. Determine the first principal component Corresponding to the maximum eigenvalue The eigenvector of represents the main direction of data change, and the first principal component score (the principal component value of each region) is calculated: : is the first principal component score of the i-th region, T is the matrix transpose, and the dispatch survival correlation index is calculated as follows: ; In the formula, GH is the dispatch survival correlation index, is the sum of the total eigenvalues.

[0072] When the dispatch survival correlation index value is larger and closer to 1, it means that the first principal component can explain most of the changes in the data, the correlation between the dispatch success rate and the patient survival rate is strong, and the impact of dispatch accuracy on patient prognosis is very significant. This means that improving the dispatch success rate (such as reducing misdispatches and shortening response time) can directly improve patient survival rates and is a key link in optimizing the emergency system. Areas with high DSCI values ​​need to focus on maintaining dispatch accuracy and further improve emergency efficiency to continuously improve patient prognosis.

[0073] When the dispatch survival correlation index value is smaller and closer to 0, it means that the correlation between the dispatch success rate and the patient survival rate is weak, and the dispatch accuracy has little effect on the patient's prognosis. There may be other more important factors (such as response time and emergency start time) that dominate the change of patient survival rate. In this case, it is difficult to significantly improve the prognosis by optimizing the dispatch accuracy alone. It should be combined with other key variables (such as equipment deployment density and community emergency resources) for systematic optimization to improve the overall response capability of the emergency system.

[0074] S4: Evaluate the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in emergency resource utilization and the impact of scheduling accuracy on patient prognosis.

[0075] The AED equipment utilization deviation index and the dispatch survival correlation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the accuracy value label of the limited AED resources allocated by the intelligent algorithm in the Internet of Things emergency rescue system as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the limited AED resources allocated by the intelligent algorithms in all Internet of Things emergency rescue systems as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy value of the limited AED resources allocated by the intelligent algorithm in the Internet of Things emergency rescue system is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0076] The method for obtaining the accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system is to obtain the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: ; In the formula, is the output function of the model, HK is the AED equipment utilization deviation index, GH is the dispatch survival correlation index, Accuracy values ​​for allocating limited AED resources to intelligent algorithms in IoT emergency rescue systems.

[0077] Compare the obtained accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system with the first standard threshold and the second standard threshold respectively;

[0078] If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is greater than the second standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is high. At this time, an accuracy signal is generated, indicating that the effect of the intelligent algorithm in allocating resources is good and no adjustment is required.

[0079] If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is average, and an incomplete accuracy signal is generated at this time; indicating that the system performance is acceptable but there is room for improvement and requires regular monitoring.

[0080] If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is less than the first standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is low. At this time, an inaccuracy signal is generated, indicating that the performance of the intelligent algorithm is poor and needs to be optimized first.

[0081] S5: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

[0082] The error severity is defined as the difference between the actual allocation accuracy value and the ideal allocation accuracy value, expressed as: : represents the severity of the error in the ith region, is the actual allocation accuracy value of the i-th region, is the ideal allocation accuracy value for the ith region.

[0083] The errors in all regions are combined into a global error to measure the overall deviation of the system: ; In the formula, The global error severity, n is the total number of regions. The formula for setting the prediction error is: ; The predicted error of the ith region, f is the functional form of the prediction model, is the AED equipment utilization deviation index in the i-th region, is the dispatch survival correlation index in the i-th region, is the current algorithm parameter (such as weight or scheduling priority). The error prediction model is trained by historical data so that it can be based on the comprehensive feature vector Accurate prediction error.

[0084] Dynamically adjust algorithm parameters based on prediction error , so that the global error of the prediction is minimized: ; Use the gradient descent method to adjust the algorithm parameters so that the prediction error is gradually reduced: ; The parameter value of the i-th region in the t-th iteration, is the learning rate, which controls the amplitude of parameter updates. is the gradient of the error with respect to the parameters.

[0085] Adjusted parameters It can be used to optimize the AED resource sharing strategy in different areas: High error area: increase the resource allocation priority, increase equipment deployment or optimize the scheduling path. Low error area: maintain the current strategy and save resources.

[0086] Based on the optimized parameters , dynamically adjust the AED resource sharing strategy in different areas. Resource allocation priority is calculated based on error and algorithm parameters: ; α, β are weight factors, indicating the relative importance of errors and parameters. Dynamically adjust the ratio of regional shared resources according to resource allocation priority: ; In the formula, The proportion of resources allocated to the i-th region.

[0087] In this embodiment, first, the data related to the utilization rate of emergency resources (such as the dispatch data and task completion rate of AED equipment) are extracted from the Internet of Things system, and the patient treatment data and actual rescue data of the emergency team and hospital are combined to fully grasp the resource utilization and patient prognosis information. Secondly, based on geographic information technology, the AED equipment utilization and response time distribution in each region are analyzed, the blind spots and deviations of resource allocation are marked, and the degree of deviation of emergency resource utilization is quantified. Subsequently, in areas with high deviations, the correlation between the dispatch success rate and the patient survival rate is deeply analyzed to evaluate the impact of dispatch accuracy on patient prognosis. Then, the evaluation results of the degree of deviation and dispatch accuracy are comprehensively used to calculate the accuracy value of the intelligent algorithm for allocating limited AED resources. Finally, if the allocation accuracy value is low, the algorithm parameters are dynamically adjusted by predicting the severity of the error, and the AED resource sharing strategy in different regions is optimized to continuously improve the system's emergency efficiency and resource utilization effect.

[0088] Embodiment 2, the AED sharing management system for OHCA emergency treatment based on the Internet of Things described in this embodiment includes a data extraction module, a deviation detection module, a correlation analysis and scheduling accuracy assessment module, an allocation accuracy assessment module and a dynamic optimization module;

[0089] Data extraction module: extracts data related to the utilization rate of emergency resources from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; collects data related to the improvement rate of OHCA patient survival prognosis from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment;

[0090] Deviation detection module: Based on geographic information technology, it analyzes the utilization rate and response time distribution of AED equipment in different regions, marks the blind spots and deviations of resource allocation, and determines the degree of deviation in the utilization rate of emergency resources;

[0091] Correlation analysis and dispatch accuracy assessment module: If the degree of deviation is high, the correlation between the dispatch success rate and the patient survival rate is analyzed to determine the impact of dispatch accuracy on patient prognosis;

[0092] Allocation accuracy evaluation module: evaluates the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in the utilization rate of emergency resources and the impact of scheduling accuracy on patient prognosis;

[0093] Dynamic optimization module: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in allocating limited AED resources by the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

[0094] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things, characterized in that: The following steps are involved: S1: Extract data related to the utilization rate of emergency resources from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; collect data related to the improvement rate of OHCA patient survival prognosis from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment; S2: Based on geographic information technology, analyze the utilization rate and response time distribution of AED equipment in different regions, mark the blind spots and deviations of resource allocation, and determine the degree of deviation in the utilization rate of emergency resources; S3: If the degree of deviation is high, analyze the correlation between dispatch success rate and patient survival rate to determine the degree of influence of dispatch accuracy on patient prognosis; S4: Evaluate the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in the utilization of emergency resources and the impact of scheduling accuracy on patient prognosis; S5: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

2. According to the Internet of Things-based OHCA emergency AED sharing management method according to claim 1, it is characterized by: In S2, after analyzing the AED equipment utilization deviation in the area with high demand but delayed response time in the demand deviation graph, an AED equipment utilization deviation index is generated. The method for obtaining the AED equipment utilization deviation index is: The utilization rate of AED equipment in the i-th area is marked as ; The response time of the i-th region is marked as ; Construct the decision matrix: ; S represents the value of the i-th region on the j-th indicator; the weighted matrix S is obtained by weighted standardization : ; represents the weighted and standardized utilization rate of AED equipment, Represents the weighted normalized response time; determines the ideal solution and negative ideal solution; ideal solution: ; Negative ideal solution: ; Calculate the distance to the ideal solution and the negative ideal solution. The distance to the ideal solution is: ; is the Euclidean distance between the ith region and the ideal solution; Distance to negative ideal solution: : is the Euclidean distance between the ith region and the negative ideal solution, m represents the number of indicators, represents the weighted normalized value of the i-th sample on the j-th index, represents the ideal solution of the jth index; is the negative ideal solution of the jth index; Calculate the AED equipment utilization deviation index, the expression is: ; HK is the AED equipment utilization deviation index.

3. According to claim 2, a method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things, characterized in that: In S3, the correlation between the dispatch success rate and the patient survival rate is analyzed to generate a dispatch survival correlation index, and the method for obtaining the dispatch survival correlation index is: The dispatch success rate of each region is marked as , and the patient survival rate in each region is marked as , marking the emergency response time as , mark the first aid start time as ; Represent the data of each region as a multidimensional vector and construct a data matrix : ; n is the number of regions, m is the number of variables, ; After standardizing each column of data, calculate the covariance matrix, the formula is: ; In the formula, is the standardized data matrix, Covariance matrix, each element in the covariance matrix Represents the covariance of the j-th and k-th variables; perform eigenvalue decomposition on the covariance matrix Σ: ; is the eigenvalue diagonal matrix, sorted from large to small, P is the eigenvector matrix, each column corresponds to the direction of a principal component, and the variance contribution rate of each eigenvalue is calculated , the expression is: ; Variance contribution rate indicates the ability of the i-th principal component to explain the change of the total variable; Determine the first principal component Corresponding to the maximum eigenvalue , calculate the score of the first principal component: : is the first principal component score of the i-th region, T is the matrix transpose, and the dispatch survival correlation index is calculated as follows: ; In the formula, GH is the dispatch survival correlation index, is the sum of the total eigenvalues.

4. The method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things according to claim 3 is characterized in that: In S4, the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources is evaluated based on the degree of deviation in the utilization rate of emergency resources and the impact of scheduling accuracy on patient prognosis. Specifically: The AED equipment utilization deviation index and the dispatch survival correlation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the accuracy value label of the limited AED resources allocated by the intelligent algorithm in the Internet of Things emergency rescue system as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the limited AED resources allocated by the intelligent algorithms in all Internet of Things emergency rescue systems as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The accuracy value of the limited AED resources allocated by the intelligent algorithm in the Internet of Things emergency rescue system is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. The method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things according to claim 4 is characterized in that: Compare the obtained accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the accuracy value of the intelligent algorithm for allocating limited AED resources in the IoT emergency rescue system with the first standard threshold and the second standard threshold respectively; If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is greater than the second standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is high, and an accuracy signal is generated at this time without adjustment; If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is average, and an incomplete accuracy signal is generated at this time; regular monitoring is required; If the accuracy value of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is less than the first standard threshold, it means that the accuracy of the intelligent algorithm in the IoT emergency rescue system for allocating limited AED resources is low. At this time, an inaccuracy signal is generated and optimization is required.

6. The method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things according to claim 1, characterized in that: In S5, if the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas, specifically: The error severity is defined as the difference between the actual allocation accuracy value and the ideal allocation accuracy value, expressed as: : represents the severity of the error in the ith region, is the actual allocation accuracy value of the i-th region, is the ideal allocation accuracy value for the i-th region; The errors in all regions are combined into a global error to measure the overall deviation of the system: ; In the formula, The global error severity, n is the total number of regions; the formula for setting the prediction error is: ; The predicted error of the ith region, f is the function form of the prediction model, is the AED equipment utilization deviation index in the i-th region, is the dispatch survival correlation index in the i-th region, is the current algorithm parameter; the error prediction model is trained through historical data so that it can be based on the comprehensive feature vector Accurate prediction error.

7. The method for sharing and managing AEDs for OHCA emergency treatment based on the Internet of Things according to claim 6, characterized in that: Dynamically adjust algorithm parameters based on prediction error , so that the global error of the prediction is minimized: ; Use the gradient descent method to adjust the algorithm parameters so that the prediction error is gradually reduced: ; The parameter value of the i-th region in the t-th iteration, is the learning rate, which controls the amplitude of parameter updates. is the gradient of the error to the parameter; the adjusted parameter Used to optimize the AED resource sharing strategy in different areas: high error area: improve resource allocation priority; low error area: maintain the current strategy and save resources; based on the optimized parameters , dynamically adjust the AED resource sharing strategy in different areas; resource allocation priority is calculated based on error and algorithm parameters: ; α, β are weight factors, which dynamically adjust the ratio of regional shared resources according to the resource allocation priority: ; In the formula, The proportion of resources allocated to the i-th region.

8. An OHCA emergency AED sharing management system based on the Internet of Things, used to implement an OHCA emergency AED sharing management method based on the Internet of Things as claimed in any one of claims 1 to 7, characterized in that: It includes data extraction module, deviation detection module, correlation analysis and scheduling accuracy assessment module, allocation accuracy assessment module and dynamic optimization module; Data extraction module: extracts data related to the utilization rate of emergency resources from the IoT system, including the dispatch data of AED equipment and the task completion rate data of AED equipment; collects data related to the improvement rate of OHCA patient survival prognosis from the emergency team and hospital, including patient treatment data and actual rescue data of AED equipment; Deviation detection module: Based on geographic information technology, it analyzes the utilization rate and response time distribution of AED equipment in different regions, marks the blind spots and deviations of resource allocation, and determines the degree of deviation in the utilization rate of emergency resources; Correlation analysis and dispatch accuracy assessment module: If the degree of deviation is high, the correlation between the dispatch success rate and the patient survival rate is analyzed to determine the impact of dispatch accuracy on patient prognosis; Allocation accuracy evaluation module: evaluates the accuracy of the intelligent algorithm in the IoT emergency rescue system in allocating limited AED resources based on the degree of deviation in the utilization rate of emergency resources and the impact of scheduling accuracy on patient prognosis; Dynamic optimization module: If the intelligent algorithm in the IoT emergency rescue system allocates limited AED resources inaccurately, the severity of the error in allocating limited AED resources by the intelligent algorithm in the IoT emergency rescue system is predicted, and the algorithm parameters are dynamically adjusted according to the prediction results to adjust the sharing strategy of AED resources in different areas.

Images