A cardiopulmonary resuscitation and defibrillation integrated machine and a control method thereof

Abstract

The application provides a cardiopulmonary resuscitation defibrillation integrated machine and a control method thereof, which comprises a base plate, a pressing push rod, a negative pressure suction disc, a plurality of curved plastic columns, a pair of electrode pads and a host, the host is connected with the base plate through the plurality of curved plastic columns, the host comprises a shell, a display screen and a control device, two holes are arranged below the shell, the two holes are connected with the pair of electrode pads, the display screen is used for displaying the vital signs of a current patient in real time, and the pressing push rod is connected with the negative pressure suction disc; the cardiopulmonary resuscitation instrument and the defibrillation instrument are combined into one, the sensor feedback data are intelligently analyzed to obtain the best defibrillation time and the most suitable machine pressing strength, the quality of external chest compression is improved, the chest wall rebound is ideal, the cardiopulmonary resuscitation defibrillation integrated machine does not need to be repeatedly disassembled and assembled during use, precious rescue time is saved, space is saved, the limitation on the rescue space is reduced, and the rescue efficiency is improved. The cardiopulmonary resuscitation defibrillation integrated machine has multiple functions, good cardiopulmonary resuscitation defibrillation control effect and the rescue efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cardiopulmonary resuscitation defibrillation, and in particular to a cardiopulmonary resuscitation defibrillation all-in-one machine and a control method thereof. BACKGROUND

[0002] Modern cardiopulmonary resuscitation was established in the early 1960s, and has gone through a six-decade journey after continuous improvement and popularization. It has made great achievements in the rescue of patients with respiratory and cardiac arrest. According to the 2020 AHA guideline standard, high-quality cardiopulmonary resuscitation (CPR) is implemented for patients with cardiac arrest, and if the patient develops ventricular fibrillation during the compression process, electrical defibrillation is required. However, manual CPR has high requirements for operators, and often fails to achieve the desired effect due to factors such as incorrect compression site, improper force method, poor compression depth control, and irregular compression frequency, and even causes serious complications such as fractures, pneumothorax, and hemothorax. At the same time, the time consumed in the process of switching between CPR and defibrillation is very valuable for the patient, and it is difficult for the general public to master CPR and defibrillation technology. A cardiopulmonary resuscitation machine that rescues quickly, positions accurately, and compresses moderately is necessary, and its timeliness and accuracy can significantly improve the treatment effect.

[0003] There are some cardiopulmonary resuscitation machines at home and abroad, but their cost is high, the machine is heavy, and defibrillation is not considered to be integrated. During out-of-hospital emergency care, CPR is mostly continued without feedback, and emergency personnel mainly rely on their senses and experience to perform rescue. To improve the degree of external chest compression perfusion, the compression depth of the chest is required to be greater than 5 cm, but this also increases the risk of fractures in patients, so there is a lack of trade-off control mechanism.

[0004] There is no cardiopulmonary resuscitation defibrillation all-in-one machine on the market, only some defibrillation ventilator monitoring all-in-one machines, but manual compression is still required, which requires high skills and physical strength of the operator, and also requires a certain switching time, and the device is large in size and not convenient. As for the cardiopulmonary resuscitation machine, it is mostly adjustable in stroke, but it is not automatically adjusted by feedback, and the energy delivered by the AED is fixed and cannot be adjusted according to the chest impedance, the function is single, and the control effect is poor. SUMMARY

[0005] In view of the deficiencies in the above related technologies, the present application provides a cardiopulmonary resuscitation defibrillation all-in-one machine with multiple functions and good cardiopulmonary resuscitation defibrillation control effect, and a control method thereof.

[0006] In order to solve the above technical problems, the first aspect, the embodiment of the present application provides a cardiopulmonary resuscitation defibrillation integrated machine, comprising: a backing plate, a pressing push rod, a negative pressure suction disc, a plurality of curved plastic columns, a pair of electrode pads and a host computer, the host computer is connected with the backing plate through the plurality of curved plastic columns, the host computer comprises a shell, a display screen and a control device, two holes are arranged below the shell, the two holes are connected with the pair of electrode pads, the display screen is used for displaying the vital signs of the current patient in real time, and the pressing push rod is connected with the negative pressure suction disc.

[0007] Preferably, one end of the pressing push rod is connected with a DC servo cylinder, and the other end of the pressing push rod is connected with the negative pressure suction disc.

[0008] Preferably, the cardiopulmonary resuscitation defibrillation integrated machine further comprises a plurality of sensors, and the plurality of sensors comprise a PETCO2 sensor, an electrocardio sensor, a thoracic impedance sensor, a displacement sensor and a pressure sensor.

[0009] Preferably, the plurality of curved plastic columns are 4, and the host computer is buckled and connected with the backing plate through the 4 curved plastic columns.

[0010] The second aspect, the embodiment of the present application further provides a control method of a cardiopulmonary resuscitation defibrillation integrated machine, which is applied to the cardiopulmonary resuscitation defibrillation integrated machine, and the control method comprises the following steps:

[0011] S1, the neural network PID of the cardiopulmonary resuscitation defibrillation integrated machine is initialized, and the thoracic impedance feedback defibrillation instrument is initialized;

[0012] S2, the patient is dressed with the cardiopulmonary resuscitation defibrillation integrated machine for high-quality cardiopulmonary resuscitation compression, and two electrode pads are attached to appropriate positions;

[0013] S3, whether the patient has ventricular fibrillation is judged according to the characteristics of the collected electrocardiogram, if yes, S4 is executed, and if no, S2 is returned;

[0014] S4, discharging defibrillation is started;

[0015] S5, the stop key is pressed after discharging is completed.

[0016] Preferably, the S5 further comprises the following steps:

[0017] When discharging is not completed, S4 is returned;

[0018] If the stop key is not pressed, S2 is returned.

[0019] Preferably, the S1 specifically comprises the following sub-steps:

[0020] S11, outputting three parameters through PID control;

[0021] S12, obtaining new three parameters according to the three parameters and through training and adjusting weighting coefficients of a neural network;

[0022] S13, sending the new three parameters to a controlled object for PID control.

[0023] Preferably, the control method further comprises the following sub-steps:

[0024] S14, obtaining a preset compression force and displacement of a cardiopulmonary resuscitation instrument according to a preset PETCO2 sensor and AHA database;

[0025] S15, obtaining an actual PETCO2 value of a patient through the PETCO2 sensor;

[0026] S16, a controlled object prediction mathematical model establishing module;

[0027] S17, establishing a neural network structure combined with the PID controller;

[0028] S18, after obtaining a heading deviation through training and weighting processing according to the neural network structure, outputting a PID control system response taking the heading deviation angle as input through the PID controller.

[0029] Preferably, the S1 further comprises the following sub-steps:

[0030] S19, injecting a human body through an electrode sheet by taking a constant current source circuit as an excitation signal;

[0031] S20, measuring a thoracic impedance signal on a loop through another electrode sheet;

[0032] S21, amplifying, demodulating and filtering other interference signals of the thoracic impedance signal through a preamplifier circuit and a filter circuit;

[0033] S22, discharging a weak signal through a discharging circuit to obtain a target thoracic impedance signal;

[0034] S23, sending the target thoracic impedance signal to an MCU for processing.

[0035] Compared with the related art, the host machine is connected with the gasket through the plurality of curved plastic columns, the host machine comprises a shell, a display screen and a control device, two holes are arranged below the shell and connected with the pair of electrode sheets, the display screen is used for displaying the current patient's vital signs in real time, and the pressing push rod is connected with the negative pressure suction disc; the cardiopulmonary resuscitation device and the defibrillation device are combined into one, the best time for patient defibrillation and the most suitable machine pressing force are intelligently analyzed through sensor feedback data, the quality of external chest compression is improved, the chest wall rebound is ideal, the secondary injury to the patient is reduced to the greatest extent, the pressing and defibrillation can be completed without high skills of the operator, repeated disassembly is not needed during use, the valuable rescue time is saved, space is saved, the limitation on the rescue space is reduced, and the rescue efficiency is improved. Through the control method of the cardiopulmonary resuscitation and defibrillation integrated machine, the cardiopulmonary resuscitation control effect is good, the safety is high, and the adaptation range is wide. BRIEF DESCRIPTION OF DRAWINGS

[0036] The present application will be described in detail below with reference to the drawings. The above or other aspects of the present application will become more apparent and more readily appreciated through detailed description, taken in conjunction with the following drawings, in which:

[0037] Figure 1 It is a structural schematic diagram of the cardiopulmonary resuscitation and defibrillation integrated machine of the present application;

[0038] Figure 2 It is a method flowchart of the control method of the cardiopulmonary resuscitation and defibrillation integrated machine of the present application;

[0039] Figure 3 It is a whole method flowchart of the control method of the cardiopulmonary resuscitation and defibrillation integrated machine of the present application;

[0040] Figure 4 It is a module diagram of the control system of the cardiopulmonary resuscitation and defibrillation integrated machine of the present application;

[0041] Figure 5 It is a specific method flowchart of step S1 of the present application;

[0042] Figure 6 It is another specific method flowchart of step S1 of the present application;

[0043] Figure 7 It is another specific method flowchart of step S1 of the present application;

[0044] Figure 8 It is a schematic diagram of the neural network PID control algorithm of the present application;

[0045] Figure 9 It is a schematic diagram of the neural network structure used in combination with the PID controller of the present application;

[0046] Figure 10 This is a schematic diagram of the PID control system response input to the present invention;

[0047] Figure 11 This is a schematic diagram of the equivalent model of the human body impedance model of the present invention;

[0048] Figure 12 This is a schematic diagram showing the placement of the two electrode plates in this invention;

[0049] Figure 13 This is a block diagram of the impedance detection system of the present invention;

[0050] Figure 14 This is a schematic diagram of the parallelogram law for voltage values ​​according to the present invention.

[0051] In the diagram, 1 is the main unit, 2 is the push rod, 3 is the bending plastic column, 4 is the negative pressure suction cup, 5 is the electrode plate, 6 is the pad, 7 is the hole, and 8 is the outer shell. Detailed Implementation

[0052] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0053] The specific embodiments / examples described herein are specific implementations of the present invention, used to illustrate the concept of the invention, and are illustrative and exemplary, and should not be construed as limiting the implementation methods or scope of the present invention. In addition to the embodiments described herein, those skilled in the art can employ other obvious technical solutions based on the content disclosed in the claims and specification of this application. These technical solutions include those that make any obvious substitutions and modifications to the embodiments described herein, all of which are within the protection scope of the present invention.

[0054] Example 1

[0055] like Figure 1 As shown, the present invention provides a cardiopulmonary resuscitation defibrillator integrated machine, including: a pad 6, a compression push rod 2, a negative pressure suction cup 4, multiple curved plastic columns 3, a pair of electrode pads 5, and a main unit 1. The main unit 1 is connected to the pad through the multiple curved plastic columns 3. The main unit 1 includes a housing 8, a display screen (not shown), and a control device (not shown). Two holes 7 are provided at the bottom of the housing 8, and the two holes 7 are connected to the pair of electrode pads 5. The display screen is used to display the patient's current vital signs in real time. The compression push rod 2 is connected to the negative pressure suction cup 4.

[0056] Specifically, the pad 6 is used to place the patient and provides support during chest compressions. The electrode pad 5 is located in two holes 7 at the bottom of the outer casing 8, making it easy to pull out for use. The negative pressure suction cup 4 helps the chest wall recoil better during the compression and recovery process. The main unit 1 and the pad are connected by the multiple curved plastic pillars 3 for fixation. This integrated cardiopulmonary resuscitation and defibrillation machine combines cardiopulmonary resuscitation and defibrillation into one device. Through sensor feedback data, it intelligently analyzes the optimal timing for defibrillation and the most suitable machine compression pressure, improving the quality of chest compressions, achieving ideal chest wall recoil, and minimizing secondary injury to the patient. It requires minimal operator skill to perform compressions and defibrillation, eliminates the need for repeated disassembly and assembly, saving valuable rescue time, saving space, reducing restrictions on rescue space, and improving rescue efficiency.

[0057] Specifically, the display screen can be an LED display screen used to display patient data, including waveform data, event data, audio recording data, emergency data, and data acquisition sensors. Waveform data is used to analyze ECG rhythm; event data is used for AED analysis, CPR operation and prompts, system operation and prompts, etc. Audio recording data is used for audio data during the emergency response. Emergency data describes the emergency response time, CPR duration, number of discharges, etc. Data acquisition sensors are used to collect data such as PETCO2, compression displacement, and compression force.

[0058] Specifically, performing CPR, especially effective chest compressions, for a period of time before defibrillation increases blood circulation to the coronary arteries and brain, thereby improving the success rate of defibrillation. CPR is also crucial after defibrillation, as most patients experience asystole (AS) or pulseless electrical activity (PEA) after defibrillation; CPR can convert these rhythms into perfusion rhythms. The integrated system enables rapid switching between the automated external defibrillator (AED) and CPR, reducing the switching time. Furthermore, due to automation, no manual operation is required after installation and activation, lowering the skill requirements for the operator. Combined use also improves the overall effectiveness of both parties' work.

[0059] In this embodiment, one end of the pressing push rod 2 is connected to a DC servo electric cylinder, and the other end of the pressing push rod 2 is connected to the negative pressure suction cup 4.

[0060] Specifically, the pressing part adopts an electrically controlled driving mode, and a direct current servo cylinder is used for pressing, so that the weight and volume are further reduced under the premise of meeting the pressing frequency, pressing force and pressing depth in the AHA guideline, and the device is more portable; the suction cup 4 type silica gel pressing head is combined with the automatic mechanical device, so that the chest recoil is more sufficient, and higher quality CPR can be realized.

[0061] The defibrillation part shares a power supply with the cardiopulmonary resuscitation device, and the electrode pads 5 are located in two holes 7 below the main machine 1 housing 8, and a structure similar to a tape measure is used, which can be telescopic, more portable, and the electrode pads 5 can be attached during pressing, and defibrillation can be performed when the defibrillation effect is detected, which greatly reduces the conversion time of the cardiopulmonary resuscitation device and the AED.

[0062] Optionally, the negative pressure suction cup 4 is a suction cup 4 type silica gel pressing head.

[0063] In the embodiment, the cardiopulmonary resuscitation and defibrillation all-in-one machine further comprises a plurality of sensors, including a PETCO2 sensor, an electrocardio sensor, a thoracic impedance sensor, a displacement sensor and a pressure sensor. The blood perfusion degree can be judged by detecting the partial pressure of end-tidal CO2 (PETCO2) of the patient in real time, and compared with the preset degree, so as to adjust and control the pressing depth of the motor in real time to realize the control of the blood perfusion degree. The PID control based on the neural network has the advantages of small overshoot, fast regulation speed and high accuracy, so as to achieve ideal control effect. The output energy can be judged by detecting the thoracic impedance of the patient in real time, so as to avoid the myocardial stunning and electrical burn caused by the excessive defibrillation current output by the AED for the patient with low thoracic impedance, and the defibrillation current output is too small to achieve the effect when the impedance is high.

[0064] In the embodiment, the plurality of curved plastic columns 3 are 4, and the main machine 1 is connected by buckling type connection of the gasket plate 6 through the 4 curved plastic columns.

[0065] Embodiment two

[0066] As shown in Figures 2-4 The embodiment of the application further provides a control method of a cardiopulmonary resuscitation and defibrillation all-in-one machine, which is applied to the cardiopulmonary resuscitation and defibrillation all-in-one machine of embodiment one, and the control method comprises the following steps:

[0067] S1, the neural network PID of the cardiopulmonary resuscitation and defibrillation all-in-one machine and the defibrillation instrument of the thoracic impedance feedback are initialized.

[0068] Specifically, the control method uses a control system mainly including two aspects: one is a cardiopulmonary resuscitation device based on a neural network PID; and the other is an AED (defibrillator) based on thoracic impedance feedback.

[0069] Specifically, the control of compression and defibrillation is performed by a microprocessor, the collected data are fed back to the microprocessor by a sensor module, the processed data are sent to an upper computer information management platform by the microprocessor, and data management, data sampling storage, data classification identification, system processing control and the like are performed by the upper computer information management platform, so that the control and management effect is good and the safety is high.

[0070] The sensor module includes a PETCO2 sensor, an electrocardiogram sensor, a thoracic impedance sensor, a displacement sensor and a pressure sensor. The blood perfusion degree can be judged by detecting the partial pressure of end-tidal CO2 (PETCO2) of a patient in real time, and compared with a preset degree, so as to adjust and control the compression depth of the motor in real time to realize the control of the blood perfusion degree. The PID control based on the neural network has the advantages of small overshoot, fast regulation speed and high accuracy, so as to achieve an ideal control effect. The output energy can be judged by detecting the thoracic impedance of the patient in real time, so as to avoid the myocardial stunning and electrical burn caused by the too large defibrillation current output by the AED for a patient with low thoracic impedance, and the too small defibrillation current output for a patient with high impedance and unable to achieve the effect.

[0071] The microprocessor is used for controlling servo motor driving, AED high-voltage charging / discharging, data storage, limit detection, human-computer interaction, physiological characteristic display and a serial communication port. The serial communication port is used for communication connection with a port of the upper computer, so as to facilitate data uploading.

[0072] S2, wearing the cardiopulmonary resuscitation and defibrillation all-in-one machine on the patient to perform high-quality cardiopulmonary resuscitation compression, and pasting two electrode pads on appropriate positions.

[0073] S3, judging whether the patient has ventricular fibrillation according to the characteristics of the collected electrocardiogram, if yes, performing S4, and if no, returning to the front of S2.

[0074] S4, starting to discharge defibrillation.

[0075] S5, pressing the stop key after the discharge is completed.

[0076] Specifically, when the patient has a cardiac arrest, the rescuer quickly wears the cardiopulmonary resuscitation and defibrillation integrated machine on the patient to perform high-quality cardiopulmonary resuscitation compression, and at the same time, two electrode pads are attached to the appropriate position. Whether the patient has ventricular fibrillation is determined according to the characteristics of the collected electrocardiogram (ECG). When ventricular fibrillation (VF) occurs, the controller stops the compression of the cardiopulmonary resuscitation device and starts defibrillation; when VF does not occur, continuous compression is performed. By integrating the cardiopulmonary resuscitation device and defibrillation, intelligent analysis of the best defibrillation opportunity and the most suitable machine compression force of the patient is realized through sensor feedback data, so that the quality of external chest compression is improved, the chest wall rebound is ideal, and the secondary damage to the patient is minimized. At the same time, the operator's skills do not need to be very high to complete the compression and defibrillation, and the device does not need to be repeatedly disassembled during use, which saves valuable rescue time, saves space, reduces the limitation on the rescue space, and improves the rescue efficiency. Through the control method of the cardiopulmonary resuscitation and defibrillation integrated machine, the cardiopulmonary resuscitation control effect is good, the safety is high, and the adaptation range is wide.

[0077] Preferably, the S5 further comprises the following steps:

[0078] When the discharge is not completed, return to the S4; if the stop key is not pressed, return to the S2.

[0079] Specifically, when the discharge is not completed, the discharge defibrillation needs to be continued to protect the safety of the patient. However, when the discharge is completed, the stop key is not pressed, and high-quality cardiopulmonary resuscitation compression needs to be continuously performed on the patient.

[0080] In this embodiment, as shown in the S1, the S1 specifically comprises the following sub-steps: Figure 5

[0081] S11, output three parameters through PID control.

[0082] Specifically, the three parameters of the PID controller are the three outputs KP, KI, and KD of the neural network, which are used to control the controlled object.

[0083] S12, obtain new three parameters according to the three parameters and by training and adjusting the weighting coefficients of the neural network. Through the training and adjustment of the weighting coefficients of the neural network, the outputs KP, KI, and KD of the neural network are stabilized to correspond to the PID controller parameters under a certain optimal control law.

[0084] S13, send the new three parameters to the controlled object for PID control.

[0085] Specifically, as shown in the S1, Figure 8 ​As shown, when the sternum is soft, increase the compression depth, and obtain better blood perfusion under the condition of safety; when the sternum is hard, reduce the compression depth, and reduce the risk of sternum fracture while maintaining the necessary blood perfusion. Therefore, it is a crucial point to study and determine the appropriate compression depth.

[0086] Since the neural network can understand the structure, parameters, uncertainties and nonlinearities of the system through its own learning process, and give the control law required by the system, the controller composed of the neural network has good regulation ability and robustness. PID control based on neural network model shows good adaptability and robustness to deterministic systems and parameter-varying controlled systems with time delay.

[0087] Further, as shown in Figure 6 The control method further includes the following sub-steps:

[0088] S14, according to the preset PETCO2 sensor and AHA database, the compression force and displacement of the cardiopulmonary resuscitation instrument are obtained.

[0089] S15, the actual PETCO2 value of the patient is obtained through the PETCO2 sensor.

[0090] S16, the controlled object prediction mathematical model establishment module.

[0091] S17, the neural network structure combined with the PID controller is established.

[0092] S18, according to the neural network structure, training and weighting processing are carried out, and after the heading deviation, the PID control system response with the heading deviation angle as the input is output through the PID controller.

[0093] Specifically, according to the preset PETCO2 and AHA guidelines, we can obtain the compression force and displacement of the cardiopulmonary resuscitation instrument. The actual PETCO2 value of the patient can be obtained through the PETCO2 sensor. The controlled object prediction mathematical model establishment module: the neural network generally needs to use the prediction output of the system to train (adjust the weighting coefficient) the neural network.

[0094] The neural network structure for combining with the PID controller is as shown in Figure 9 .

[0095] The neural network is a BP neural network (three-layer structure), and the output layer contains three nodes because the output needs to correspond to the three parameters of the PID controller. The input of the input node is the state quantity of the system running, and two nodes are selected according to the system. The number of hidden layer nodes is selected according to the complexity of the system, which is three here. Since K P , KI , K D The value of is non-negative, so the activation function of the output layer neuron is a non-negative Sigmoid function, and the activation function of the hidden layer neuron is a positive-negative symmetrical Sigmoid function.

[0096] The deviation can be obtained by difference analysis of the preset value and the actual value And the change rate of the deviation is calculated

[0097] Taking the actual PETCO2 value of the patient as the benchmark, the deviation is divided into two types. When the actual PETCO2 is greater than the preset PETCO2, it indicates that the perfusion degree is good, and the compression depth can be reduced to reduce the damage to the chest. At this time, the preset value is set to Similarly, When the actual PETCO2 is less than the preset PETCO2, it indicates that the compression is not enough, and the compression needs to be increased.

[0098] The input expression of the BP neural network is as follows in combination with the specific situation of control:

[0099]

[0100]

[0101] The input and output expressions of the hidden layer neurons of the network are as follows:

[0102]

[0103]

[0104] In the formula, w ij is the threshold value, and g[·] is the activation function, The superscript (1) in the formula represents the input layer, (2) represents the hidden layer, and (3) represents the output layer.

[0105] The input and output expressions of the output layer neurons of the network are as follows:

[0106]

[0107]

[0108]

[0109]

[0110]

[0111] In the formula, w is the output layer weighting coefficient; w lQ (Q=0,1,2) is the threshold value; f[·] is the activation function,

[0112] The performance metric function takes the following expression:

[0113]

[0114] The weighting coefficients of the network are corrected using the fastest descent method, and the inertia term should be chosen to enable the search to converge quickly.

[0115]

[0116] In the formula, η is the learning rate, η>0; α is the inertia coefficient, 0<α<1.

[0117]

[0118] The prediction model of the online estimation system (using the linear identification method) can be derived.

[0119] The formula above can be rearranged to obtain the formula for calculating the weighting coefficients of the output layer of a BP neural network, as shown below:

[0120]

[0121]

[0122] The formula for calculating the weighting coefficients of the hidden layer is as follows:

[0123]

[0124]

[0125] In the formula;

[0126]

[0127] Specifically, such as Figure 10 As shown, after obtaining the heading deviation, the PID controller outputs the heading deviation angle. The input is the PID control system response. Different deviations and rates of change of deviation represent different patient states. Based on these two pieces of information, we need to impose different requirements on the performance indicators of the PID controller. This patent adjusts the parameters of the PID controller according to the system operating state, and the output of the neural network corresponds to the K of the PID controller. p K i K d Three adjustable parameters are used to optimize the PID parameters through the self-learning of the neural network and the adjustment of the weight coefficients. The PID controller is then used to control the depth and force of the cardiopulmonary resuscitation device.

[0128] In this embodiment, as shown in Figure 7 S1 further comprises the following sub-steps:

[0129] S19, injecting the human body through an electrode sheet by a constant current source circuit as an excitation signal.

[0130] S20, measuring the thoracic impedance signal on the loop through another electrode sheet.

[0131] S21, amplifying, demodulating and filtering other interference signals of the thoracic impedance signal through a preamplifier circuit and a filter circuit.

[0132] S22, discharging the weak signal through a discharge circuit to obtain the target thoracic impedance signal.

[0133] S23, sending the target thoracic impedance signal to the MCU for processing.

[0134] Specifically, the thoracic impedance detection technology is a non-invasive detection technology that uses the electrical properties of human organs and tissues and the change law of the electrical properties to obtain human medical information. The thoracic impedance detection technology is to apply a certain frequency sinusoidal voltage or current signal to the chest through the electrode sheet attached to the chest, detect the current or voltage signal at the electrode end at the same time, and then calculate the thoracic impedance and its change through the amplitude and phase of the current or voltage signal.

[0135] By establishing a human body impedance model, there are various model theories of human body impedance model, and the three-element biological impedance model is widely recognized. Cells are composed of cell membranes and intracellular fluid, and cell membranes are wrapped by extracellular fluid and intercellular substance. The intracellular fluid and extracellular fluid are various ionic electrolytes, and in terms of their electrical properties, the extracellular fluid and intracellular fluid are equivalent to resistors; the cell membrane is a phospholipid bilayer structure, and its electrical properties can be equivalent to a capacitor; the intercellular substance is composed of fibrous macromolecular substances, and it is generally considered to be an electrical insulator. Its equivalent model is shown in Figure 11 , wherein: R1 is the resistance of the extracellular fluid; C1 is the capacitance of the extracellular fluid; R2 is the resistance of the cell membrane; C2 is the capacitance of the cell membrane; R3 is the resistance of the intracellular fluid; C3 is the capacitance of the intracellular fluid.

[0136] Specifically, the detection method of thoracic impedance: In the field of external defibrillation, the method of thoracic impedance detection is usually to use two-electrode method, which is to apply a low-amplitude, high-frequency sinusoidal excitation current to the patient's chest through a pair of electrodes, and then detect the voltage signal at both ends of the electrode. The voltage signal is a modulated signal, which is modulated together with the thoracic impedance signal by the high-frequency sinusoidal excitation current signal as the carrier, and this process is also called amplitude modulation.

[0137] As shown in Figure 12As shown, the placement of the two electrode pads is in accordance with the defibrillation standard, the left electrode pad is placed under the right clavicle of the human body, and the right electrode pad is placed at the left apex of the human body. The bioimpedance between the two electrodes is the thoracic impedance Z, and the sinusoidal excitation current Is applied between the two electrodes forms a modulated voltage signal Uz related to the thoracic impedance at the two ends of the electrodes after passing through the thoracic impedance Z, that is, Uz = Is x Z. The modulated signal Uz at the two ends of the modulation electrode is detected by the thoracic impedance signal detection module, and the amplitude thereof is linearly related to the thoracic impedance, and then the thoracic impedance signal is demodulated by the demodulation circuit.

[0138] As shown in Figure 13 , a typical impedance detection system is composed of four parts: constant current source, signal acquisition, amplification and demodulation.

[0139] Specifically, the constant current source circuit is a positive selection wave signal, which is injected into the human body through the electrode as an excitation signal; the thoracic impedance signal on the measurement loop is measured through another electrode, and the signal is amplified, demodulated and filtered from other interference signals through the preamplifier and filter circuit; the weak signal is amplified through the amplification circuit, and the thoracic impedance signal is obtained, and then transmitted to the MCU for processing.

[0140] To eliminate the impedance calculation error caused by the excitation current error, the following calculation model is used: a standard resistor Rs and a human bioelectric resistance Zx are connected in series, an AC sinusoidal wave current Is of about 50 kHz is used to flow into the series impedance, and the voltages at the two ends of Rs, Zx and (Rs+Zx) are detected respectively to obtain voltage values Ur, Uz and Ui. According to the circuit principle, Ur, Uz and Ui should satisfy the parallelogram rule, which is shown in Figure 14 .

[0141] Let the current output by the current source be Is, then:

[0142]

[0143] In the formula, |U Z |, |I R | respectively represent the amplitude or effective value of U Z and U R ; θ represents the phase difference between U Z and U R .

[0144] According to the above formula, the amplitude of Z X is:

[0145]

[0146] The phase angle θ of Z X is:

[0147]

[0148] Therefore, by measuring Rs, Zx and the voltage amplitude or effective value between (Rs+Zx), the value of thoracic impedance Zx can be obtained. The calculation method eliminates the error introduced by the excitation current, greatly improving the accuracy of impedance detection.

[0149] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A control method for a cardiopulmonary resuscitation and defibrillation integrated machine, wherein the cardiopulmonary resuscitation and defibrillation integrated machine is characterized in that, include: The device comprises a pad, a pressing rod, a negative pressure suction cup, multiple curved plastic columns, a pair of electrode pads, and a main unit. The main unit is connected to the pad via the multiple curved plastic columns. The main unit includes a housing, a display screen, and a control device. Two holes are provided at the bottom of the housing, connecting to the pair of electrode pads. The display screen shows the patient's vital signs in real time. The pressing rod is connected to the negative pressure suction cup. The cardiopulmonary resuscitation defibrillator also includes multiple sensors, including a PETCO2 sensor, an ECG sensor, a chest impedance sensor, a displacement sensor, and a pressure sensor. The control method includes the following steps: S1. Initialize the neural network PID cardiopulmonary resuscitation device and the chest impedance feedback defibrillator of the integrated cardiopulmonary resuscitation and defibrillation machine. S2. Put the cardiopulmonary resuscitation defibrillator on the patient and perform high-quality cardiopulmonary resuscitation compressions, while attaching the two electrode pads to the appropriate positions. S3. Determine whether the patient has ventricular fibrillation based on the characteristics of the collected electrocardiogram. If yes, proceed to S4; otherwise, return to the point before S2. S4. Begin defibrillation; S5. Press the stop button after the discharge is complete; S1 specifically includes the following sub-steps: S11. Output three parameters through the PID controller; S12. Based on the three parameters, and by training the neural network and adjusting the weighting coefficients, obtain three new parameters; S13. Send the three new parameters to the controlled object for PID control; The control method further includes the following sub-steps: S14. Based on the preset PETCO2 sensor and AHA database, obtain the preset compression force and displacement of the cardiopulmonary resuscitation device. S15. Obtain the patient's actual PETCO2 value through the PETCO2 sensor; S16. Establish a predictive mathematical model based on the controlled object; S17. Establish a neural network structure that is combined with the PID controller; S18. After training and weighting the neural network structure to obtain the heading deviation, the PID controller outputs the PID control system response with the heading deviation angle as input. The three parameters are defined as follows: , , , , , Since the value of is non-negative, the activation function of the output layer neurons should be a non-negative sigmoid function, while the activation function of the hidden layer neurons is a positive-negative symmetric sigmoid function. The deviation can be obtained by performing a difference analysis between the preset value and the actual value. And calculate the rate of change of the deviation. ; The input expression for the neural network is as follows: ； ； The input-output expressions of the hidden layer neurons in the network are as follows: ； ](i=0,1,2,3)； In the formula As the threshold, g[ ] is the activation function, g[ ]=tanh(x)= In the formula, the superscript (1) represents the input layer, (2) represents the hidden layer, and (3) represents the output layer; The input-output expressions of the neurons in the output layer of the network are as follows: ； ](l=0,1,2)； = ； = ； = ； In the formula These are the weighting coefficients of the output layer; (Q=0,1,2) is the threshold; f[ ] is the activation function, f[ ]=(1 / 2)[1+tanh(x)]= ; The performance metric function takes the following expression: J= 2= z2(k+1); The weighting coefficients of the network are corrected using the fastest descent method, and the inertia term should be chosen to enable the search to converge quickly. In the formula For learning rate, >0; The inertia coefficient, 0 < <1; The prediction model of the online estimation system can be derived ; The formula above can be rearranged to obtain the formula for calculating the weighting coefficients of the neural network output layer, as shown below: The formula for calculating the weighting coefficients of the hidden layer is as follows: ； In the formula; ； 。 2. The control method for the cardiopulmonary resuscitation and defibrillation integrated machine as described in claim 1, characterized in that, S5 further includes the following steps: If the discharge is incomplete, return to step S4; If the stop button is not pressed, return to step S2.

3. The control method for the cardiopulmonary resuscitation and defibrillation integrated machine as described in claim 1, characterized in that, S1 specifically includes the following sub-steps: S19. The constant current source circuit is used as the excitation signal and injected into the human body through an electrode plate. S20. Measure the chest impedance signal on the circuit using another electrode. S21. The chest impedance signal is amplified, demodulated, and other interference signals are filtered out by passing it through a preamplifier circuit and a filter circuit. S22. The weak signal is discharged through the discharge circuit to obtain the target chest impedance signal; S23. Send the target chest impedance signal to the MCU for processing.

4. The control method for the cardiopulmonary resuscitation and defibrillation integrated machine as described in claim 1, characterized in that, One end of the pressing push rod is connected to a DC servo electric cylinder, and the other end of the pressing push rod is connected to the negative pressure suction cup.

5. The control method for the cardiopulmonary resuscitation and defibrillation integrated machine as described in claim 1, characterized in that, The multiple curved plastic columns consist of four pieces, and the main unit is connected to the pad through the four curved plastic columns in a snap-fit ​​manner.

Images