A tactile perception training method for cardiopulmonary resuscitation first aid

Abstract

The present invention provides a tactile perception training method for cardiopulmonary resuscitation first aid, which pads an extremely thin skin-like sensor on the chest and determines the pressing situation by detecting the change in the size of the capacitor when pressing. The pressing force, pressing frequency and pressing position are compared with the standard pressing to intuitively reflect the distance between the current pressing and the standard pressing level, and display the areas that need improvement through an interactive interface. The present invention further obtains the pressing force, frequency and position information by analyzing the pressing data collected by the sensor, so that the rescuer can intuitively feel the gap between his pressing level and the standard pressing level when receiving training.

Description

Technical Field

[0001] The invention relates to the field of cardiopulmonary resuscitation training, and in particular to a tactile perception training method for cardiopulmonary resuscitation first aid. Background Art

[0002] Cardiopulmonary resuscitation is a technique used to maintain vital signs in an emergency when the victim has stopped breathing and the heart has stopped beating. It can provide the important support the human body needs during the rescue time and give the victim a chance to receive further medical treatment. Cardiopulmonary resuscitation is a very important first aid technique that can save lives in an emergency. This technique is usually performed by medical personnel, but other people can also perform cardiopulmonary resuscitation after receiving proper training.

[0003] When performing cardiopulmonary resuscitation, it is very important to have moderate force. If the force is too great, it may cause sternal injury or internal organ damage. If the force is too small, there may not be enough oxygen entering the heart and lungs, thereby reducing the success rate of resuscitation. At the same time, the frequency of cardiopulmonary resuscitation compression is also particular. During cardiopulmonary resuscitation, the heart obtains blood supply through chest compression. If the compression frequency is too low, the blood supply received by the heart will be reduced, which will cause cardiac ischemia. If the compression frequency is too high, the heart tissue will be subjected to excessive pressure, which may cause damage to the heart tissue. Therefore, it is very necessary for every rescuer, especially students and doctors who are receiving cardiopulmonary resuscitation training, to master the force and frequency of cardiopulmonary resuscitation. At this stage, the international index for cardiopulmonary resuscitation compression is a compression depth greater than 5 cm. This index is too superficial, which makes it impossible for rescuers to have an effective understanding of their compression level. For the accuracy of cardiopulmonary resuscitation implementation, beginners and even doctors who have studied many times still do not have a very good judgment. Since the force, frequency and even location of compressions during CPR are not directly visible, it is a problem that needs to be designed to efficiently and intuitively display this information and evaluate the accuracy of each CPR implementation.

[0004] Application number: CN202211049681.1, patent name: A cardiopulmonary resuscitation assessment and training device and method of use, including a main screen, a sub-screen, an identity scanner, a host and a simulator, the simulator is used for receiving compression and blowing in actual operation, simulating pulse beats, and simulating pupil simulation functions. The present invention combines the existing on-site first aid standards to establish an effective teaching and training system; combined with a virtual simulation simulator, real-time voice feedback on the pressing position and strength; combined with the pressing rhythm indication and the pressing rhythm music prompts, the user can quickly master the accurate cardiopulmonary resuscitation pressing rhythm; combined with the data curve chart and the pressing index, the pressing-related data is displayed to help the user correctly understand the training situation in real time; various incentive systems such as scoring and rankings are provided, which effectively increases the chances of users reusing this product.

[0005] It combines the existing first aid standards to establish an effective teaching and training system. Based on the traditional CPR standards, this article re-sets more concrete indicators on compression force, position and frequency based on the data collected by the sensor. By generating compression curves and heat maps, trainees can intuitively recognize the gap between their compression level and the standard compression, which can avoid the original CPR indicators being too superficial, which makes rescuers unable to have an effective understanding of their own compression level.

[0006] Application number: RU2016108154, patent name: cardiopulmonary resuscitation quality feedback system, automatic resuscitation device, machine-readable information storage medium. The system includes a measuring unit, a processor, an indicator unit, a sensor for recording the depth of chest compressions, and a display for displaying a signal indicating the depth of compressions. An automatic resuscitation device, including a chest compression device, a system for providing feedback about chest wall compressions, and a processor. While performing cardiopulmonary resuscitation, the patient's arterial blood pressure is obtained over a period of time. A blood pressure cardiopulmonary resuscitation quality indicator is calculated. The cardiopulmonary resuscitation quality indicator is compared with a quality standard determined as a threshold value of a target interval. If the blood pressure cardiopulmonary resuscitation quality indicator is outside the quality standard, or if the blood pressure cardiopulmonary resuscitation quality indicator meets the quality standard, a low quality indication signal is sent or a high quality indication signal is sent. The compression depth of cardiopulmonary resuscitation is recorded and a signal indicating the compression depth is displayed. The need to perform a boost and a de-boost of the compression depth relative to a previously determined optimal compression depth is indicated. A new optimal compression depth is selected based on the three CPR quality indicators obtained as the depth with the highest blood pressure CPR quality indicator value or as the minimum depth with the blood pressure CPR quality indicator exceeding the CPR quality indicator value. A target blood pressure CPR quality indicator value. Effect: By using the blood pressure characteristic as a CPR quality indicator, the present invention group allows optimal CPR of the patient.

[0007] It uses a sensor for recording the depth of chest compressions and a display for displaying a signal indicating the depth of compressions. While performing cardiopulmonary resuscitation, the patient's arterial blood pressure is obtained over a period of time. The blood pressure cardiopulmonary resuscitation quality index is calculated, and the cardiopulmonary resuscitation quality indicator is compared with the quality standard of the alarm value determined as the target interval. This article uses the method of evaluating the compression pressure, frequency and position of cardiopulmonary resuscitation, which can protect patients from sternal injury or internal organ injury due to excessive force during cardiopulmonary resuscitation, or insufficient oxygen entering the heart and lungs due to insufficient force, thereby reducing the success rate of resuscitation. Summary of the invention

[0008] In order to solve the above problems, the present invention proposes a tactile perception training method for cardiopulmonary resuscitation first aid, which pads a very thin skin-like sensor on the chest and detects the change in capacitance when pressing to determine the pressing situation. The pressing force, pressing frequency and pressing position are compared with the standard pressing to intuitively reflect the distance between the current pressing and the standard pressing level, and the areas that need improvement are displayed through an interactive interface.

[0009] To achieve the above object, the technical solution adopted by the present invention is:

[0010] A tactile perception training method for cardiopulmonary resuscitation first aid, characterized by comprising the following steps:

[0011] Step 1: Design a skin-like mold to fit the object used for cardiopulmonary resuscitation;

[0012] Step 2: Make a sensor to detect pressure, frequency and speed information;

[0013] Step 3. Set the activation function to obtain the pressing information, so that the pressing force, frequency and position information can be displayed on Arduino in real time, and the data of each point, including the unactivated points, can be output to Excel or directly to the system for the next step of data analysis.

[0014] The activation function is as follows:

[0015] An activation function is set for the output data of each point according to the pressing position and pressing force. When the pressing pressure value received by the point is higher than a certain value, it can be activated. Only the activated point can output the pressure value, and the output values ​​of the other points are all 0;

[0016] Step 4: record the standard cardiopulmonary resuscitation compression data of multiple doctors, process the data appropriately and use it as a reference and comparison standard, set the threshold range of the standard value, and judge the rationality of cardiopulmonary resuscitation by designing a certain algorithm as a reference and indicator of standard compression;

[0017] The data processing process described in step 4 is as follows:

[0018] For the 120 grid points obtained by 10*12, each point outputs a corresponding pressure value. The pressure value is 0 when it is not activated or pressed, and it is the pressure value of pressing when activated. The pressure value of the 120 dot matrix will change with time. After the 120-dimensional data is input into Excel or matlab, the data composed of the 120 dot matrix is ​​reduced in dimension by the PCA algorithm. The principal component after dimension reduction and time t are fitted into a function image that changes with time to determine whether the pressing force and frequency are qualified. At the same time, for the pressing position, the correctness of the pressing position is determined by setting a dynamic heat map, and the pressing process can be intuitively displayed;

[0019] The specific steps of dimensionality reduction are as follows:

[0020] Taking 10*12 grid data as an example, we get 120-dimensional index data. The samples in each row are the pressure data of 120 points at each sampling time, which can be expressed in matrix form as follows:

[0021] The 120-dimensional data needs to be reduced in dimensionality. The points that have never been activated need to be deleted directly before dimensionality reduction, and the remaining points need to be reduced in dimensionality. In one press, the points that are activated are often only in a single digit. Each indicator, that is, each point, has the same attribute and the same unit and the value is not much different. Therefore, after dimensionality reduction, there is only one principal component left, that is, find the following linear transformation:

[0022] W'=a0W (0,0) +a1W (0,1) +...+a 119 W (9,11) , where the coefficients before the unactivated points are 0. Use it as the final indicator for curve fitting;

[0023] Step 5. According to the algorithm in step 4, the pressing data of the trainee is processed, and a certain algorithm is designed to determine the pressing force, position, frequency and other values ​​to determine the defects of the unqualified trainees, and whether the trainee is qualified and the areas that need improvement are displayed in the corresponding user interface.

[0024] As a further improvement of the present invention, the mold described in step 1 needs to be designed on a dummy, the mold needs to cover the required pressing position, and be appropriately extended to determine whether the pressing position is accurate or not, and the mold should match the design of the sensor in step 2.

[0025] As a further improvement of the present invention, the sensor in step 2 is as follows:

[0026] Requirements for the sensor: The sensor is formed by casting silicone on a mold. It should be as thin as possible and fit the human skin without leaking the wires.

[0027] As a further improvement of the present invention, the sensor in step 2 is as follows:

[0028] The sensor production steps are as follows: After applying a shallow layer of silicone on the mold, lay about 12 wires vertically, then lay about 10 wires horizontally, and finally lay another layer of silicone on the mold to form a skin-like sensor. When the two electrode wires cross, a capacitor is formed at the intersection. The size of the capacitor is changed by touching and pressing, thereby detecting the pressure and position information.

[0029] As a further improvement of the present invention, the algorithm described in step 5 is as follows:

[0030] After completing PCA dimensionality reduction, choose to use Fourier series to fit the compression curve, that is:

[0031]

[0032] By judging the size of the correlation coefficient R square obtained by fitting the compression data at the 8th order or below Fourier series, we can judge whether the changes in the compression fluctuations of the rescuer are generally qualified: if R square is less than 0.8, it means that the amplitude deviation during compression is too large, which is an unqualified compression; if R square is greater than 0.8, then the subsequent compression intensity and compression frequency are judged, and then by judging whether the difference between the absolute value of the maximum compression and the standard compression is within 7, that is, observing the difference in the peak value of the peak, and judging whether the compression frequency is within 100-120 per minute, that is, observing the number of fluctuations of the fitting curve. In the time t seconds of data collection, assuming that the number of peaks is n, the compression frequency f can be expressed as: That is, determine whether f is between the interval [1.67,2];

[0033] If any one of the items is not met, it is considered an unqualified press; finally, the thermal map is used to determine whether the press position is correct. If it is correct, it can be said that this is a qualified press that meets the indicators.

[0034] Beneficial effects:

[0035] (1) The present invention can reduce and alleviate the problem that some doctors, due to their lack of cardiopulmonary resuscitation skills, are unable to control the frequency and strength of compression when performing cardiopulmonary resuscitation on patients in emergency situations, causing some unnecessary pain or damage to patients.

[0036] (2) With respect to the standardized judgment of the conventional compression depth of about 5 cm during cardiopulmonary resuscitation, the present invention provides a more detailed method for quantifying information such as compression intensity and frequency. By generating a dynamic heat map and a compression fitting curve, rescue personnel can have more intuitive indicators to practice cardiopulmonary resuscitation, and further help patients relieve symptoms efficiently in emergency situations.

[0037] (3) The sensor designed by the present invention is made of silicone. Due to its extremely thin thickness and skin-like touch, compared with other sensors, the doctor will not feel any discomfort when pressing. At the same time, due to its extremely thin thickness, no additional force is required to overcome the resistance of the sensor, which will not cause each medical staff to face a big difference from the training when facing real rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is the design diagram of the sensor of the present invention;

[0039] Figure 2 This is a process diagram of the present invention for determining whether the rescuer's compression is qualified;

[0040] Figure 3 It is the overall flow chart of the present invention. DETAILED DESCRIPTION

[0041] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0042] The present invention discloses a compression visualization method and compression evaluation standard for cardiopulmonary resuscitation first aid training, which further obtains compression force, frequency and position information by analyzing the compression data collected by a sensor, so that rescuers can intuitively feel the gap between their compression level and the standard compression level when receiving training.

[0043] like Figure 3 As shown, the system of the present invention includes sensor production, standard data collection and comparison and discrimination algorithm. The designed sensor is used for data collection, and has the advantages of thin thickness and skin-like touch. After that, data collection begins, and unnecessary errors and disturbances are avoided by setting the activation function. The compression data of experienced rescuers are collected as a comparison index during cardiopulmonary resuscitation training, and the compression status of each trainee is intuitively compared with the standard compression. The comparison and discrimination algorithm in the present invention intuitively shows the shortcomings of the presser through the dimension reduction algorithm and curve fitting, as well as the generation of dynamic thermal maps.

[0044] The sensors used to collect data are composed of Figure 1Intuitive display, the electrode wires represented by thin lines are arranged crosswise as shown in the figure, thus forming grid points one by one, each point can generate capacitance, and the force, frequency and position of the pressing are obtained by analyzing the change in the capacitance. The front and back of the electrode wires are covered with silicone. Due to its extremely thin thickness and skin-like touch, rescuers do not need to exert extra force during training to overcome the resistance of the sensor, and the pressing tactile sensation during actual rescue will not be too different from the tactile sensation of this sensor.

[0045] For specific discrimination algorithms, such as Figure 2 As shown, the following steps are included:

[0046] (1) Analyze the rescuer's compression data, set the activation function, store the pressure data obtained by compression in Excel, and perform principal component analysis on the data in Excel. Take the 10*12 dimensional line laying method as an example:

[0047]

[0048] Among them, each point is used as an indicator to be reduced in dimension, the purpose is to find out:

[0049] W'=a0W (0,0) +a1W (0,1) +...+a 119 W (9,11)

[0050] Unlike the general PCA, the 120-dimensional indicators here are variables with the same unit and attribute. Therefore, after removing the samples that have never been activated (the value of each frame is 0), the principal component obtained by PCA on the data is basically only one, so the obtained data Perform Fourier fitting to generate a pressure curve that varies with time.

[0051] (2) For the obtained curve, first analyze the size of its correlation coefficient R square to judge whether the change of the rescuer's compression fluctuation is generally qualified. If R square is greater than 0.8, then judge the compression intensity and compression frequency. A peak of the fitting curve is a compression. If the compression frequency is within the range of 100 to 120 per minute, it means that the frequency is in compliance. That is, within the time t seconds of data collection, assuming that the number of peaks is n, the compression frequency f can be expressed as: Whether the pressing frequency is appropriate is determined by judging whether f is in the interval [1.67,2]. At the same time, if the difference between the maximum value of the peak of the pressing fitting curve and the absolute value of the peak of the standard value is within 7, it can be determined that the pressure is acceptable.

[0052] (3) At the same time, the pressing position is judged by generating a dynamic heat map of the pressure and pressing position changing over time based on the pressure data in Excel, so as to intuitively judge whether there is a deviation in the pressing position.

[0053] (4) If all the above indicators are met, it is considered a qualified press. Otherwise, there are certain problems with the press. The indicators with problems will be listed on the user interface to inform the trainer of the current deficiencies.

[0054] The above description is only a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent change made based on the technical essence of the present invention still falls within the scope of protection required by the present invention.

Claims

1. A tactile perception training method for cardiopulmonary resuscitation first aid, characterized in that: The following steps are involved: Step 1: Design a skin-like mold to fit the object used for cardiopulmonary resuscitation; Step 2: Make a sensor to detect pressure, frequency and speed information; Step 3. Set the activation function to obtain the pressing information, so that the pressing force, frequency and position information can be displayed on Arduino in real time, and the data of each point, including the unactivated points, can be output to Excel or directly to the system for the next step of data analysis. The activation function is as follows: An activation function is set for the output data of each point according to the pressing position and pressing force. When the pressing pressure value received by the point is higher than a certain value, it can be activated. Only the activated point can output the pressure value, and the output values ​​of the other points are all 0; Step 4: Record the standard CPR compression data of multiple doctors, process the data appropriately and use it as a reference and comparison standard, set the threshold range of the standard value, and design a certain algorithm to judge the rationality of CPR as a reference and indicator of standard compression; Step 4: Data processing is as follows: For the 120 grid points obtained by 10*12, each point outputs a corresponding pressure value. The pressure value is 0 when it is not activated or pressed, and it is the pressure value of pressing when activated. The pressure value of the 120 dot matrix will change with time. After the 120-dimensional data is input into Excel or matlab, the data composed of the 120 dot matrix is ​​reduced in dimension by the PCA algorithm. The principal component after dimension reduction and time t are fitted into a function image that changes with time to determine whether the pressing force and frequency are qualified. At the same time, for the pressing position, the correctness of the pressing position is determined by setting a dynamic heat map, and the pressing process can be intuitively displayed; The specific steps of dimensionality reduction are as follows: With 10*12 grid data, we get 120-dimensional index data. Each row of samples is the pressure data of 120 points at each sampling time, which can be expressed in matrix form as follows: The 120-dimensional data needs to be reduced in dimensionality. The points that have never been activated need to be deleted directly before dimensionality reduction, and the remaining points need to be reduced in dimensionality. In one press, the points that are activated are often only in a single digit. Each indicator, that is, each point, has the same attribute and the same unit and the value is not much different. Therefore, after dimensionality reduction, there is only one principal component left, that is, find the following linear transformation: W'=a0W (0,0) +a1W (0,1) +...+a 119 W (9,11) , where the coefficients before the unactivated points are 0, where W' = Use it as the final indicator for curve fitting; Step 5: According to the algorithm of step 4, the pressing data of the trainee is processed, and the pressing force, position, and frequency values ​​are determined to determine the defects of the unqualified trainees, and whether the trainees are qualified and the areas that need improvement are displayed on the corresponding user interface.

2. A tactile perception training method for cardiopulmonary resuscitation first aid according to claim 1, characterized in that: The mold described in step 1 needs to be designed on the dummy. The mold needs to cover the required pressing position and be appropriately extended to determine whether the pressing position is accurate or not. The mold should match the design of the sensor in step 2.

3. A tactile perception training method for cardiopulmonary resuscitation first aid according to claim 1, characterized in that: The sensors described in step 2 are as follows: Requirements for the sensor: The sensor is formed by casting silicone on a mold. It should be as thin as possible and fit the human skin without leaking the wires.

4. A tactile perception training method for cardiopulmonary resuscitation first aid according to claim 3, characterized in that: The sensors described in step 2 are as follows: The sensor production steps are as follows: After laying a layer of silicone on the mold, lay 12 wires vertically, then lay 10 wires horizontally, and finally lay another layer of silicone on the mold to form a skin-like sensor. When the two electrode wires cross, a capacitor is formed at the intersection. The size of the capacitor is changed by touching and pressing, thereby detecting the pressure and position information.

5. A tactile perception training method for cardiopulmonary resuscitation first aid according to claim 3, characterized in that: In step 5: After completing PCA dimensionality reduction, choose to use Fourier series to fit the compression curve, that is: The size of the correlation coefficient R-square obtained by fitting the compression data at the 8th order or below Fourier series is used to judge whether the change in the compression fluctuation of the rescuer is qualified: if the R-square is less than 0.8, it means that the amplitude deviation during compression is too large, which is an unqualified compression; if the R-square is greater than 0.8, the subsequent compression intensity and compression frequency are judged, and then the difference between the maximum compression value and the absolute value of the standard compression is judged to be within 7, that is, the difference in the peak value of the peak is observed, and at the same time, the compression frequency is judged to be within 100-120 per minute, that is, the number of fluctuations of the fitting curve is observed. In the time t seconds of data collection, assuming that the number of peaks is n, the compression frequency f is expressed as: That is, determine whether f is between the interval [1.67,2]; If any one of the items is not met, it is considered an unqualified press; finally, the thermal map is used to determine whether the press position is correct. If it is correct, it means that this is a qualified press that meets the indicators.

Images