An aviation medical rescue simulation training adaptability evaluation method and system

Abstract

The application provides an aviation medical rescue simulation training adaptability evaluation method and system, relates to the medical rescue simulation training evaluation technical field, and mainly comprises the following steps: collecting visual data of rescue operation of a trainee and heart rate data of the trainee in real time; extracting gravity center data of the trainee from the visual data of rescue operation of the trainee; analyzing the heart rate data and the gravity center data of the trainee, and outputting an adaptability evaluation result of the trainee. Through experimental data, it can be seen that the adaptability evaluation method provided by the application can more accurately complete the adaptability simulation training evaluation of rescue personnel, help the rescue personnel better understand the physical state and adaptability in the training process, improve the rescue skills of the rescue personnel, and has important significance for improving the aviation medical emergency rescue training quality and efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical rescue simulation training evaluation, in particular to an aviation medical rescue simulation training adaptability evaluation method and system. BACKGROUND

[0002] The preferred carrier for aviation medical emergency rescue is a helicopter. Medical personnel need to perform first aid rescue treatment for patients on the helicopter. However, most rescue personnel have not received rescue training on the helicopter. The vertical take-off and landing of the helicopter may cause overloading and weightlessness to the rescue personnel and equipment. In addition, the helicopter is affected by various factors during flight, which may cause severe jolts and loud noises. These sudden conditions during flight may cause the medical personnel to have stress reactions such as shock, resulting in muscle tension, increased heart rate, and shortness of breath, and even confusion or loss of consciousness, which may result in failure to remember the current operation procedure or unconscious error operation, which is not conducive to rescue. Therefore, it is necessary to carry out adaptability simulation training and evaluation in aviation medical rescue training to help medical personnel improve their ability to respond to sudden conditions and harsh environments. Current aviation medical rescue simulation training is mostly simulation training of operation procedures, and evaluation of personnel adaptability is mainly through manual and self-evaluation of participants, lacking scientific and systematic and intelligent evaluation methods. SUMMARY

[0003] To partially or completely solve the above technical problems, the present application proposes an aviation medical rescue simulation training adaptability evaluation method and system based on heart rate and center of gravity data, which can effectively complete the adaptability evaluation of participants.

[0004] In one aspect, the present application provides an aviation medical rescue simulation training adaptability evaluation method, comprising: collecting real-time visual data of rescue operation of a participant and heart rate data of the participant; extracting center of gravity data of the participant from the visual data of rescue operation of the participant; analyzing the heart rate data and the center of gravity data of the participant, and outputting an adaptability evaluation result of the participant.

[0005] Further, the collected heart rate data of the participant is converted into RR interval, and the RR interval is used to calculate heart rate characteristic parameters of the participant: mean heart rate MEAN and root mean square of adjacent RR interval difference RMSSD. The calculation formula of the mean heart rate MEAN includes:

[0006] ;

[0007] The calculation formula of the root mean square of adjacent RR interval difference RMSSD includes:

[0008] ;

[0009] wherein, represents the first RR interval value; represents the first RR interval value, represents the total number of RR intervals counted.

[0010] The mean heart rate MEAN and the root mean square of successive differences RMSSD of adjacent RR intervals can reflect the dynamic changes of heart rate regulation and the regulation ability of the human autonomic nervous system.

[0011] Further, the method for extracting the barycenter data from the real-time collected visual data of the rescuer operation of the participant is:

[0012] Step a1, using the MultiPose human posture recognition algorithm to recognize and calculate the human key point coordinates from the visual data of the rescuer operation of the participant;

[0013] Step a2, based on the human key point coordinates obtained in step a1, supplementing the head coordinates, hand coordinates, foot heel coordinates, two shoulder midpoint coordinates and two hip midpoint coordinates to obtain 19 human key point coordinates required for calculating the human barycenter position coordinates by the method of moment synthesis;

[0014] Further, the hand coordinates are obtained by elongation according to the elbow coordinates and the wrist coordinates, and the specific formula includes:

[0015] ;

[0016] ;

[0017] wherein, represents the x-axis value of the elbow joint coordinates; represents the y-axis value of the elbow joint coordinates; represents the x-axis value of the wrist joint coordinates; represents the y-axis value of the wrist joint coordinates; represents the position ratio of the palm.

[0018] Further, the two shoulder midpoints are the midpoint coordinates of the double shoulder line, and the specific formula includes:

[0019] , ;

[0020] wherein, represents the x-axis value of the left shoulder joint coordinates; represents the y-axis value of the left shoulder joint coordinates; represents the x-axis value of the right shoulder joint coordinates; represents the y-axis value of the right shoulder joint coordinates.

[0021] Step a3: Use the moment synthesis method to calculate the coordinates of the trainee's center of gravity in each frame of the image. The specific formulas include:

[0022] ;

[0023] ;

[0024] in, Indicates the first The relative weight coefficient corresponding to an individual's joints. , Corresponding to the first The horizontal and vertical coordinates of key points on an individual's body.

[0025] Using the moment synthesis method to calculate the trainee's center of gravity eliminates the need for building a separate model for each trainee, significantly improving efficiency. Combined with the multiplier method, it can determine the trainee's center of gravity position data with relatively high accuracy. This method allows for the extraction of the trainee's center of gravity coordinates from each frame of the video.

[0026] Furthermore, after obtaining a series of time-sequential center-of-gravity coordinates, the center-of-gravity data can be analyzed to calculate the characteristic parameters of the trainee's human balance ability. The specific method is as follows:

[0027] Step b1: Calculate the length of the trainee's center of gravity movement trajectory in the visual data of the rescue operation. The specific formula includes:

[0028] ;

[0029] in, These represent the coordinates of the human body's center of gravity. This indicates that the index number of the human body's center of gravity position is extracted from the visual data. This represents the total number of coordinates of the human body's center of gravity that were collected.

[0030] The length of the trajectory of the center of gravity position of the human body reflects the amplitude of the body sway during actual training, and the degree to which the center of gravity deviates from its own center position reflects the stability of the trainee's own balance.

[0031] Step b2, calculate the area of ​​the trainee's center of gravity coordinate trajectory in the visual data of the rescue operation, which refers to the area of ​​the closed region enclosed by the center of gravity coordinate trajectory. The specific formula includes:

[0032] ;

[0033] in, This represents the greater distance between two adjacent center-of-gravity coordinates and the average center-of-gravity coordinate, with the average center-of-gravity coordinate being the origin. The area of ​​the trajectory of the center-of-gravity coordinates reflects the degree of body sway; generally, during training, increased body sway leads to an increase in the trajectory area.

[0034] Step b3, calculate the length of the trainee's center of gravity trajectory per unit area in the visual data of the rescue operation, which refers to the ratio of the length of the trajectory of the center of gravity position movement to the area of ​​the trajectory. The specific formula includes:

[0035] ;

[0036] The length of the center of gravity trajectory per unit area reflects the test subject's ability to maintain body balance with subtle postural control. This ratio decreases as the trainee's control ability improves.

[0037] Step b4: Calculate the variance of the center-of-gravity coordinates of all trainees in the visual data of the rescue operation. The specific formula includes:

[0038] ;

[0039] in, The average displacement along the x-axis represents the center-of-gravity coordinates of all human bodies. This represents the average y-axis displacement of the center of gravity of all individuals. A larger variance indicates poorer balance.

[0040] Furthermore, an adaptability assessment of the participants was conducted based on the aforementioned heart rate and balance ability characteristics, including the following equations:

[0041] ;

[0042] ;

[0043] ;

[0044] ;

[0045] ;

[0046] in, An adaptive evaluation equation representing the mean heart rate (MEAN); An adaptive evaluation equation representing the root mean square slowdown (RMSSD) of the difference between adjacent RR intervals of heart rate; An adaptive evaluation equation representing the variance of the center of gravity displacement; An adaptive evaluation equation representing the length of the centroid trajectory per unit area; MEAN represents the average heart rate. The root mean square (RMSSD) represents the difference between adjacent RR intervals of heart rate. Indicates the variance of the center of gravity displacement; This represents the length of the centroid trajectory per unit area. When the calculated adaptive evaluation results are greater than the maximum value in the predetermined adaptive evaluation table, the maximum value is taken; when the calculated adaptive evaluation results are less than the minimum value in the predetermined adaptive evaluation table, the minimum value is taken.

[0047] The system can automatically evaluate the adaptability of trainees using the adaptive evaluation equation and output the evaluation results without human intervention.

[0048] Preferably, the method for constructing the adaptive evaluation equations for the above-mentioned characteristic parameters is as follows:

[0049] Step c1: Using the mean heart rate (MEAN), the root mean square (RMSSD) of the difference between adjacent RR intervals of heart rate, the length of the centroid trajectory per unit area, and the variance of the centroid coordinate displacement as independent variables, and the adaptive evaluation results as the dependent variable, an adaptive evaluation equation is constructed using linear regression.

[0050] ;

[0051] in, This indicates the different independent variables selected. This indicates the results of the adaptation evaluation. Represents the regression coefficient. Represents the constant term. Indicates the index number of the independent variable;

[0052] Step c2 involves collecting multiple sets of training data from participants through experiments, using a predetermined adaptive evaluation form to obtain the participants' self-evaluation and manual evaluation results, and obtaining the coefficients and constant terms of each adaptive evaluation equation from the training data and evaluation results through statistical methods, thereby obtaining the adaptive equations for each characteristic parameter.

[0053] The experimental data shows that using the adaptive equation obtained by this method to evaluate the adaptability of trainees can yield relatively accurate evaluation results.

[0054] On the other hand, the present invention provides an adaptive evaluation system for aviation medical rescue simulation training, comprising:

[0055] The real-time data acquisition unit is used to collect visual data of the trainees' rescue operations and heart rate data in real time.

[0056] The center of gravity data extraction unit is used to extract the center of gravity data of trainees from the visual data of trainees' rescue operations.

[0057] The adaptability assessment unit is used to analyze the heart rate and center of gravity data of trainees and output the adaptability assessment results of trainees.

[0058] By adopting the above technical solution, the present invention has the following beneficial effects:

[0059] This invention provides a method and system for evaluating the adaptability of aviation medical rescue simulation training. It selects heart rate data, reflecting the function of the human autonomic nervous system, and center of gravity data, reflecting the body's balance, as evaluation elements. By collecting trainees' physiological indicators and visual data in real time during training, heart rate and center of gravity data are extracted. Based on the analysis of these data, characteristic parameters are obtained to further complete the adaptability evaluation and output quantitative evaluation results. Experimental data shows that the adaptability evaluation method provided by this invention can accurately evaluate the adaptive simulation training of rescue personnel, helping them better understand their physical state and adaptability during training, improving their rescue skills, and is of great significance for improving the quality and efficiency of aviation medical emergency rescue training. Attached Figure Description

[0060] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0061] Figure 1 A flowchart of an adaptive evaluation method for aviation medical rescue simulation training provided in an embodiment of the present invention;

[0062] Figure 2 This is a scenario for visual data acquisition by trainees during simulated training, as provided in an embodiment of the present invention.

[0063] Figure 3 The diagram shows a comparison between the coordinates of human key points identified by the MultiPose human pose recognition algorithm provided in this embodiment of the invention and the coordinates of human key points in the multiplier coefficient model used in the torque synthesis method. Figure a is a schematic diagram of human key points in the multiplier coefficient model, and Figure b is a schematic diagram of human key points identified by MultiPose.

[0064] Figure 4 This is a centroid trajectory diagram from a single simulated training session provided in an embodiment of the present invention.

[0065] Figure 5 A graph showing the comparison of mean heart rate (MEAN) under three experimental conditions provided in this embodiment of the invention.

[0066] Figure 6 The diagram shows the comparison of RMSSD data under three experimental conditions provided in this embodiment of the invention. Detailed Implementation

[0067] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] The present invention will be further explained below with reference to specific embodiments.

[0069] Heart rate variability (HRV) reflects the intensity and balance of the cardiac sympathetic and vagus nerve activities, and can be used to assess the activity level of the autonomic nervous system. In recent years, many studies have applied HRV to various fields, including mental fatigue, psychological stress, emotion recognition, and motor status. These studies have revealed a correlation between HRV and psychological state, making it a valuable set of adaptive assessment indicators. Balance ability reflects the ability to stably control movement; good balance can provide support for rescue operations in aviation medical emergency rescue. Maintaining good balance in unstable environments is an important assessment criterion for evaluating a trainee's adaptability to the aviation medical emergency rescue flight environment. In a dynamic environment, the position of the body's center of gravity directly reflects balance ability. Therefore, this invention selects heart rate and center of gravity data as evaluation criteria to design an assessment of the effectiveness of adaptive simulation training for aviation medical emergency rescue.

[0070] Example 1:

[0071] like Figure 1 The diagram illustrates an embodiment of the adaptive evaluation method for aviation medical rescue simulation training according to the present invention, which includes: real-time acquisition of visual data of the trainee's rescue operation and heart rate data of the trainee; extraction of the trainee's center of gravity data from the visual data of the trainee's rescue operation; analysis of the trainee's heart rate data and center of gravity data, and output of the trainee's adaptive evaluation results.

[0072] In practical implementation, before conducting adaptive training for trainees, it is essential to first construct an adaptive simulation training environment and operational procedures. This invention uses a six-degree-of-freedom motion gimbal to load a flight spectrum within a flight simulator to simulate helicopter flight. The enclosed environment and unexpected events may cause trainees to experience physical maladjustment or stress reactions, such as noise, changes in light and darkness, sudden weightlessness, or severe turbulence. Within this simulated environment, trainees undergo medical rescue simulation training, while video data and physiological indicators are collected in real-time during rescue operations. After the simulation training concludes, data collection is stopped, and the video data and physiological indicators are analyzed and evaluated to assess the trainees' adaptive training effectiveness.

[0073] In this embodiment, the purpose of collecting video data from trainees is to extract their center of gravity data, and the main physiological indicator collected is heart rate data.

[0074] In practice, a monocular camera is used to record the trainee performing first aid procedures. The camera is positioned directly in front of the trainee. To improve the versatility of the corresponding actions, the camera is kept roughly in front of the trainee during recording, although the position and angle can vary slightly. When collecting data in real time, a simulated training process is selected where the trainee performs first aid procedures from a fixed position. For example, when performing CPR compressions, the trainee is required to remain fixed to one side of the injured person to reduce data interference. Figure 2 This demonstrates the data acquisition scenario of this embodiment.

[0075] In practical implementation, a Bleak-based heart rate data acquisition method is adopted. The heart rate data acquisition module mainly includes a heart rate acquisition module and a Bluetooth module. The heart rate acquisition module can use professional heart rate acquisition equipment or rely on the real-time heart rate function of a suitable smart bracelet, such as the Huawei Smart Bracelet 6, which relies on HUAWEI TruSeen 4.0 heart rate monitoring technology to continuously, quickly, and accurately monitor the user's heart rate, ensuring the accuracy of experimental data. Through the bracelet's heart rate broadcast function, after the host computer determines the bracelet's device address, it can listen to the Bluetooth device in real time, enabling real-time reading of the bracelet's heart rate information.

[0076] In this embodiment, the heart rate data of the trainees is converted into RR intervals, and the characteristic parameters of heart rate are calculated using RR intervals: the mean heart rate (MEAN) and the root mean square (RMSSD) difference between adjacent RR intervals.

[0077] First, heart rate data is acquired through Huawei smart bands worn by trainees, with one data point obtained every second. The relative RR interval is calculated using the heart rate data. HRV characteristic parameters are analyzed based on the relative RR interval to obtain the root mean square (RMSSD) difference between the mean heart rate (MEAN) and adjacent RR intervals. Finally, the adaptive evaluation results of the heart rate characteristic parameters are obtained through an adaptive equation.

[0078] In this embodiment, the heart rate variability analysis employs a time-domain index analysis method. Time-domain analysis involves statistically analyzing the RR intervals between adjacent heartbeats, extracting characteristic parameters from the collected heart rate data, and normalizing the calculated parameter results to measure the activity level of the autonomic nervous system. The parameters selected in this invention are the mean heart rate (MEAN) and the root mean square (RMSSD) of the difference between adjacent RR intervals.

[0079] Mean heart rate (MEAN) aims to reflect the average level of the RR interval. Generally, a person's average heart rate can reflect the body's adaptability to the environment and various stresses. Under normal circumstances, well-adapted individuals usually have a relatively stable and appropriate average heart rate level. When faced with stress, exercise, or other stressful situations, the body will adjust its heart rate to adapt to changes in the external environment and maintain internal homeostasis. Therefore, by monitoring changes in average heart rate, one can gain a preliminary understanding of an individual's adaptability and physical condition. The formula for calculating mean heart rate (MEAN) is:

[0080] ;

[0081] The root mean square (RMSSD) of the difference between adjacent heart rate intervals reflects the fast-changing component of heart rate velocity (HRV), as well as the activity of the autonomic nervous system and the balance between the sympathetic and vagus nerves. A higher RMSSD value is generally considered a sign of good autonomic nervous system function and high adaptability. Conversely, a lower RMSSD value may indicate autonomic nervous system dysfunction or interference from adverse factors, resulting in poor adaptability. The formula for calculating the root mean square (RMSSD) of the heart rate difference is as follows:

[0082] ;

[0083] in, Indicates the first One RR interval value; Indicates the first RR interval values, This represents the total number of RR intervals in the statistics.

[0084] In this embodiment, the method for extracting center-of-gravity data from real-time collected video data of trainees is as follows:

[0085] Step a1: Use the MultiPose human pose recognition algorithm to identify and calculate the coordinates of key human points from the visual data of the trainee's rescue operation.

[0086] The Multipose human pose recognition method can calculate the two-dimensional position information of 17 human body key points, such as... Figure 3As shown in Figure b, the human body's center of gravity position information is solved based on the key point information of the human body extracted by this method. It is suitable for graphical analysis. Therefore, the torque synthesis method is selected in this invention to calculate the center of gravity position of the trainee. There is no need to build a separate model for each trainee, which can effectively improve efficiency. Combined with the multiplier method, the center of gravity position of the trainee can be determined more accurately.

[0087] Step a2: Based on the human body key point coordinates obtained in step a1, supplement the coordinates of the head, hands, heels, midpoints of the two shoulders, and midpoints of the two hips, resulting in the coordinates of 19 human body key points required for calculating the position of the human body's center of gravity using the torque synthesis method. Figure 3 As shown in Figure a;

[0088] The torque composition method, based on a joint model of the human body, divides the body into multiple segments, each considered a rigid body. This means that during movement or under stress, the shape, size, and relative positions of internal points within each segment remain unchanged. Different human joint models may vary slightly. According to the torque theorem, the resultant torque at the body's center of gravity is equal to the algebraic sum of the torques of the component forces in each segment, as shown in the following formula:

[0089] ;

[0090] ;

[0091] in, It is the weight of the human body. These are the X and Y direction values ​​of the body's center of gravity coordinates, respectively. These are the coordinates of the center of gravity of various parts of the body. It is the weight of the link.

[0092] The coordinates of the body's center of gravity can be obtained using the following formula:

[0093] ;

[0094] ;

[0095] To calculate the center of gravity coordinates of a human body using the formula for center of gravity coordinates, the center of gravity coordinates of each part of the body are required. The center of gravity coordinates of each part of the body can be obtained by solving the coordinates of the center point of each part of the body through the two joints, and then the center of gravity coordinates can be calculated based on the coordinates of the center point.

[0096] The multiplier method for determining the coordinates of the human body's center of gravity reduces the process of finding the center point coordinates of two joints in each human body segment, enabling the rapid acquisition of the body's center of gravity coordinates. Based on the principle of the multiplier method, and combined with the coordinates of 17 human key points obtained from the Multipose human pose estimation used in this paper, the commonly used human joint inertia parameters are used as the relative weight coefficients in the multiplier method in this embodiment, as shown in Table 1.

[0097]

[0098] like Figure 3 Figure b shows the coordinates of 17 human keypoints output by the Multipose human pose recognition algorithm. This does not perfectly match the 19 human keypoints required for calculating the center of gravity coordinates in the multiplier model used in the moment synthesis method shown in Figure a. Specifically, Multipose outputs 5 more human keypoints than the multiplier model: 1 nose, 2 eyes, and 2 ears. It also lacks the location information for 7 human keypoints: 1 head keypoint, 2 hand keypoints, 2 heel keypoints, 1 midpoint between the shoulders, and 1 midpoint between the hips. Therefore, preprocessing and supplementing of the missing and redundant human keypoints are necessary. The specific method is as follows:

[0099] For head center of gravity coordinate missing point compensation, Multipose outputs human keypoints that do not include head keypoints, but will output five human keypoints for the nose, eyes, and ears. According to biological analysis, the position of the ears actually accounts for a very small percentage of head mass, while the nose and eyes are located near the central area of ​​the head and occupy a larger area. In actual data acquisition, it was found that the coordinates of the nose are basically in the center of the head. Therefore, the coordinates of the nose are selected as the coordinates of the head center of gravity, and the coordinates of the ears and eyes are discarded.

[0100] Compensation for missing coordinates of the hand's center of gravity and heel: The compensation methods for these two key human body points are similar, so they will be explained together. The center of gravity of the hand is obtained by extending the distance from the elbow joint to the wrist joint. Let the elbow joint coordinate be... The wrist joint coordinates are The specific calculation method for the coordinates of the center of gravity of the palm is shown in the following formula;

[0101] ;

[0102] ;

[0103] in, This indicates the ratio of hand position. According to human biological analysis, the length of the hand is approximately 1 / 4 of the forearm length, and the center of gravity of the hand is biased towards the palm. Take 1 / 9;

[0104] According to human biological analysis, the distance from the heel to the ankle is approximately 1 / 4 of the calf length, and the weight coefficient of the ankle is relatively small. The impact of the compensation difference on the overall center of gravity is not significant, so the calculation method and the hand center of gravity coordinate preprocessing filling can be consistent.

[0105] For compensation of missing midpoints of the two shoulders and the two hips, since both coordinates are the midpoints of the connecting lines, the processing procedure is the same. Let the coordinate of the left shoulder joint be... The coordinates of the right shoulder joint are The coordinates of the midpoint of the line connecting the two shoulders are shown in the following formula;

[0106] ;

[0107] ;

[0108] Similarly, the midpoints of the two hips can be calculated using the known coordinates of the two hip joints.

[0109] Using the coordinates of 17 key points output by the Multipose model, the coordinates of 19 human body key points required for calculating the center of gravity using the moment synthesis method are obtained after the above preprocessing process. The coordinates of the human body center of gravity are then calculated with reference to the relative weight coefficients in Table 1.

[0110] Step a3: Use the moment synthesis method to calculate the coordinates of the trainee's center of gravity in each frame of the image. The specific formula is:

[0111] ;

[0112] ;

[0113] in, Indicates the first The relative weight coefficients corresponding to key points of an individual's body. , Corresponding to the first The horizontal and vertical coordinates of key points on an individual's body.

[0114] The key points of the human body in the above formula correspond to the human body segments in Table 1, the coefficient table of the multiplier method.

[0115] The above method is used to obtain the coordinates of the trainee's center of gravity during the rescue operation from video frames of the rescue operation video data, thus obtaining a series of center of gravity coordinates with a time sequence. In this embodiment, center of gravity characteristic parameters are obtained by analyzing the center of gravity data, which are used to evaluate the trainee's human balance ability. The specific steps are as follows:

[0116] Step b1: Calculate the length of the trainee's center of gravity movement trajectory in the visual data of the rescue operation. The specific formula is:

[0117] ;

[0118] in, These represent the coordinates of the human body's center of gravity. This indicates that the index number of the human body's center of gravity position is extracted from the visual data. This represents the total number of coordinates of the human body's center of gravity that were collected.

[0119] The length of the center of gravity movement trajectory, Lng, represents the length of the trajectory of the body's center of gravity coordinate movement during adaptive training. Generally, the larger the length of the center of gravity movement trajectory, the greater the amplitude of body swaying during actual training, the further the center of gravity deviates from its own center position, and the poorer the trainee's balance stability.

[0120] Step b2: Calculate the area of ​​the trainee's center of gravity coordinate trajectory in the visual data of the rescue operation. This refers to the area of ​​the closed region enclosed by the center of gravity coordinate trajectory. The specific formula is:

[0121] ;

[0122] in, This represents the greater distance between the coordinates of two adjacent centroid positions and the coordinates of the average centroid position of the human body. This calculation assumes that the average centroid position of the human body is at the origin of the coordinate system.

[0123] The trajectory area (Area) refers to the enclosed region bounded by the trajectory lines of the trainee's center of gravity coordinate changes during adaptive training. It reflects the degree of body sway; generally, increased body sway during training leads to an increase in the trajectory area, which can assess the trainee's balance and stability performance during the test. In actual calculations, assuming the average center of gravity is at the origin, the area enclosed by the coordinates of the center of gravity obtained from two adjacent video frames and the coordinates of the average center of gravity is approximately a sector (distance R between the more distant points). These areas are then summed to obtain the trajectory area.

[0124] Step b3: Calculate the length of the trainee's center of gravity trajectory per unit area in the visual data of the rescue operation. This refers to the ratio of the length of the trajectory of the center of gravity movement to the area of ​​the trajectory. The specific formula is as follows:

[0125] ;

[0126] The length of the trajectory per unit area (Unit) reflects the trainee's minute control over their posture while maintaining balance. This ratio decreases as the trainee's control improves.

[0127] Step b4: Calculate the variance of the coordinates of the center of gravity of all trainees in the visual data of the rescue operation. The specific formula is as follows:

[0128] ;

[0129] in, The average displacement along the x-axis represents the center-of-gravity coordinates of all human bodies. This represents the average displacement along the y-axis of the center of gravity of all human bodies.

[0130] The center of gravity displacement variance is used to assess a person's balance ability by calculating the variance of all collected center of gravity coordinates during the current training. The larger this value, the worse the balance ability.

[0131] The displacement variance evaluation standard of this invention is formulated based on reference materials and actual measurement values ​​of this invention, as shown in Table 2.

[0132]

[0133] In this embodiment, the evaluation method for the adaptability of aviation medical emergency rescue uses MEAN and RMSSD of heart rate data and the unit area trajectory length and center of gravity displacement variance of center of gravity data as characteristic parameters for evaluation.

[0134] First, construct the adaptive evaluation equations for each characteristic parameter. The specific method is as follows:

[0135] Step c1: Using the mean heart rate (MEAN), the root mean square (RMSSD) of the difference between adjacent RR intervals of heart rate, the length of the centroid trajectory per unit area, and the variance of the centroid coordinate displacement as independent variables, and the adaptive evaluation result as the dependent variable, an adaptive evaluation equation is constructed using linear regression as follows:

[0136] ;

[0137] in, This indicates the different independent variables selected. This indicates the results of the adaptation evaluation. Represents the regression coefficient. Represents the constant term. Indicates the index number of the independent variable.

[0138] Step c2: Collect multiple sets of training data from the trainees, use a predetermined adaptive evaluation table to obtain the trainees' self-evaluation and manual evaluation results, and obtain the coefficients and constant terms of each adaptive evaluation equation by fitting the training data and evaluation results.

[0139] Table 3 shows the predetermined adaptability evaluation table in this embodiment, which divides the adaptability effect evaluation into several scoring intervals for evaluation.

[0140]

[0141] Following the experimental method described below, data from multiple training sessions were collected, including real-time heart rate and visual data collected during simulated rescue operations. Pre-assessments for each training session were provided using both participant self-evaluation and post-training human evaluation. These training data and pre-assessment results were imported into SPSS software, and linear regression was selected. The regression model coefficients are shown in Table 4. In Table 4, models 1, 2, 3, and 4 are linear regression models with mean heart rate (MEAN), RMSSD, center of gravity displacement variance, and trajectory length per unit area as independent variables, respectively. The dependent variable is the adaptive evaluation corresponding to each independent variable. The significance of these independent variables is less than 0.05, indicating statistical significance.

[0142]

[0143] The regression equation obtained from Table 4 is as follows:

[0144] ;

[0145] ;

[0146] ;

[0147] ;

[0148] in, An adaptive evaluation equation representing the mean heart rate (MEAN); The evaluation equation representing the root mean square slowdown (RMSSD) of the difference between adjacent RR intervals of heart rate; The evaluation equation representing the variance of the center of gravity displacement; An evaluation equation representing the length of the centroid trajectory per unit area; MEAN represents the average heart rate. The root mean square (RMSSD) represents the difference between adjacent RR intervals of heart rate. Indicates the variance of the center of gravity displacement; This represents the length of the centroid trajectory per unit area. When the calculated evaluation results of each characteristic parameter are greater than the maximum value in the predetermined adaptability evaluation table, the maximum value is used; when the calculated evaluation results of each characteristic parameter are less than the minimum value in the predetermined adaptability evaluation table, the minimum value is used. For example, when... >5 o'clock, order ,when When <2, let ; , and Similarly.

[0149] The final evaluation result is obtained by averaging the scores of mean heart rate (MEAN), RMSSD, trajectory length per unit area, and center of gravity displacement variance. The specific evaluation equation for the trainee's adaptability is as follows:

[0150] ;

[0151] This regression equation, based on heart rate and center of gravity data, can be used to evaluate the training effectiveness of trainees' aviation medical emergency rescue adaptability training, helping trainees to better understand and master their training situation.

[0152] Example 2:

[0153] In this embodiment, an experiment is conducted according to the specific implementation plan of Embodiment 1, providing experimental data and result analysis. The purpose of the experiment is to use the heart rate and center of gravity characteristic parameters designed in this invention to conduct an overall adaptive assessment of the trainees' performance in simulated training, obtain a large amount of sample data, and obtain an adaptive equation through statistical fitting, while verifying the feasibility and reliability of the adaptive equation assessment results.

[0154] The experimental environment is an aviation simulator or a similar testing environment, using an aviation medical emergency rescue simulation platform to simulate the actual flight environment. Adaptive training experiments are conducted on healthy participants without any diseases affecting balance or mental disorders.

[0155] In this embodiment, real-time visual data acquisition utilizes an industrial-grade camera based on an RGB camera, with a maximum resolution of 2K. The hardware specifications include an industrial-grade high-definition 1-megapixel camera connected to the host via a USB interface. Heart rate data acquisition uses a Huawei Smart Band 6, based on HUAWEI TruSeen 4.0 heart rate monitoring technology, capable of continuously, quickly, and accurately monitoring the user's heart rate. Real-time acquisition of video and heart rate data of the participant performing CPR compressions is also performed.

[0156] During the acclimatization test, participants undergo training in a specific sequence. First, trainers need to relax, maintain a calm mood, and keep their bodies stable for 2 minutes; this is the static state, during which heart rate data is collected. It is important to avoid any disruptive behavior during this time, such as giving instructions or asking questions. If any platform or data collection errors occur during the training, the session must be ended and restarted. Before the second training session, participants need to rest for 5 minutes to ensure a stable state and avoid any influence from the previous training, thus ensuring the usability of the collected data.

[0157] After each training session, participants evaluated their own level of adaptability using the adaptability evaluation form in Table 3, obtaining self-evaluation results. At the same time, a pre-assessment of the current training session was given to the participants using manual evaluation after each training session.

[0158] The experimental data collected in real time in this experiment were processed using SPSS 26 software (Statistical Product and Service Solutions, a statistical software package for social sciences). The results are expressed as mean ± standard deviation. One-way ANOVA was used to compare data among multiple groups, and independent samples t-tests were used for comparisons between groups. Generally, a p-value < 0.05 is considered statistically significant.

[0159] When analyzing heart rate data, real-time heart rate data of trainees in static (before training), during the first simulated training (first training), and during simulated training after multiple training sessions are collected in real time. The heart rate RR interval is calculated, and then time-domain heart rate specificity analysis is performed to obtain the root mean square (RMSSD) of the difference between the mean heart rate (MEAN) and adjacent RR intervals in each state.

[0160] Figure 5 The graph shows a comparison of the average heart rate (MEAN) of trainees under the three training conditions described above. It can be seen from the graph that the average heart rate of trainees during their initial training was significantly higher than the resting heart rate. After multiple training sessions, the average heart rate decreased, indicating that trainees' adaptability to the training environment was not as good as their static adaptability. However, after repeated training, the trainees' adaptability to the aviation simulation environment improved.

[0161] Figure 6 This chart compares the RMSSD of the participants' heart rate under the three training conditions described above. The participants' RMSSD during the initial training was significantly higher than their resting RMSSD, but after multiple training sessions, the participants' RMSSD decreased. This differs from the normal physiological phenomenon where a person's stronger adaptability results in a higher RMSSD value, for two reasons:

[0162] (1) The adaptability in the RMSSD physiological index refers to the adaptability of human physiological capabilities, not the adaptability to the air rescue environment improved through multiple training sessions in this embodiment.

[0163] (2) Since the RR interval of a normal human body will vary each time, the heart rate data in this embodiment is the heart rate per minute after being collected and processed by the smart bracelet. The measurement accuracy is limited. In this embodiment, the square root of the mean square of the difference in the statistical heart rate data is used to reflect the fluctuation of the heart rate curve. It can also represent the human body's vagal nerve tension level and breathing rhythm reflected by the RMSSD in the HRV index mentioned above. Therefore, when the trainee's breathing rhythm is abnormal and the heart rate accelerates when he is not adapted to the air rescue environment, the square root of the mean square of the difference in the heart rate RR interval will rise, which is consistent with the collected results.

[0164] Based on the changes in average heart rate, we can conclude that repeated training can reduce the training load of trainees in the flight simulation training environment, help trainees alleviate adverse psychological conditions such as tension and fright, and also indicate that trainees' adaptability to the flight simulation environment has improved after repeated training. The experimental results are consistent with expectations.

[0165] The mean heart rate (MEAN) and RMSSD of multiple test groups were statistically analyzed. SPSS software calculation showed a significant difference (P < 0.05), indicating that the differences in the experimental data reflected the true differences between groups.

[0166] When analyzing the center of gravity data, training videos of the participants were first collected. Multipose was used to estimate the human posture. The center of gravity coordinates of the participants in the videos were extracted using the torque synthesis method mentioned earlier. Then, the trajectory length per unit area and the variance of the center of gravity shift were calculated based on the center of gravity coordinate data. The data for each test index are shown in Table 5. A center of gravity trajectory diagram was then plotted based on the calculation results. Figure 4 The image shows the center of gravity trajectory after a single test. The center of gravity trajectory after the test can help determine the training effect.

[0167]

[0168] The results in Table 5 show that the length of the center of gravity trajectory after multiple training sessions is shorter than that after the initial training. This indicates that the deviation of the center of gravity from its normal position decreases and stability improves after repeated training. The area of ​​the center of gravity trajectory after multiple training sessions is also smaller than that after the initial training, suggesting a decrease in swaying and improved balance control. The length of the trajectory per unit area after multiple training sessions is also shorter than that after the initial training, further demonstrating that repeated training improves balance. Similarly, the variance of the center of gravity displacement is also lower after multiple training sessions than after the initial training, indicating that repeated training enhances the trainees' adaptability to balance in dynamic simulation environments. Therefore, it can be concluded that repeated training can help trainees improve their adaptability to balance in aviation simulation environments, and the experimental results are consistent with expectations.

[0169] Example 3:

[0170] This embodiment provides an adaptive evaluation system for aviation medical rescue simulation training, including a real-time data acquisition unit, a center of gravity data extraction unit, and an adaptive evaluation unit.

[0171] The real-time data acquisition unit is used to collect visual data of the trainees' rescue operations and heart rate data in real time.

[0172] The center of gravity data extraction unit is used to extract the center of gravity data of trainees from the visual data of trainees' rescue operations.

[0173] The adaptability assessment unit is used to analyze the heart rate and center of gravity data of trainees and output the adaptability assessment results of trainees.

[0174] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for evaluating the adaptability of aviation medical rescue simulation training, characterized in that, include: Real-time collection of visual data on the rescue operation of trainees and heart rate data of trainees; Extract the trainees' center of gravity data from the visual data of the trainees' rescue operations; The heart rate data of the trainees were analyzed to obtain the heart rate characteristic parameters of the trainees; By analyzing the center of gravity data of the trainees, characteristic parameters of their human balance ability can be obtained. Based on the heart rate characteristic parameters and the human balance ability characteristic parameters, the trainee's adaptability evaluation results are output. The characteristic parameters of human balance ability include the length of the center of gravity trajectory per unit area; The length of the center of gravity trajectory per unit area refers to the ratio of the length of the trajectory of the human body's center of gravity to the area of ​​the closed region enclosed by the trajectory of the human body's center of gravity.

2. The method according to claim 1, characterized in that, The heart rate data of the participants were converted into RR intervals, and the characteristic parameters of the heart rate were calculated using the RR intervals: mean heart rate (MEAN) and root mean square (RMSSD) of the difference between adjacent RR intervals. The formula for calculating the mean heart rate (MEAN) includes: ； The formula for calculating the root mean square (RMSSD) difference between adjacent RR intervals includes: ； in, Indicates the first One RR interval value; Indicates the first RR interval values, This represents the total number of RR intervals in the statistics.

3. The method according to claim 1, characterized in that, The method for extracting the centroid data includes: Step a1: Use the MultiPose human pose recognition algorithm to identify and calculate the coordinates of key human points from the visual data of the trainee's rescue operation; Step a2: Based on the coordinates of the human body key points, supplement the coordinates of the head, hands, heels, midpoints of the two shoulders, and midpoints of the two hips to obtain the coordinates of the 19 human body key points required for calculating the coordinates of the center of gravity of the human body using the torque synthesis method. Step a3: Use the moment synthesis method to calculate the coordinates of the trainee's center of gravity in each frame of the image. The specific formulas include: ； ； in, Indicates the first The relative weight coefficient corresponding to key points of an individual's body. , Corresponding to the first The horizontal and vertical coordinates of key points on an individual's body.

4. The method according to claim 3, characterized in that, The hand coordinates are obtained by extending the elbow and wrist coordinates, and the specific formula includes: ； ； in, The x-axis value represents the coordinates of the elbow joint; The y-axis value represents the coordinates of the elbow joint; The x-axis value represents the wrist joint coordinates; The y-axis value represents the wrist joint coordinates; This indicates the ratio of the position of the palm.

5. The method according to claim 3, characterized in that, The midpoint between the two shoulders is the coordinate of the midpoint of the line connecting the two shoulders. The specific formula includes: ， ； in, The x-axis value represents the coordinates of the left shoulder joint; The y-axis value represents the coordinates of the left shoulder joint; The x-axis value represents the coordinates of the right shoulder joint; The y-axis value represents the coordinates of the right shoulder joint.

6. The method according to claim 1, characterized in that, Methods for calculating the characteristic parameters of a trainee's human balance ability include: Step b1: Calculate the length of the trainee's center of gravity movement trajectory in the visual data of the rescue operation. The specific formula includes: ； in, These represent the coordinates of the human body's center of gravity. This indicates the extraction of the index number of the human body's center of gravity from visual data. This represents the total number of coordinates of the human body's center of gravity collected. Step b2, calculate the area of ​​the trainee's center of gravity coordinate trajectory in the visual data of the rescue operation, which refers to the area of ​​the closed region enclosed by the center of gravity coordinate trajectory. The specific formula includes: ； in, This represents the greater distance between the coordinates of the centroid position and the coordinates of the average centroid position of the human body in two adjacent frames of visual data. This calculation assumes that the average centroid position of the human body is at the origin of the coordinate system. Step b3, calculate the length of the trainee's center of gravity trajectory per unit area in the visual data of the rescue operation, which refers to the ratio of the length of the trajectory of the center of gravity position movement to the area of ​​the trajectory. The specific formula includes: ； Step b4: Calculate the variance of the center-of-gravity coordinates of all trainees in the visual data of the rescue operation. The specific formula includes: ； in, The average displacement along the x-axis represents the center-of-gravity coordinates of all human bodies. This represents the average displacement along the y-axis of the center of gravity of all people.

7. The method according to claim 6, characterized in that, The equations for adaptive evaluation based on the participants' heart rate and balance ability characteristics include: ； ； ； ； ； in, An adaptive evaluation equation representing the mean heart rate (MEAN); An adaptive evaluation equation representing the root mean square slowdown (RMSSD) of the difference between adjacent RR intervals of heart rate; An adaptive evaluation equation representing the variance of the center of gravity displacement; An adaptive evaluation equation representing the length of the centroid trajectory per unit area; MEAN represents the average heart rate. The root mean square (RMSSD) represents the difference between adjacent RR intervals of heart rate. Indicates the variance of the center of gravity displacement; This represents the length of the centroid trajectory per unit area. When the calculated adaptive evaluation results are greater than the maximum value in the predetermined adaptive evaluation table, the maximum value is taken; when the calculated adaptive evaluation results are less than the minimum value in the predetermined adaptive evaluation table, the minimum value is taken.

8. The method according to claim 7, characterized in that, The method for constructing the fitness evaluation equations for each feature parameter is as follows: Step c1: Using the mean heart rate (MEAN), the root mean square (RMSSD) of the difference between adjacent RR intervals of heart rate, the length of the centroid trajectory per unit area, and the variance of the centroid coordinate displacement as independent variables, and the adaptive evaluation result as the dependent variable, an adaptive evaluation equation is constructed using linear regression. ； in, This indicates the different independent variables selected. This indicates the results of the adaptation evaluation. Represents the regression coefficient. Represents the constant term. Indicates the index number of the independent variable; Step c2: Collect multiple sets of training data from the trainees, use a predetermined adaptive evaluation table to obtain the trainees' self-evaluation and manual evaluation results, and obtain the coefficients and constant terms of each adaptive evaluation equation by fitting the training data and evaluation results.

9. An adaptive evaluation system for aviation medical rescue simulation training using any one of the methods described in claims 1-8, characterized in that, include: The real-time data acquisition unit is used to collect visual data of the trainees' rescue operations and heart rate data in real time. The center of gravity data extraction unit is used to extract the center of gravity data of trainees from the visual data of trainees' rescue operations. The heart rate characteristic parameter analysis unit is used to analyze the heart rate data of trainees and obtain their heart rate characteristic parameters. The human balance ability characteristic parameter analysis unit is used to analyze the center of gravity data of trainees to obtain the human balance ability characteristic parameters of trainees; An adaptive evaluation unit is used to combine the heart rate characteristic parameters and the human balance ability characteristic parameters to output the adaptive evaluation results of the trainees; The characteristic parameters of human balance ability include the length of the center of gravity trajectory per unit area; The length of the center of gravity trajectory per unit area refers to the ratio of the length of the trajectory of the human body's center of gravity to the area of ​​the closed region enclosed by the trajectory of the human body's center of gravity.

Images