Aviation medical rescue assessment method and system based on MR technology

Abstract

The invention discloses an aeronautical medical rescue assessment method and system based on an MR technology, and relates to the technical field of aeronautical medical rescue. Comprising a multi-dimensional data acquisition module, a fusion data display module, a task real-time evaluation module, a task automatic adjustment module, a remote cooperation adjustment module and a task record evaluation module. According to the scheme, data such as vital signs of a patient, the flight state of an aircraft, rescue worker operation and medical resource configuration are comprehensively integrated through multi-dimensional data acquisition, a comprehensive data source ensures that a decision maker can obtain comprehensive, real-time and accurate situation awareness, and information one-sidedness caused by a single data source in the prior art is avoided; the data are cleaned and fused through a Kalman filter, and the precision and timeliness of the data are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aviation medical rescue, and in particular to an aviation medical rescue assessment method and system based on MR technology. Background Art

[0002] Aeromedical rescue assessment technology refers to the use of modern information technology to monitor, analyze and optimize the health status of patients, the standardization of rescue personnel's operations, the flight status of aircraft and the use of medical resources in real time during the aeromedical rescue process, so as to improve the rescue efficiency and patient survival rate. This technology is widely used in air ambulance, battlefield rescue, disaster medical rescue and other scenarios, and can help decision makers make accurate judgments quickly in emergency situations.

[0003] Traditional aviation medical rescue mainly relies on a single data source and manual decision-making. Medical rescue personnel usually collect patient data through portable vital signs monitoring equipment, report the situation to the ground medical team through walkie-talkies or satellite communications, and ground experts provide guidance based on limited text or voice information. In addition, the flight status of the aircraft is manually adjusted by the crew, and the management of medical resources relies on manual recording and deployment. The overall rescue process lacks intelligence, data fusion and visualization display methods, resulting in limited rescue efficiency.

[0004] The existing technology has the following major defects: isolated data, lack of integration, lack of visualization means, delayed evaluation and adjustment, insufficient remote collaboration capabilities, and lack of intelligent optimization of the rescue process. Summary of the invention

[0005] In view of the above problems, the present invention proposes an aviation medical rescue assessment method and system based on MR technology, which combines MR technology to realize real-time visual assessment, intelligent task adjustment and remote collaborative support, thereby improving the intelligence level and rescue efficiency of aviation medical rescue.

[0006] An aviation medical rescue assessment method based on MR technology includes the following steps: Step S1: multi-dimensional data collection, real-time collection of multi-dimensional data related to aviation medical rescue, including patient vital signs data, aircraft status data, rescue personnel operation data and medical resource allocation data; Step S2: Fusion data display: Use Kalman filter to clean and fuse the collected multi-dimensional data to obtain fused data, use MR technology to generate a visualization interface, and display the fused data in real time in the MR device. The display content includes the patient's health status, aircraft flight status and resource allocation for reference by decision makers and medical staff; Step S3: Real-time task evaluation: input the fused data into the trained aviation rescue evaluation model for real-time evaluation. The evaluation content includes the patient's health status, the rescuer's operational standardization, the aircraft's flight status and the efficiency of medical resource utilization. The real-time evaluation results are displayed on the MR device and provide feedback to the decision makers. Step S4: Automatic task adjustment: According to the real-time evaluation results, the aviation medical rescue task is automatically adjusted. The adjustment content includes flight path optimization, medical plan adjustment, and reallocation of medical resources. All adjustment contents provide real-time dynamic feedback to the rescue team and decision makers through the MR device. Step S5: Remote collaborative adjustment, using MR technology to achieve real-time collaboration and guidance between remote experts and on-site medical staff, supporting remote experts to view on-site data through MR equipment, and providing operation guidance, medical advice and task adjustment suggestions based on real-time fusion data, and automatically generating auxiliary adjustment suggestions; Step S6: Mission record evaluation. In the aviation medical rescue mission, the complete rescue mission process is recorded, the effectiveness of the mission execution is evaluated, and improvement suggestions are made based on the evaluation results. The evaluation content includes mission execution time, medical effect and resource utilization efficiency. After the mission is completed, the evaluation results and improvement suggestions are displayed through the MR equipment.

[0007] As a preferred technical solution of the present invention, the multidimensional data specifically includes: The patient's vital signs data collected by biosensors include the patient's heart rate, blood pressure, body temperature, blood oxygen saturation and respiratory rate, and the biosensors include electrocardiogram monitors, pulse oximeters, temperature sensors, and blood pressure sensors; Aircraft status data collected by the aircraft’s monitoring equipment, including the aircraft’s flight altitude, speed, range, flight trajectory, aircraft position, fuel status, and weather data. The monitoring equipment includes flight recorders, black boxes, GPS navigation equipment, barometric altimeters, and flight weather monitoring systems; The rescuer's operation data collected by the motion capture device includes the rescuer's operation path, surgical steps, medical intervention operations, drug use and equipment deployment, and the motion capture device includes a head-mounted display, a gesture recognition system, a motion sensor and a three-dimensional positioning system; The medical resource configuration data collected through the Internet of Things monitoring equipment includes the usage of medical equipment, drug inventory, equipment status, medical resource allocation, and material requirements; the Internet of Things monitoring equipment includes a smart medicine cabinet, a medical equipment management system, a bed management system, and a resource allocation monitoring system.

[0008] As a preferred technical solution of the present invention, the data cleaning and fusion includes: Pre-process the collected multi-dimensional data to remove noise and outliers to ensure data accuracy and consistency, including removing noise from patients' vital signs data; removing data points beyond the physiological range by setting threshold ranges; removing noise from aircraft flight trajectories; and checking the data integrity of medical resource allocation data; The Kalman filter is used to fuse multidimensional data to improve the accuracy and timeliness of the data. The multidimensional data collected in real time is taken as the observation value, and the Kalman filter model is used to fuse the observation value. The state vector is , the observed value is , the data fusion process is carried out through the following derivation formula: ,in is the predicted value, is the Kalman gain, is the observation matrix, is the estimated value after fusion; The different data in the multi-source data are weighted and fused. The weight of each data is adjusted according to the accuracy and reliability of the data acquisition equipment. The contribution value of each data is fused using the weighted average method. The formula is: ,in is the weight of the data, For the i data observations, is the number of data; By comparing the synchronized data of different data, a consistency check is performed. If the difference value is found to be outside the set range, the exception handling mechanism is triggered.

[0009] As a preferred technical solution of the present invention, the visual interface includes the following units: Patient health status unit: used to display the patient's real-time vital signs data, present the numerical change trend of each vital sign through a graphical interface, and provide health warning prompts; Aircraft flight status unit: used to display the flight status data of the aircraft, present it through dynamic maps or 3D models, and update the flight data in real time; Medical resource allocation unit: used to display medical resource allocation data, display the allocation of medical resources through bar charts or pie charts, and automatically mark scarce resources; Rescuer operation path unit: used to display rescuer operation data, display rescuer's actions and operation paths in real time during the rescue process through 3D models or path diagrams, and mark key medical operation steps; Real-time evaluation result unit: used to display the results of real-time evaluation through a combination of charts, numbers and graphs, and to identify the execution status of tasks through color coding; Mission progress and adjustment suggestion unit: used to display the progress and suggested adjustments of the aviation medical rescue mission, and display the various stages of the mission through a progress bar and timeline.

[0010] As a preferred technical solution of the present invention, the structure of the aviation rescue assessment model includes: Input layer: used to receive fused data including patients’ vital signs data, aircraft flight status data, rescue personnel operation data and medical resource allocation data, and convert the fused data into a unified format through standardized processing for further analysis by the aviation rescue assessment model; Feature extraction layer: used to extract key features from fused data for evaluation, including time window sliding features of patient vital signs data, flight range / flight path features of flight status data, action path features of rescuer operation data, and usage and allocation features of medical resource configuration data; Evaluation layer: used to conduct real-time evaluation based on the extracted feature data, including health status evaluation, operational normative evaluation, flight status evaluation and utilization efficiency evaluation; Output layer: used to present the evaluation results to decision makers in a visual way to help them make subsequent task adjustment decisions.

[0011] As a preferred technical solution of the present invention, the training of the aviation rescue assessment model includes the following steps: Step A1: Prepare training data. The training process is based on historical mission data of aviation medical rescue. The mission data includes patient vital signs data, flight status data, rescuer operation data and medical resource configuration data. After preprocessing and feature extraction of the mission data, a training data set is formed. ,in is the feature vector of multidimensional data, The corresponding labels are the results of task evaluation; the training dataset is divided into training set, validation set and test set using the holdout method; Step A2. Model initialization and training: Random forest and support vector machine training are used to construct the initial aviation rescue assessment model. The structure of the model is randomly initialized at the beginning of training, and the initial values ​​of the weight matrix W and the bias term b are given. The model is continuously updated through the training process to obtain an optimized aviation rescue assessment model. Step A3. Use the loss function to measure the error between the model's prediction results and the label. The model is updated using the loss function as follows: ,in is the predicted value of the model, is the loss function, which is used to measure the gap between the model prediction value and the label. The total number of data samples used for training; Loss function in the aviation rescue evaluation model: For regression tasks, the mean square error is used, and the formula is expressed as: ; For classification tasks, cross entropy loss is used, and the formula is expressed as: , where C is the number of categories, For data samples i In category k Tags, is the corresponding prediction probability; the regression task includes health status assessment and flight status assessment, and the classification task includes operation normativeness assessment and use efficiency assessment; Step A4. Use the gradient descent method to optimize the loss function and update the weights and bias terms of the model through back propagation according to the following rules: ,in Represents the learning rate, which is used to control the step size of weight update in each iterative training; Step A5. Adjust the model complexity based on the performance of the aviation rescue evaluation model on the validation set, specifically by adjusting the parameters through L2 regularization The formula is: ; Step A6. After each training, the test set is used to evaluate the performance of the aviation rescue evaluation model, and a comprehensive evaluation is performed through the mean square error and accuracy until the comprehensive evaluation value exceeds the set threshold, that is, the training is completed and the final aviation rescue evaluation model is obtained.

[0012] As a preferred technical solution of the present invention, the automatic adjustment of the aviation medical rescue mission includes: Flight path optimization: matching the flight path according to the flight status and the patient's health status; avoiding areas with adverse climate; Adjust the medical plan. By combining the patient's health status with the rescuer's operational norms, it can be determined whether the patient's condition has changed dramatically, and a strategy for adjusting the medical plan can be automatically generated. Medical resources are reallocated based on the efficiency evaluation. If the evaluation shows that resources are being consumed too quickly or are in short supply, the system automatically calculates the required resources and generates a resource adjustment strategy. Automatic adjustment strategies are generated by combining real-time evaluation results with linear programming algorithms; The generation and implementation of automatic adjustment strategies are fed back to the rescue team and decision makers in real time through MR equipment; during the implementation process, if the environment of the aeromedical rescue mission changes, the adjustment strategy is automatically updated based on the new evaluation results, and the latest adjustment information is pushed in real time.

[0013] As a preferred technical solution of the present invention, the task record evaluation specifically includes: By recording the time of each stage in the execution of the aviation medical rescue mission, the overall time efficiency of the rescue mission is evaluated, including: the time from the start to the completion of the mission, the time of the key mission stage; Based on the time efficiency evaluation results, make suggestions for optimizing the task process; Evaluate the effects of automatic adjustments in tasks, including changes in patient health status, comparison of treatment outcomes, and recommendations for optimizing medical plans; Evaluate the efficiency of medical resource utilization in tasks, including resource consumption and efficiency analysis, resource allocation rationality assessment, and resource optimization suggestions; After the task is completed, a task summary report is generated, which records every step of the task execution process in detail and puts forward suggestions for improvement, including a summary of the task completion, problems encountered during the execution process, and suggestions for improvement of subsequent tasks; Through MR equipment, mission evaluation results and improvement suggestions can be displayed to help decision makers and rescue teams review the execution of the mission and make adjustments and optimizations to future missions based on the evaluation results.

[0014] An aviation medical rescue assessment system based on MR technology includes the following modules: Multi-dimensional data acquisition module: used to collect multi-dimensional data related to aviation medical rescue in real time; Fusion data display module: used to clean and fuse the collected multi-dimensional data through Kalman filter to obtain fused data, generate a visualization interface using MR technology, and display the fused data in real time in the MR device; Mission real-time assessment module: used to input fusion data into the trained aviation rescue assessment model for real-time assessment, display the real-time assessment results through MR equipment, and provide feedback to decision makers; Automatic mission adjustment module: used to automatically adjust the aviation medical rescue mission according to the real-time evaluation results, and provide real-time dynamic feedback to the rescue team and decision-makers through MR equipment; Remote collaboration adjustment module: Realize real-time collaboration and guidance between remote experts and on-site medical staff through MR technology; Mission record evaluation module: In aviation medical rescue missions, the complete rescue mission process is recorded, the effectiveness of mission execution is evaluated, and improvement suggestions are made based on the evaluation results.

[0015] The present invention has the following advantages: This solution uses multi-dimensional data collection to comprehensively integrate data such as patient vital signs, aircraft flight status, rescue personnel operations, and medical resource allocation. The comprehensive data source ensures that decision makers can obtain a comprehensive, real-time, and accurate situational awareness, avoiding the one-sided information brought about by a single data source in traditional technology. The Kalman filter is used to clean and fuse these data, which significantly improves the accuracy and timeliness of the data, helping rescue teams make more scientific and reliable decisions.

[0016] This solution uses MR technology to visualize data as three-dimensional images, allowing rescue workers and remote experts to intuitively view mission progress, patient health status, aircraft flight status, and resource allocation. MR equipment provides a more immersive and interactive interface, improving information transmission efficiency and decision-making response speed; rescue workers can obtain real-time on-site dynamics through MR equipment and respond quickly, reducing decision-making delays and errors.

[0017] The present invention uses an aviation rescue assessment model to evaluate mission progress in real time based on fused data; it instantly monitors and evaluates the patient's health status, flight status, and medical resource utilization efficiency, and provides real-time feedback, so that the rescue mission can adjust its strategy at any time during execution to achieve accurate dynamic optimization of the mission.

[0018] The present invention realizes a more efficient and flexible rescue process through automatic adjustment combined with remote collaborative adjustment, maximizes resource utilization efficiency, ensures that patients receive the most effective treatment in the shortest time, provides accurate operational guidance and medical advice, and helps on-site teams respond quickly to emergencies.

[0019] The present invention records the entire process of the aviation medical rescue mission and provides real-time evaluation and subsequent optimization suggestions. After the mission is completed, a complete rescue mission report is generated to evaluate the execution effect and make optimization suggestions. Through the intelligent mission recording and evaluation mechanism, the rescue team can deeply analyze the performance of each mission, optimize the process and strategy of future missions, thereby continuously improving rescue capabilities and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The present invention is a schematic diagram of the structure of an aviation medical rescue assessment system based on MR technology used in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0022] Embodiment 1, an aviation medical rescue assessment method based on MR technology, comprising the following steps: Step S1: multi-dimensional data collection, real-time collection of multi-dimensional data related to aviation medical rescue, including patient vital signs data, aircraft status data, rescue personnel operation data and medical resource allocation data; The multidimensional data specifically includes: The patient's vital signs data collected by biosensors include the patient's heart rate, blood pressure, body temperature, blood oxygen saturation and respiratory rate, and the biosensors include electrocardiogram monitors, pulse oximeters, temperature sensors, and blood pressure sensors; Aircraft status data collected by the aircraft’s monitoring equipment, including the aircraft’s flight altitude, speed, range, flight trajectory, aircraft position, fuel status, and weather data. The monitoring equipment includes flight recorders, black boxes, GPS navigation equipment, barometric altimeters, and flight weather monitoring systems; The rescuer's operation data collected by the motion capture device includes the rescuer's operation path, surgical steps, medical intervention operations, drug use and equipment deployment, and the motion capture device includes a head-mounted display, a gesture recognition system, a motion sensor and a three-dimensional positioning system; The medical resource configuration data collected through the Internet of Things monitoring equipment includes the usage of medical equipment, drug inventory, equipment status, medical resource allocation, and material requirements; the Internet of Things monitoring equipment includes a smart medicine cabinet, a medical equipment management system, a bed management system, and a resource allocation monitoring system.

[0023] Step S2: Fusion data display: Use Kalman filter to clean and fuse the collected multi-dimensional data to obtain fused data, use MR technology to generate a visualization interface, and display the fused data in real time in the MR device. The display content includes the patient's health status, aircraft flight status and resource allocation for reference by decision makers and medical staff; The data cleaning and fusion include: Pre-process the collected multi-dimensional data to remove noise and outliers to ensure data accuracy and consistency, including removing noise from patient vital signs data, such as using filters to remove artifacts in electrocardiogram data; removing unreasonable data points or data points that exceed the physiological range by setting threshold ranges, such as abnormal data with a heart rate higher than 200 beats / minute or lower than 30 beats / minute; denoising the flight trajectory of aircraft to remove abnormal flight trajectory points caused by weather or sensor failure; and performing data integrity checks on medical resource allocation data to ensure that equipment usage and drug inventory data are not empty or erroneous values; The Kalman filter is used to fuse multidimensional data to improve the accuracy and timeliness of the data. The multidimensional data collected in real time is taken as the observation value, and the Kalman filter model is used to fuse the observation value (such as the patient's health status and the aircraft position). The state vector is , the observed value is , the data fusion process is carried out through the following derivation formula: ,in is the predicted value, is the Kalman gain, is the observation matrix, is the estimated value after fusion; The different data in the multi-source data are weighted and fused. The weight of each data is adjusted according to the accuracy and reliability of the data acquisition equipment. The contribution value of each data is fused using the weighted average method. The formula is: ,in is the weight of the data, For the i data observations, is the number of data; By comparing the synchronized data of different data, a consistency check is performed. If the difference value is found to be outside the set range, the exception handling mechanism is triggered.

[0024] The specific method is: Compare the data collected by different sensors at the same time (such as the synchronized time points in biosensors and motion capture devices). If inconsistent data is found, use interpolation or replace it with high-priority data.

[0025] Through the above data cleaning and fusion process, the accuracy, timeliness and consistency of multidimensional data are ensured, providing reliable data support for subsequent evaluation and decision-making.

[0026] The visualization interface includes the following units: Patient health status unit: used to display the patient's real-time vital signs data, present the numerical change trend of each vital sign through a graphical interface, and provide health warning prompts (display a warning sign when exceeding the set threshold); Aircraft flight status unit: used to display the flight status data of the aircraft, present it through dynamic maps or 3D models, and update the flight data in real time to ensure that decision makers are aware of the flight status in a timely manner; Medical resource allocation unit: used to display medical resource allocation data, display the allocation of medical resources through bar charts or pie charts, and automatically mark scarce resources; Rescuer operation path unit: used to display rescuer operation data, display rescuer's actions and operation paths in real time during the rescue process through 3D models or path diagrams, and mark key medical operation steps; Real-time evaluation result unit: used to display the results of real-time evaluation through a combination of charts, numbers and graphs, and to identify the execution status of tasks through color coding (green for normal, yellow for warning, and red for emergency); Mission progress and adjustment suggestion unit: used to display the progress and suggested adjustments of the aviation medical rescue mission. It displays the various stages of the mission through a progress bar and timeline to help decision makers track the progress of the rescue mission.

[0027] Step S3: Real-time task evaluation: input the fused data into the trained aviation rescue evaluation model for real-time evaluation. The evaluation content includes the patient's health status, the rescuer's operational standardization, the aircraft's flight status and the efficiency of medical resource utilization. The real-time evaluation results are displayed on the MR device and provide feedback to the decision makers. The structure of the aviation rescue assessment model includes: Input layer: used to receive fused data including patients’ vital signs data, aircraft flight status data, rescue personnel operation data and medical resource allocation data, and convert the fused data into a unified format through standardized processing for further analysis by the aviation rescue assessment model; Feature extraction layer: used to extract key features from fused data for evaluation, including time window sliding features of patient vital signs data, flight range / flight path features of flight status data, action path features of rescuer operation data, and usage and allocation features of medical resource configuration data; The features include: Vital signs data: Extract changing trends, abnormal fluctuations, and features of rapid changes through time series analysis to assess health risks.

[0028] Flight status data: Optimize flight paths and ensure safety through trajectory analysis, flight speed changes, and weather data feature extraction.

[0029] Rescuer operation data: Ensure operational efficiency and accuracy through action path, operation timeliness and standardization evaluation.

[0030] Medical resource allocation data: Optimize resource allocation and avoid waste through analysis of resource usage and allocation efficiency.

[0031] The specific features extracted include: Feature extraction of patient vital signs data: Time window sliding feature extraction: For continuous data such as heart rate, blood oxygen, body temperature, and respiratory rate, a sliding time window (10 seconds, 30 seconds) is used to extract short-term statistical features, including: Mean: The average value of vital signs in each time window, which is used to reflect the stability of the patient.

[0032] Variance: It is used to measure the fluctuation of data. The larger the variance, the more drastic the changes in vital signs, which may indicate a sharp change in the condition.

[0033] Max / Min: Extracts the highest and lowest values ​​within each time window to capture extreme changes.

[0034] Rise rate / fall rate: Calculates the rate of change of vital signs within each time window. Sudden conditions (rapid heartbeat or shortness of breath) are usually accompanied by rapid changes.

[0035] Frequency domain features: Fast Fourier transform (FFT) is applied to extract frequency domain features (heartbeat frequency) and identify potential periodic patterns.

[0036] Abnormal pattern recognition: By comparing the current time window with historical data, a threshold-based rule algorithm is used to identify abnormal fluctuations and mark them as potential health risks.

[0037] Feature extraction of flight status data: Range / Flight Path Features: Extract aircraft flight trajectory data: Position change (displacement): Calculate the displacement of each moment relative to the previous moment to infer the change of the flight path.

[0038] Remaining distance: Calculate the remaining distance from the current location to the destination and estimate the flight time based on the current flight speed.

[0039] Flight speed change: Calculates the acceleration or deceleration of the aircraft at each moment to analyze whether there are abnormal speed fluctuations during the flight.

[0040] Meteorological data features: Extract wind speed, air pressure, and temperature information from meteorological data to analyze the impact of weather on flight path or time. This includes the average, maximum, and change rate features of meteorological data.

[0041] Aircraft health status: Extract features based on flight status and system parameters (fuel, engine temperature, body load), and use multivariate analysis to determine if the aircraft has potential risks.

[0042] Feature extraction of rescuer operation data: Motion path features: Extract the rescuer’s operation path features through motion capture data: Path length: Calculates the total distance a rescuer moves while performing a mission, measuring the overall amount of activity performed during the mission.

[0043] Speed ​​change: Based on the time interval and distance between each motion capture point, the rescuer’s movement speed and acceleration are calculated to evaluate the operation efficiency.

[0044] Operation time: Calculate the execution time of each operation step. By analyzing the duration of each step, determine whether there are delays or irregular operations.

[0045] Action sequence and standardization: By identifying the order of action sequences (sequence matching technology), it is detected whether the rescuers are following the prescribed medical steps.

[0046] Key operation points: Identify key operation points during the task (drug injection, equipment preparation, surgical cutting), and determine whether the operation is standardized by evaluating the time interval and accuracy of each key operation.

[0047] Feature extraction of medical resource configuration data: Resource usage characteristics: Frequency of use: Statistics on the frequency of use of medical equipment and drugs, the number of times the equipment is activated and the duration of use.

[0048] Equipment status change: Extract the working status characteristics of the equipment (on / off frequency, maintenance records) to determine whether the equipment is in a faulty state or requires maintenance.

[0049] Drug consumption: Calculate the drug consumption rate and correlate the consumption with factors such as task progress and number of patients to analyze whether there is a risk of drug overuse or shortage.

[0050] Resource allocation characteristics: Resource allocation efficiency: Calculate the proportion and efficiency of resource allocation based on task requirements and resource usage. Use linear programming to evaluate whether there is resource waste or imbalanced allocation.

[0051] Material supply chain characteristics: Extract inventory data (drug inventory, equipment availability), evaluate the health of the material supply chain, and predict possible material shortages in the future.

[0052] Evaluation layer: used to conduct real-time evaluation based on the extracted feature data, including health status assessment, which evaluates the patient's emergency status based on the characteristics of vital signs data and automatically determines whether immediate intervention is needed; operational normativeness assessment, which combines the rescue personnel's operating path to evaluate the timeliness and normativeness of their operations, and whether there are redundant operations or delays; flight status assessment, which combines flight data and meteorological information to evaluate the safety of the aircraft, whether there are risks such as course deviation and excessive flight time, and utilization efficiency assessment, which evaluates the rationality of resource allocation based on the configuration and use of medical resources to ensure that resources are optimally utilized.

[0053] Output layer: used to present the evaluation results to decision makers in a visual way to help them make subsequent task adjustment decisions.

[0054] The training of the aviation rescue assessment model in step S3 includes the following steps: Step A1: Prepare training data. The training process is based on historical mission data of aviation medical rescue. The mission data includes patient vital signs data, flight status data, rescuer operation data and medical resource configuration data. After preprocessing and feature extraction of the mission data, a training data set is formed. ,in is the feature vector of multidimensional data, The corresponding labels are the results of task evaluation; the training dataset is divided into training set, validation set and test set using the holdout method; Step A2. Model initialization and training: Random forest and support vector machine training are used to construct the initial aviation rescue assessment model. The structure of the model is randomly initialized at the beginning of training, and the initial values ​​of the weight matrix W and the bias term b are given. The model is continuously updated through the training process to obtain an optimized aviation rescue assessment model. Step A3. Use the loss function to measure the error between the model's prediction results and the label. The model is updated using the loss function as follows: ,in is the predicted value of the model, is the loss function, which is used to measure the gap between the model prediction value and the label. The total number of data samples used for training; Loss function in the aviation rescue evaluation model: For regression tasks, the mean square error is used, and the formula is expressed as: ; For classification tasks, cross entropy loss is used, and the formula is expressed as: , where C is the number of categories, For data samples i In category k Tags, is the corresponding prediction probability; the regression task includes health status assessment and flight status assessment, and the classification task includes operation normativeness assessment and use efficiency assessment; Step A4. Use the gradient descent method to optimize the loss function and update the weights and bias terms of the model through back propagation according to the following rules: ,in Represents the learning rate, which is used to control the step size of weight update in each iterative training; Step A5. Adjust the model complexity based on the performance of the aviation rescue evaluation model on the validation set, specifically by adjusting the parameters through L2 regularization The formula is: ; Step A6. After each training, the test set is used to evaluate the performance of the aviation rescue evaluation model, and a comprehensive evaluation is performed through the mean square error and accuracy until the comprehensive evaluation value exceeds the set threshold, that is, the training is completed and the final aviation rescue evaluation model is obtained.

[0055] Through the above training process, the aviation rescue assessment model can automatically learn various data features in the assessment task based on historical mission data, gradually improve the accuracy of the prediction, and provide reliable real-time assessment support for subsequent rescue missions.

[0056] Step S4: Automatic task adjustment: According to the real-time evaluation results, the aviation medical rescue task is automatically adjusted. The adjustment content includes flight path optimization, medical plan adjustment, and reallocation of medical resources. All adjustment contents provide real-time dynamic feedback to the rescue team and decision makers through the MR device. The automatic adjustment of the aviation medical rescue mission includes: Flight path optimization: match the flight path according to the flight status and the patient's health status; avoid areas with adverse climates; if the assessment results show that the patient's health status is critical and there is no rescue condition at the current flight destination, recommend a new flight route to reduce flight time and ensure faster arrival at the destination; Medical plan adjustment: By combining the patient's health status with the rescuer's operational norms, it is determined whether the patient's condition has changed dramatically, such as the need for emergency intervention (such as cardiopulmonary resuscitation, oxygen support, etc.), and a strategy for adjusting the medical plan is automatically generated; Medical resources are reallocated based on the efficiency evaluation. If the evaluation shows that resources (drugs, equipment, personnel) are being consumed too quickly or are in short supply, the system automatically calculates the required resources and generates a resource adjustment strategy to ensure that key resources can be reasonably allocated and used throughout the rescue mission. Automatic adjustment strategies are generated by combining real-time evaluation results with linear programming algorithms; The optimization algorithm takes into account multiple constraints such as flight path, patient condition, resource usage, etc. to arrive at the best adjustment plan. For example, if the flight path adjustment involves multiple factors such as fuel consumption and flight distance, the most time-saving and resource-saving path is automatically calculated based on the input data.

[0057] The generation and implementation of automatic adjustment strategies are fed back to the rescue team and decision makers in real time through MR equipment. During the implementation process, if the environment of the aviation medical rescue mission changes (such as sudden weather changes during the flight or further deterioration of the patient's condition), the adjustment strategy is automatically updated based on the new evaluation results, and the latest adjustment information is pushed in real time to ensure the flexibility and efficiency of mission execution.

[0058] Automatic adjustment strategies are generated and implemented through real-time evaluation results, optimization algorithms, and dynamic feedback mechanisms to ensure the efficient execution of aeromedical rescue missions and maximize the patient's chance of survival.

[0059] Step S5: Remote collaborative adjustment, using MR technology to achieve real-time collaboration and guidance between remote experts and on-site medical staff, supporting remote experts to view on-site data through MR equipment, and providing operation guidance, medical advice and task adjustment suggestions based on real-time fusion data, automatically generating auxiliary adjustment suggestions, and ensuring that on-site rescue complies with best medical practices; Step S6: Mission record evaluation. In the aviation medical rescue mission, the complete rescue mission process is recorded, the effectiveness of the mission execution is evaluated, and improvement suggestions are made based on the evaluation results. The evaluation content includes mission execution time, medical effect and resource utilization efficiency. After the mission is completed, the evaluation results and improvement suggestions are displayed through the MR equipment.

[0060] The task record assessment specifically includes: By recording the time of each stage in the execution of the aviation medical rescue mission, the overall time efficiency of the rescue mission is evaluated, including: the time from the start of the mission to the completion of the mission, calculating the total time from the receipt of the mission to the completion of the mission, and evaluating whether it is completed within the scheduled time; the time of the key mission stage, recording the time spent on key mission nodes (patient reception, flight, medical intervention), evaluating whether there is a delay, and analyzing the time bottleneck in the mission process; Based on the time efficiency evaluation results, make suggestions for optimizing the task process, such as reducing unnecessary operations or increasing the rescue speed; Automatic adjustment effects in evaluation tasks, including changes in the patient's health status, by analyzing the patient's vital signs data (heart rate, blood oxygen, body temperature), to evaluate whether the medical measures are effective and whether the patient's health status has been improved within a reasonable time; comparison of treatment results, comparing the actual treatment effect with the predetermined treatment goal, and analyzing whether the treatment effect meets expectations; medical plan optimization suggestions, based on the medical effect evaluation, put forward optimization suggestions for the medical plan to ensure that the treatment measures in subsequent tasks are more effective; Evaluate the efficiency of the use of medical resources in the task, including resource consumption and efficiency analysis, record the consumption and usage time of various medical resources, and analyze whether there is waste or shortage of resources; evaluate the rationality of resource allocation, based on resource consumption data, evaluate whether resources are reasonably allocated, and whether it is necessary to adjust resource priorities or allocation strategies; resource optimization suggestions, based on the evaluation results, put forward suggestions for optimizing resource use, such as reducing resource waste and improving task efficiency through reasonable scheduling and allocation; After the task is completed, a task summary report is generated, which records every step of the task execution process in detail and puts forward suggestions for improvement, including: summary of task completion, whether the task was completed according to the predetermined goal, whether there are any unfinished parts or obstacles, and evaluation of the quality of task execution; problems encountered during execution, recording problems encountered during task execution (such as equipment failure, resource shortage, etc.) and their solutions, providing experience for subsequent tasks; suggestions for improvement of subsequent tasks, based on the results of task evaluation, put forward improvement measures for future tasks, such as optimizing processes, improving efficiency, reducing risks, etc.; Through MR equipment, mission evaluation results and improvement suggestions can be displayed to help decision makers and rescue teams review the execution of the mission and make adjustments and optimizations to future missions based on the evaluation results.

[0061] Through the mission record evaluation content, the implementation of the aviation medical rescue mission is comprehensively summarized to ensure that the mission execution effect is evaluated, and provide data support and improvement suggestions to optimize the implementation of subsequent rescue missions.

[0062] Example 2, an aviation medical rescue assessment system based on MR technology, see Figure 1 As shown, it includes the following modules: Multi-dimensional data acquisition module: used to collect multi-dimensional data related to aviation medical rescue in real time; Fusion data display module: used to clean and fuse the collected multi-dimensional data through Kalman filter to obtain fused data, generate a visualization interface using MR technology, and display the fused data in real time in the MR device; Mission real-time assessment module: used to input fusion data into the trained aviation rescue assessment model for real-time assessment, display the real-time assessment results through MR equipment, and provide feedback to decision makers; Automatic mission adjustment module: used to automatically adjust the aviation medical rescue mission according to the real-time evaluation results, and provide real-time dynamic feedback to the rescue team and decision-makers through MR equipment; Remote collaboration adjustment module: Realize real-time collaboration and guidance between remote experts and on-site medical staff through MR technology; Mission record evaluation module: In aviation medical rescue missions, the complete rescue mission process is recorded, the effectiveness of mission execution is evaluated, and improvement suggestions are made based on the evaluation results.

[0063] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An aviation medical rescue assessment method based on MR technology, characterized in that: The following steps are involved: Step S1: multi-dimensional data collection, real-time collection of multi-dimensional data related to aviation medical rescue, including patient vital signs data, aircraft status data, rescue personnel operation data and medical resource allocation data; Step S2: Fusion data display: Use Kalman filter to clean and fuse the collected multi-dimensional data to obtain fused data, use MR technology to generate a visualization interface, and display the fused data in real time in the MR device. The display content includes the patient's health status, aircraft flight status and resource allocation for reference by decision makers and medical staff; Step S3: Real-time task evaluation: input the fused data into the trained aviation rescue evaluation model for real-time evaluation. The evaluation content includes the patient's health status, the rescuer's operational standardization, the aircraft's flight status and the efficiency of medical resource utilization. The real-time evaluation results are displayed on the MR device and provide feedback to the decision makers. Step S4: Automatic task adjustment: According to the real-time evaluation results, the aviation medical rescue task is automatically adjusted. The adjustment content includes flight path optimization, medical plan adjustment, and reallocation of medical resources. All adjustment contents provide real-time dynamic feedback to the rescue team and decision makers through the MR device. Step S5: Remote collaborative adjustment, using MR technology to achieve real-time collaboration and guidance between remote experts and on-site medical staff, supporting remote experts to view on-site data through MR equipment, and providing operation guidance, medical advice and task adjustment suggestions based on real-time fusion data, and automatically generating auxiliary adjustment suggestions; Step S6: Mission record evaluation. In the aviation medical rescue mission, the complete rescue mission process is recorded, the effectiveness of the mission execution is evaluated, and improvement suggestions are made based on the evaluation results. The evaluation content includes mission execution time, medical effect and resource utilization efficiency. After the mission is completed, the evaluation results and improvement suggestions are displayed through the MR equipment.

2. The MR-based aviation medical rescue assessment method according to claim 1, characterized in that: The multidimensional data specifically includes: The patient's vital signs data collected by biosensors include the patient's heart rate, blood pressure, body temperature, blood oxygen saturation and respiratory rate, and the biosensors include electrocardiogram monitors, pulse oximeters, temperature sensors, and blood pressure sensors; Aircraft status data collected by the aircraft’s monitoring equipment, including the aircraft’s flight altitude, speed, range, flight trajectory, aircraft position, fuel status, and weather data. The monitoring equipment includes flight recorders, black boxes, GPS navigation equipment, barometric altimeters, and flight weather monitoring systems; The rescuer's operation data collected by the motion capture device includes the rescuer's operation path, surgical steps, medical intervention operations, drug use and equipment deployment, and the motion capture device includes a head-mounted display, a gesture recognition system, a motion sensor and a three-dimensional positioning system; The medical resource configuration data collected through the Internet of Things monitoring equipment includes the usage of medical equipment, drug inventory, equipment status, medical resource allocation, and material requirements; the Internet of Things monitoring equipment includes a smart medicine cabinet, a medical equipment management system, a bed management system, and a resource allocation monitoring system.

3. The MR-based aviation medical rescue assessment method according to claim 1, characterized in that: The data cleaning and fusion include: Pre-process the collected multi-dimensional data to remove noise and outliers to ensure data accuracy and consistency, including removing noise from patients' vital signs data; removing data points beyond the physiological range by setting threshold ranges; removing noise from aircraft flight trajectories; and checking the data integrity of medical resource allocation data; The Kalman filter is used to fuse multidimensional data to improve the accuracy and timeliness of the data. The multidimensional data collected in real time is assumed to be the observation value, and the Kalman filter model is used to fuse the observation value.

4. The MR-based aviation medical rescue assessment method according to claim 1, characterized in that: The visualization interface includes the following units: Patient health status unit: used to display the patient's real-time vital signs data, present the numerical change trend of each vital sign through a graphical interface, and provide health warning prompts; Aircraft flight status unit: used to display the flight status data of the aircraft, present it through dynamic maps or 3D models, and update the flight data in real time; Medical resource allocation unit: used to display medical resource allocation data, display the allocation of medical resources through bar charts or pie charts, and automatically mark scarce resources; Rescuer operation path unit: used to display rescuer operation data, display rescuer's actions and operation paths in real time during the rescue process through 3D models or path diagrams, and mark key medical operation steps; Real-time evaluation result unit: used to display the results of real-time evaluation through a combination of charts, numbers and graphs, and to identify the execution status of tasks through color coding; Mission progress and adjustment suggestion unit: used to display the progress and suggested adjustments of the aviation medical rescue mission, and display the various stages of the mission through a progress bar and timeline.

5. The MR-based aviation medical rescue assessment method according to claim 1, characterized in that: The structure of the aviation rescue assessment model includes: Input layer: used to receive fused data including patients’ vital signs data, aircraft flight status data, rescue personnel operation data and medical resource allocation data, and convert the fused data into a unified format through standardized processing for further analysis by the aviation rescue assessment model; Feature extraction layer: used to extract key features from fused data for evaluation, including time window sliding features of patient vital signs data, flight range / flight path features of flight status data, action path features of rescuer operation data, and usage and allocation features of medical resource configuration data; Evaluation layer: used to conduct real-time evaluation based on the extracted feature data, including health status evaluation, operational normative evaluation, flight status evaluation and utilization efficiency evaluation; Output layer: used to present the evaluation results to decision makers in a visual way to help them make subsequent task adjustment decisions.

6. The MR-based aviation medical rescue assessment method according to claim 5, characterized in that: The training of the aviation rescue assessment model includes the following steps: Step A1: prepare training data. The training process is based on historical mission data of aviation medical rescue. The mission data includes patient vital signs data, flight status data, rescuer operation data and medical resource configuration data. After preprocessing and feature extraction of the mission data, a training data set is formed. The training data set is divided into a training set, a validation set and a test set using the holdout method. Step A2. Model initialization and training: Random forest and support vector machine training are used to construct the initial aviation rescue assessment model. The structure of the model is randomly initialized at the beginning of training, and the initial values ​​of the weight matrix W and the bias term b are given. The model is continuously updated through the training process to obtain an optimized aviation rescue assessment model. Step A3. Use the loss function to measure the error between the model's prediction results and the labels; Step A4. Use the gradient descent method to optimize the loss function and update the weights and bias terms of the model through back propagation; Step A5. Adjust the model complexity according to the performance of the aviation rescue evaluation model on the validation set; Step A6. After each training, the test set is used to evaluate the performance of the aviation rescue evaluation model, and a comprehensive evaluation is performed through the mean square error and accuracy until the comprehensive evaluation value exceeds the set threshold, that is, the training is completed and the final aviation rescue evaluation model is obtained.

7. The MR-based aviation medical rescue assessment method according to claim 1, characterized in that: The automatic adjustment of the aviation medical rescue mission includes: Flight path optimization: matching the flight path according to the flight status and the patient's health status; avoiding areas with adverse climate; Adjust the medical plan. By combining the patient's health status with the rescuer's operational norms, it can be determined whether the patient's condition has changed dramatically, and a strategy for adjusting the medical plan can be automatically generated. Medical resources are reallocated based on the efficiency evaluation. If the evaluation shows that resources are being consumed too quickly or are in short supply, the system automatically calculates the required resources and generates a resource adjustment strategy. Automatic adjustment strategies are generated by combining real-time evaluation results with linear programming algorithms; The generation and implementation of automatic adjustment strategies are fed back to the rescue team and decision makers in real time through MR equipment; during the implementation process, if the environment of the aeromedical rescue mission changes, the adjustment strategy is automatically updated based on the new evaluation results, and the latest adjustment information is pushed in real time.

8. The MR-based aviation medical rescue assessment method according to claim 1, characterized in that: The task record assessment specifically includes: By recording the time of each stage in the execution of the aviation medical rescue mission, the overall time efficiency of the rescue mission is evaluated, including: the time from the start to the completion of the mission, the time of the key mission stage; Based on the time efficiency evaluation results, make suggestions for optimizing the task process; Evaluate the effects of automatic adjustments in tasks, including changes in patient health status, comparison of treatment outcomes, and recommendations for optimizing medical plans; Evaluate the efficiency of medical resource utilization in tasks, including resource consumption and efficiency analysis, resource allocation rationality assessment, and resource optimization suggestions; After the task is completed, a task summary report is generated, which records every step of the task execution process in detail and puts forward suggestions for improvement, including a summary of the task completion, problems encountered during the execution process, and suggestions for improvement of subsequent tasks; Through MR equipment, mission evaluation results and improvement suggestions can be displayed to help decision makers and rescue teams review the execution of the mission and make adjustments and optimizations to future missions based on the evaluation results.

9. An aviation medical rescue assessment system based on MR technology, characterized in that: The system applies the aviation medical rescue assessment method based on MR technology as described in any one of claims 1 to 8 above, and includes the following modules: Multi-dimensional data acquisition module: used to collect multi-dimensional data related to aviation medical rescue in real time; Fusion data display module: used to clean and fuse the collected multi-dimensional data through Kalman filter to obtain fused data, generate a visualization interface using MR technology, and display the fused data in real time in the MR device; Mission real-time assessment module: used to input fusion data into the trained aviation rescue assessment model for real-time assessment, display the real-time assessment results through MR equipment, and provide feedback to decision makers; Automatic task adjustment module: used to automatically adjust the aviation medical rescue task according to the real-time evaluation results, and provide real-time dynamic feedback to the rescue team and decision-makers through MR equipment; Remote collaboration adjustment module: Realize real-time collaboration and guidance between remote experts and on-site medical staff through MR technology; Mission record evaluation module: In aviation medical rescue missions, the complete rescue mission process is recorded, the effectiveness of mission execution is evaluated, and improvement suggestions are made based on the evaluation results.

Images