CN115970164A - Cardio-pulmonary resuscitation and defibrillation all-in-one machine control system and method based on human physiological model - Google Patents

Abstract

The invention discloses a cardio-pulmonary resuscitation and defibrillation all-in-one machine control system and a method based on a human body physiological model, wherein the method comprises the following steps: acquiring electrocardiogram characteristic data and compression characteristic data of a patient; calculating to obtain a cardiopulmonary feature result and grading levels corresponding to the BI value, the RI value and the BRI value according to the electrocardiogram feature data and the compression feature data; judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the compression process according to the grading level, if BRI is less than 5, continuing to judge BI and RI, if BI is less than 4.5, gradually increasing the compression force until BI is greater than 4.5, and if RI is greater than 3.4, gradually decreasing the compression force until RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs or not according to the cardio-pulmonary characteristic result, if yes, stopping pressing, and performing defibrillation; and judging whether ventricular fibrillation still occurs or not according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation. The invention solves the problem that the successful treatment of the patient is greatly reduced because the cardiopulmonary resuscitation instrument and the defibrillator are required to be replaced alternately to treat the patient.

Description

Cardio-pulmonary resuscitation and defibrillation all-in-one machine control system and method based on human physiological model

Technical Field

The invention relates to the technical field of medical instruments, in particular to a cardio-pulmonary resuscitation and defibrillation all-in-one machine control system and method based on a human body physiological model.

Background

The manual CPR has high requirements on operators, and often causes the factors of inaccurate compression parts, improper force application methods, poor control of compression depth, irregular compression frequency and the like, so that ideal effects are difficult to achieve, and even serious complications such as fracture, pneumothorax, hemothorax and the like are caused, and therefore, the optimal extrathoracic compression depth meeting individual characteristics of patients can be obtained only by combining the blood perfusion degree and the sternum fracture risk of the patients. However, currently, international attention to the balance problem of the external chest compression pros and cons is lacking, the compression depth of a plurality of automatic external chest devices is constant, and the optimized external chest compression after the balance of the external chest compression pros and cons cannot be carried out according to individual differences of patients. A mechanism for balancing the pros and cons of chest compressions has not been established. Meanwhile, the mouth-to-mouth artificial respiration may also cause diseases to be transmitted between the patient and the rescuer. Therefore, a cardiopulmonary resuscitator which is rapid in rescue, accurate in positioning and appropriate in compression is needed, and timeliness and accuracy of the cardiopulmonary resuscitator can remarkably improve the rescue effect.

Some products of cardiopulmonary resuscitators exist at home and abroad, but the cost is high, the machine is heavy, and defibrillation is not considered to be added to realize integration. At present, the cardiopulmonary resuscitation apparatus is operated independently, and most areas are only equipped with one of the cardiopulmonary resuscitation apparatus or the defibrillator, however, in the first aid process, the cooperation of the cardiopulmonary resuscitation apparatus and the defibrillator is very important, the purpose of performing chest compression is to manually pump blood, squeeze the blood to each internal organ of the whole body, and prevent each internal organ from being necrotic due to ischemia and oxygen deficiency. But an AED may be used to normalize a disordered heart rhythm in order to restore the patient's heartbeat. High quality cardiopulmonary resuscitation can improve the success rate of defibrillation. Most importantly, the patient does not have to defibrillate once to recover normal heart rate, so AEDs and cardiopulmonary resuscitation are performed alternately, but in short. After the device is installed, the device is pressed according to the standard of a guideline according to the programming requirement, only the pressing work is finished, if the patient has ventricular fibrillation through pressing, the patient needs to be electrically defibrillated, at the moment, the device needs to be moved away from the patient and then defibrillated, the device needs to be immediately pressed after defibrillation, at the moment, the cardio-pulmonary resuscitation device needs to be installed again, the moving away, defibrillating and reinstallation processes are the time passing, and the possibility that the patient is successfully rescued is greatly reduced.

Disclosure of Invention

Aiming at the defects, the invention provides a cardio-pulmonary resuscitation and defibrillation all-in-one machine control system and a method based on a human body physiological model, and aims to solve the problem that the successful treatment of a patient is greatly reduced because the conventional cardio-pulmonary resuscitation instrument and defibrillator need to be replaced alternately to treat the patient.

In order to achieve the purpose, the invention adopts the following technical scheme:

a control system of cardio-pulmonary resuscitation and defibrillation integrated machine based on a human physiological model comprises

The physiological signal acquisition module is used for acquiring electrocardiogram characteristic data and compression characteristic data of a patient;

the sending module is used for sending the electrocardiogram characteristic data and the pressing characteristic data to the main control module;

the main control module is used for calculating and obtaining a cardiopulmonary feature result, calculating a BI value, an RI value and a BRI value, and obtaining a corresponding grading level;

the pressing judgment module is used for judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the pressing process according to the grading level, if the BRI is less than 5, the BI and the RI are continuously judged, if the BI is less than 4.5, the pressing force is gradually increased until the BI is greater than 4.5, and if the RI is greater than 3.4, the pressing force is gradually decreased until the RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs or not according to the cardio-pulmonary characteristic result, if yes, stopping pressing, and performing defibrillation;

and the defibrillation judgment module is used for judging whether ventricular fibrillation still occurs or not according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation.

Preferably, the calculation formula of the BI value is as follows:

wherein BI represents a beneficial index and PETCO2 represents CO measured at the end of breath ₂ A concentration value;

the RI value is calculated as follows:

wherein RI represents a risk index, kchest represents a sternum elastic modulus, which refers to an action force value required by a compression unit displacement, K5 represents that the chest compression depth of an adult is 5cm, and K6.5 represents that the chest compression depth of the adult is 6.5cm;

the BRI value is calculated as follows:

where BRI represents the composite index of beneficial risk.

Preferably, the pressing judgment module comprises an impedance control submodule, and the impedance control submodule is used for establishing an impedance model; calculating to obtain a reaction elasticity error value of the chest compression, wherein the calculation formula of the reaction elasticity error value is as follows:

ΔF(t)＝F _d (t)-F(t)

wherein, delta F (t) is the error value of the reaction elastic force, F _d (t) is a constant representing the desired force, and F (t) is the reaction spring force, the calculation of which includes the following three formations:

wherein, M _d As a quality parameter, B _d As a damping parameter, K _d In order to be a parameter of the stiffness,

for the actual acceleration of the pressing head end>

The actual speed of the pressing head end, X (t) is the position of the pressing head end, and>

for desired acceleration of the pressing head end>

Desired speed of the pressing head end, X _d (t) is the desired position of the pressing head end;

inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain; and filtering the reaction elastic force signal according to the frequency domain, and converting the filtered reaction elastic force signal into a corresponding motion signal of the tail end position.

Preferably, the main control module includes an electrocardiographic signal preprocessing sub-module, and the electrocardiographic signal preprocessing sub-module is configured to locate a QRS group wave by setting an adaptive threshold, where a calculation formula of the adaptive threshold is as follows:

/>

wherein THR1 is a high threshold; mean (peak _ buffer) is the peak mean; peak is the signal peak detected at this moment; THR1_ lim is an empirical constant, represents the upper bound of threshold variation, and takes the value of 0.33; THR2 is low threshold; peak _ buffer is a buffer value for storing 8 continuous peak values before the peak value at the moment; TRH2_ lim is also an empirical constant representing the lower bound of the threshold change, and takes a value of 0.23.

Preferably, the main control module further comprises an arrhythmia identification submodule, and the arrhythmia identification submodule is used for extracting electrocardiosignal characteristics; selecting and dividing a data set to obtain a training set and a testing set; carrying out characteristic normalization processing on the data of the training set and the test set; training and testing a support vector machine model, and inputting the preprocessed electrocardiosignals into the support vector machine model for classification.

Another aspect of the present application provides a cardio-pulmonary resuscitation defibrillation all-in-one machine control method based on a human physiological model, including the following steps:

step S1: acquiring electrocardiogram characteristic data and compression characteristic data of a patient;

step S2: calculating to obtain a cardiopulmonary feature result and grading levels corresponding to the BI value, the RI value and the BRI value according to the electrocardiogram feature data and the compression feature data;

and step S3: judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the compression process according to the grading level, if BRI is less than 5, continuing to judge BI and RI, if BI is less than 4.5, gradually increasing the compression force until BI is greater than 4.5, and if RI is greater than 3.4, gradually decreasing the compression force until RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs according to the cardiopulmonary characteristic result, if yes, stopping pressing, and performing defibrillation;

and step S4: and judging whether ventricular fibrillation still occurs or not according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation.

Preferably, in step S3, the impedance control sub-step is specifically included:

step S31: establishing an impedance model;

step S32: calculating to obtain a reaction elasticity error value of the chest compression, wherein the calculation formula of the reaction elasticity error value is as follows:

ΔF(t)＝F _d (t)-F(t)

wherein Δ F (t) is a reaction elasticity error value, F _d (t) is a constant representing the desired force, and F (t) is the reaction spring force, the calculation of which includes the following three formations:

wherein M is _d As a quality parameter, B _d As a damping parameter, K _d As a parameter of the stiffness, it is,

for the actual acceleration of the pressing head end>

The actual speed of the pressing head end, X (t) is the position of the pressing head end, and>

for a desired acceleration of the pressing head end>

To press the desired speed of the head tip, X _d (t) is the desired position of the pressing head end;

step S33: inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain;

step S34: and filtering the reaction elastic force signal according to the frequency domain, and converting the filtered reaction elastic force signal into a corresponding motion signal of the tail end position.

Preferably, in step S2, the electrocardiosignal preprocessing sub-step is specifically included: positioning QRS group waves by setting an adaptive threshold, wherein the adaptive threshold is calculated according to the following formula:

wherein THR1 is a high threshold; mean (peak _ buffer) is the peak mean; peak is the peak of the signal detected at this moment; THR1_ lim is an empirical constant, represents the upper bound of threshold variation and takes the value of 0.33; THR2 is low threshold; peak _ buffer is a buffer value for storing 8 continuous peak values before the peak value at the moment; TRH2_ lim is also an empirical constant, represents the lower bound of threshold variation, and takes the value of 0.23;

further comprising an arrhythmia identification sub-step: extracting electrocardiosignal characteristics; selecting and dividing a data set to obtain a training set and a testing set; carrying out characteristic normalization processing on the data of the training set and the test set; training and testing a support vector machine model, and inputting the preprocessed electrocardiosignals into the support vector machine model for classification.

The technical scheme provided by the embodiment of the application can have the following beneficial effects:

1. according to the scheme, the blood perfusion degree and the sternum fracture risk degree of the patient are comprehensively evaluated and judged through the grading levels corresponding to the BI value, the RI value and the BRI value, so that the injury probability of the patient in the cardiopulmonary resuscitation process can be well predicted, and the injury probability of the patient is reduced.

2. Whether the physiological condition of the patient needs defibrillation operation or not is automatically judged according to the cardiopulmonary characteristic result in the pressing process, so that the time for replacing the cardiopulmonary resuscitation instrument and the defibrillator is greatly saved, and the possibility of successfully treating the patient is greatly improved.

Drawings

FIG. 1 is a schematic structural diagram of a cardio-pulmonary resuscitation and defibrillation integrated machine based on a human physiological model;

fig. 2 is a flowchart of the steps of a method for defibrillation based on a physiological model of a human body.

Wherein, 1, a host shell; 2. a display screen; 3. an electrode sheet; 4. a push rod; 5. a pressing head; 6. a base plate; 7. a bandage.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.

A control system of cardio-pulmonary resuscitation and defibrillation integrated machine based on a human physiological model comprises

The physiological signal acquisition module is used for acquiring electrocardiogram characteristic data and compression characteristic data of a patient;

the sending module is used for sending the electrocardiogram characteristic data and the pressing characteristic data to the main control module;

the main control module is used for calculating and obtaining a cardiopulmonary feature result, calculating a BI value, an RI value and a BRI value, and obtaining a corresponding grading level;

the pressing judgment module is used for judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the pressing process according to the grading level, if the BRI is less than 5, the BI and the RI are continuously judged, if the BI is less than 4.5, the pressing force is gradually increased until the BI is greater than 4.5, and if the RI is greater than 3.4, the pressing force is gradually decreased until the RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs or not according to the cardio-pulmonary characteristic result, if yes, stopping pressing, and performing defibrillation;

and the defibrillation judgment module is used for judging whether ventricular fibrillation still occurs or not according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation.

The CPR-defibrillation all-in-one machine control system based on the human physiological model adopts the FPGA controller as a carrier of the CPR-defibrillation all-in-one machine, various algorithms can be realized in parallel due to the unique hardware description language and the internal structure of the FPGA controller, in the aspect of delay, the FPGA is superior to a GPU, the FPGA can provide high bandwidth and reduce delay, and the operation speed is greatly improved. As shown in fig. 1, the integrated cardiopulmonary resuscitation and defibrillation machine comprises an FPGA controller, a host shell 1, a display screen 2, two electrode plates 3, a push rod 4, a sucker-type silica gel pressing head 5, a base plate 6 and a bandage 7, wherein the FPGA controller is arranged inside the host shell 1, the display screen 2 is arranged outside the host shell 1, and the display screen 2 is electrically connected with the FPGA controller. And the two electrode plates 3 are electrically connected with the FPGA controller. The inside of host computer shell 1 still is provided with motor and servo electric cylinder, the drive end of motor connect in servo electric cylinder, the one end of push rod 4 connect in servo electric cylinder, the other end of push rod 4 connect in the silica gel pressing head 5 of sucking disc formula. The pad 6 is used for placing a patient, and the bandage 7 is used for fixing the patient on the pad 6.

When the patient takes place the cardiac arrest, the rescuer helps the patient to dress cardiopulmonary resuscitation defibrillate all-in-one rapidly, will two electrode slice 3 pastes to the exact position, and is concrete, and anodal subsides are under patient's right side clavicle, and the negative pole pastes the below outside the nipple of patient's left side. The integrated CPR-defibrillation machine is started, a physiological signal acquisition module in a CPR-defibrillation integrated machine control system can acquire electrocardiogram characteristic data and compression characteristic data of a patient, the electrocardiogram characteristic data and the compression characteristic data are transmitted to a main control module through a transmission module, the main control module calculates to obtain a CPR characteristic result, calculates a BI value, an RI value and a BRI value, and obtains corresponding grading levels, wherein the BI value, the RI value and the BRI value are three quantitative indexes, and the BI refers to a beneficial index and is used for reflecting the degree of perfusion of blood; RI refers to a risk index for discriminating the degree of risk of sternal fracture; BRI refers to a composite index of beneficial risk for the comprehensive assessment of the pros and cons characteristic of chest compressions during cardiopulmonary resuscitation. The obtained cardiopulmonary feature results and the corresponding ranking levels of the BI value, RI value and BRI value are displayed in the display screen 2. The integrated cardiopulmonary resuscitation and defibrillation machine performs cardiopulmonary resuscitation, specifically, the motor drives the servo electric cylinder to move by starting the motor, and the servo electric cylinder drives the push rod 4 to move up and down to realize pressing. Because the push rod 4 is connected with the sucker type silica gel pressing head 5, the chest of the patient can be better protected from being damaged by the outside, and the chest of the patient can be fully rebounded, so that a better cardiopulmonary resuscitation effect is achieved.

A pressing judgment module in the cardiopulmonary resuscitation and defibrillation all-in-one machine control system can judge the cardiopulmonary resuscitation effect and the fracture risk of a patient in the pressing process according to the grading level, if the BRI is less than 5, the BI and the RI are continuously judged, if the BI is less than 4.5, the pressing strength is gradually increased until the BI is greater than 4.5, and if the RI is greater than 3.4, the pressing strength is gradually reduced until the RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs or not according to the cardio-pulmonary characteristic result, if yes, stopping pressing, and performing defibrillation; and the defibrillation judging module judges whether ventricular fibrillation still occurs according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, defibrillation is continued.

In the scheme, the pressing judgment module is arranged, so that the blood perfusion degree and the sternum fracture risk degree of the patient can be comprehensively evaluated and judged by the pressing judgment module, the injury probability of the patient in the cardio-pulmonary resuscitation process can be better predicted, and the injury probability of the patient is reduced. Through setting up the judgement module of defibrillating, the judgement module of defibrillating can be at the feedback of pressing the in-process and judging from cardiopulmonary characteristic result, judge whether patient's physiological status needs to defibrillate automatically and operate, has saved the time of changing cardiopulmonary resuscitation appearance and defibrillator so greatly for the patient is improved greatly by the possibility of successful treatment.

Preferably, the calculation formula of the BI value is as follows:

wherein BI represents a beneficial indexPETCO2 stands for CO measured at the end of breath ₂ A concentration value;

the RI value is calculated as follows:

wherein RI represents a risk index, kchest represents a sternum elastic modulus, which refers to an action force value required by a compression unit displacement, K5 represents that the chest compression depth of an adult is 5cm, and K6.5 represents that the chest compression depth of the adult is 6.5cm;

the BRI value is calculated as follows:

wherein BRI represents a composite index of beneficial risk.

In this example, the BI values represent a beneficial index, reflecting the degree of blood perfusion. The content of the carbon dioxide at the end of respiration PETCO2 can be used as an effective physiological index for detecting the cardio-pulmonary resuscitation quality in the cardio-pulmonary resuscitation process, and because PETCO2 has the advantages of simple detection, wide use and strong correlation with the blood perfusion degree of an important index of the cardio-pulmonary resuscitation, PETCO2 is used as a physiological parameter for feeding back the blood perfusion degree. To quantify the beneficial degree of chest compressions, PETCO2 is taken as a BI value and quantified to a range of 0-10, BI can be classified into a 3-grade according to the degree of perfusion of blood reflected by PETCO2 and the likelihood of predicting success in return of spontaneous circulation, the greater the PETCO2 value, the better the degree of perfusion, the higher the beneficial grade, the greater the score, specifically, the first-grade PETCO2 ranges from 0mmHg to 10mmHg, the second-grade PETCO2 ranges from 10mmHg to 15mmHg, and the third-grade PETCO2 ranges from 15mmHg to 21mmHg, or higher than 21mmHg.

RI represents a risk index for identifying the degree of risk of sternal fracture. The sternal modulus of elasticity, kchest, is inversely related to the risk of sternal fracture. Therefore, in order to quantify the risk of sternal fracture due to the sternal elastic modulus Kchest, the RI is quantified to the range of 0 to 10 according to the size of the sternal elastic modulus Kchest, and is classified into 4 grades according to the degree of risk of sternal fracture caused by the size of the sternal elastic modulus Kchest. The 2010 AHA cardiopulmonary resuscitation and cardiovascular first aid guidelines clearly indicate that the chest compression depth should be such that the sternum is depressed by at least 5cm. According to this requirement, chest compressions with a compression depth of less than 5cm do not normally cause harm to the typical human sternum. Clinical practice has shown that most rescuers do not have chest compressions of more than 6cm in depth, and chest compressions of more than 6cm are generally considered to be over-compressions, easily resulting in sternal fractures. Therefore, kchests in the range of K0-K5 are considered harmless, and Kchests in this range are classified into 1 st stages, and RI values of the 1 st stages are 0. The value of the highest sternal elastic modulus at the highest level is from K6 to K6.5, or above K6.5. While the sternum modulus of elasticity between K5 and K6 is considered to be of intermediate order. To achieve better RI resolution, the middle level of RI is again divided into two risk levels, the greater and lesser, the second level having a sternal modulus of elasticity from K5 to K5.5 and the third level having a sternal modulus of elasticity from K5.5 to K6.

In order to comprehensively evaluate the benefit and disadvantage characteristics of chest compression during cardiopulmonary resuscitation, a comprehensive risk index BRI is provided. BRI is a quantitative indicator of the quality of chest compressions and can be found on the basis of BI and RI, if the BRI ranges from 0 to 10. If BRI is more than 5, the benefit is better than the disadvantage; if BRI is less than 5, the disadvantage is better than the benefit.

Preferably, the pressing judgment module comprises an impedance control submodule, and the impedance control submodule is used for establishing an impedance model; calculating a reaction elasticity error value of the chest compression, wherein the calculation formula of the reaction elasticity error value is as follows:

ΔF(t)＝F _d (t)-F(t)

wherein, delta F (t) is the error value of the reaction elastic force, F _d (t) is a constant representing the desired force, and F (t) is the reaction spring force, the calculation of which includes the following three formations:

wherein M is _d As a quality parameter, B _d As a damping parameter, K _d In order to be a parameter of the stiffness,

for the actual acceleration of the pressing head end>

Is the actual speed of the pressing head end, X (t) is the position of the pressing head end, and>

for a desired acceleration of the pressing head end>

Desired speed of the pressing head end, X _d (t) is the desired position of the pressing head end;

inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain; and filtering the reaction elastic force signal according to the frequency domain, and converting the filtered reaction elastic force signal into a corresponding motion signal of the tail end position.

In this embodiment, when the compression head 5 compresses the patient, the compression head 5 causes the chest of the patient to sag and receives a reaction force from the chest. Equation (3) represents a model in which the deviations of position, velocity, and acceleration are considered at the same time, and when impedance control is used, this method expresses the relationship between the error of the environmental force and the position error during compression, and chest compression is performed not only to control the compression depth (position) but also to control the force. Establishing an impedance model; calculating to obtain the error value of the reaction elasticity of the chest compression; inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain; according to the frequency domain, the reaction elastic force signal is filtered, the filtered reaction elastic force signal is converted into a corresponding motion signal of the tail end position, and the relationship between the elastic force and the position between the tail end of the pressing head and the chest cavity can be adjusted by adjusting the quality parameter, the damping parameter and the rigidity parameter of impedance control. This scheme utilization impedance control is planned the motion of press head, can realize pressing the in-process and control pressing depth and dynamics simultaneously.

Preferably, the main control module includes an electrocardiographic signal preprocessing sub-module, and the electrocardiographic signal preprocessing sub-module is configured to locate a QRS group wave by setting an adaptive threshold, where a calculation formula of the adaptive threshold is as follows:

wherein THR1 is a high threshold; mean (peak _ buffer) is the peak mean; peak is the peak of the signal detected at this moment; THR1_ lim is an empirical constant, represents the upper bound of threshold variation and takes the value of 0.33; THR2 is low threshold; peak _ buffer is a buffer value that stores 8 consecutive peaks before the peak at that moment; TRH2_ lim is also an empirical constant representing the lower bound of the threshold change, and takes a value of 0.23.

In this embodiment, in order to complete the positioning of the QRS complex in the electrocardiographic signal preprocessing, an adaptive threshold is mainly set. Because the electrocardiosignal is a non-stationary signal, when detection is carried out, all QRS waves are difficult to meet only by a fixed threshold value, false detection and missed detection are easy to cause, and the problem can be better solved by setting a self-adaptive threshold value. Further, the adaptive threshold is to be changed immediately along with the waveform, and in order to make the threshold more accurate, a dual-threshold method is adopted, specifically, two thresholds, one high threshold and one low threshold, are set. When a certain peak in the signal exceeds a set low threshold, a QRS wave is judged, and the size of the threshold is adjusted according to the relation between the high and low thresholds and the amplitude of the peak. In order to keep the waveform variation of the electrocardiograph signal stable, the threshold value needs to be adjusted according to the amplitude variation of the correct peak detected before, so as to ensure that the threshold value is not too high or too low. Setting the high and low thresholds can capture more levels of peaks to a greater extent than setting only one threshold. In the aspect of preventing error detection, a refractory period mode is adopted. When the two wave crests are close, only the larger wave crest is selected. This process refers to time intervals below 0.24s, the length of the refractory period; the presence of a lower bound on the value of the double threshold also prevents some noise from being mistakenly detected as a QRS wave to some extent.

Preferably, the main control module further comprises an arrhythmia identification submodule, and the arrhythmia identification submodule is used for extracting electrocardiosignal characteristics; selecting and dividing a data set to obtain a training set and a testing set; carrying out feature normalization processing on data of the training set and the test set; training and testing a support vector machine model, and inputting the preprocessed electrocardiosignals into the support vector machine model for classification.

In this embodiment, the arrhythmia identification submodule is configured to improve the sensitivity of the classification of the electrocardiographic signals. Specifically, the electrocardiosignal characteristics are extracted, an ECG signal data set is loaded firstly, then 5-order wavelet decomposition is carried out on each heart beat in the data set, and a wavelet function utilizes db6 wavelets. After 5-order wavelet decomposition and 2-time sampling, the 'approximate' coefficient of the original signal in the wavelet transform coefficient, namely the a coefficient after 5-order decomposition, is taken, and the coefficients are selected as the characteristic value of each heartbeat. And selecting and dividing a data set to obtain a training set and a testing set, wherein the training set and the testing set respectively comprise 10000 samples. The random selection can be realized by establishing a randderm function to randomly disorder sample indexes, then intercepting samples corresponding to the former 10000 indexes as a training set, and taking the rest as a test set. In order to accelerate the convergence of the SVM, the data of the training set and the test set are subjected to feature normalization processing, during actual operation, a normal mode of a normalization function is used firstly, the feature of the training set is normalized to be between 0 and 1, after the normalized training set is obtained, an application mode of the normalization function is used, normalization information obtained by the training set is used to the test set, and the normalization of the test set is completed. Training and testing the support vector machine model, and calling a libsvmtrain function and a libsvmpredict function to train and test the model. The default kernel function trained by the Libsvmtran function is an RBF kernel function, and 2 hyper-parameters, namely a penalty factor coefficient c and a kernel function parameter g, need to be set artificially, and are set to be 2 and 1 respectively. Different values of c and g may result in larger difference, and if a better effect is to be obtained, the parameter adjustment must be performed, so as to reduce the problems of under-fitting and over-fitting. And inputting the preprocessed electrocardiosignals into a support vector machine model for classification, judging that ventricular fibrillation occurs when a complete QRS waveform is not detected within 2s, immediately stopping pressing, and performing defibrillation operation. The defibrillation operation needs to be carried out for three times, the first defibrillation energy is 200J, judgment is carried out after defibrillation is finished, if ventricular fibrillation still occurs, the second defibrillation energy is 300J, whether ventricular fibrillation still occurs or not is continuously judged, if ventricular fibrillation still occurs, the third defibrillation energy is 360J, pressing operation is carried out after defibrillation is finished for the third time, and whether ventricular fibrillation occurs or not is continuously judged after 30 s. If ventricular fibrillation still occurs, the defibrillation procedure is continued for three times.

Another aspect of the present application provides a cardio-pulmonary resuscitation defibrillation all-in-one machine control method based on a human physiological model, including the following steps:

step S1: acquiring electrocardiogram characteristic data and compression characteristic data of a patient;

step S2: calculating to obtain a cardiopulmonary feature result and grading levels corresponding to the BI value, the RI value and the BRI value according to the electrocardiogram feature data and the compression feature data;

and step S3: judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the compression process according to the grading level, if BRI is less than 5, continuing to judge BI and RI, if BI is less than 4.5, gradually increasing the compression force until BI is greater than 4.5, and if RI is greater than 3.4, gradually decreasing the compression force until RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs or not according to the cardio-pulmonary characteristic result, if yes, stopping pressing, and performing defibrillation;

and step S4: and judging whether ventricular fibrillation still occurs or not according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation.

According to the control method of the integrated machine for cardio-pulmonary resuscitation and defibrillation based on the human physiological model, as shown in fig. 2, the blood perfusion degree and the sternum fracture risk degree of a patient are comprehensively evaluated and judged through the grading levels corresponding to the BI value, the RI value and the BRI value, so that the injury probability of the patient in the cardio-pulmonary resuscitation process can be well predicted, and the injury probability of the patient is reduced. Whether the physiological condition of the patient needs defibrillation operation or not is automatically judged according to the cardiopulmonary characteristic result in the pressing process, so that the time for replacing the cardiopulmonary resuscitation instrument and the defibrillator is greatly saved, and the possibility of successfully treating the patient is greatly improved.

Further, the calculation formula of the BI value is as follows:

wherein BI represents a beneficial index and PETCO2 represents CO measured at the end of breath ₂ A concentration value;

the RI value is calculated as follows:

wherein RI represents a risk index, kchest represents a sternum elastic modulus, which refers to an action force value required by a compression unit displacement, K5 represents that the chest compression depth of an adult is 5cm, and K6.5 represents that the chest compression depth of the adult is 6.5cm;

the BRI value is calculated as follows:

wherein BRI represents a composite index of beneficial risk.

Preferably, in step S3, the method specifically includes the sub-steps of impedance control:

step S31: establishing an impedance model;

step S32: calculating to obtain a reaction elasticity error value of the chest compression, wherein the calculation formula of the reaction elasticity error value is as follows:

ΔF(t)＝F _d (t)-F(t)

wherein, delta F (t) is the error value of the reaction elastic force, F _d (t) is a constant representing the desired force, and F (t) is the reaction spring force, the calculation of which includes the following three formations:

wherein M is _d As a quality parameter, B _d As a damping parameter, K _d In order to be a parameter of the stiffness,

for the actual acceleration of the pressing head end>

Is the actual speed of the pressing head end, X (t) is the position of the pressing head end, and>

for desired acceleration of the pressing head end>

To press the desired speed of the head tip, X _d (t) is the desired position of the pressing head end;

step S33: inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain;

step S34: and filtering the reaction elastic force signal according to the frequency domain, and converting the filtered reaction elastic force signal into a corresponding motion signal of the tail end position.

In this embodiment, the relationship between the elastic force and the position between the distal end of the compression head and the thoracic cavity can be adjusted by adjusting the mass parameter, the damping parameter, and the stiffness parameter of the impedance control. This scheme utilization impedance control does the planning to the motion of press head, can realize pressing the in-process and control pressing degree of depth and dynamics simultaneously.

Preferably, in step S2, the method specifically includes a cardiac signal preprocessing substep: positioning QRS group waves by setting an adaptive threshold, wherein the adaptive threshold is calculated according to the following formula:

wherein THR1 is a high threshold; mean (peak _ buffer) is the peak mean; peak is the signal peak detected at this moment; THR1_ lim is an empirical constant, represents the upper bound of threshold variation and takes the value of 0.33; THR2 is low threshold; peak _ buffer is a buffer value that stores 8 consecutive peaks before the peak at that moment; TRH2_ lim is also an empirical constant, represents the lower bound of threshold variation, and takes the value of 0.23;

further comprising an arrhythmia identification sub-step: extracting electrocardiosignal characteristics; selecting and dividing a data set to obtain a training set and a testing set; carrying out characteristic normalization processing on the data of the training set and the test set; training and testing a support vector machine model, and inputting the preprocessed electrocardiosignals into the support vector machine model for classification.

In the embodiment, because the electrocardiosignal is a non-stationary signal, when detection is carried out, all QRS waves are difficult to meet only by a fixed threshold, false detection and missed detection are easy to cause, and the problem can be solved well by setting the self-adaptive threshold. The scheme is combined with arrhythmia identification based on a support vector machine, and the sensitivity of electrocardiosignal classification can be effectively improved.

In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.

Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.

Claims (8)

1. Cardiopulmonary resuscitation defibrillates all-in-one control system based on human physiology model, its characterized in that: comprises a physiological signal acquisition module used for acquiring electrocardiogram characteristic data and compression characteristic data of a patient;

the sending module is used for sending the electrocardiogram characteristic data and the pressing characteristic data to the main control module;

the main control module is used for calculating and obtaining a cardiopulmonary feature result, calculating a BI value, an RI value and a BRI value, and obtaining a corresponding grading level;

the compression judging module is used for judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the compression process according to the grading level, if the BRI is less than 5, the BI and the RI are continuously judged, if the BI is less than 4.5, the compression force degree is gradually increased until the BI is greater than 4.5, and if the RI is greater than 3.4, the compression force degree is gradually decreased until the RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs according to the cardiopulmonary characteristic result, if yes, stopping pressing, and performing defibrillation;

and the defibrillation judgment module is used for judging whether ventricular fibrillation still occurs or not according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation.

2. The integrated CPR-defibrillation machine control system based on the human physiological model according to claim 1, wherein: the calculation formula of the BI value is as follows:

wherein BI represents a beneficial index and PETCO2 represents CO measured at the end of breath ₂ A concentration value;

the RI value is calculated as follows:

wherein RI represents a risk index, kchest represents a sternum elastic modulus, which refers to an action force value required by a compression unit displacement, K5 represents that the chest compression depth of an adult is 5cm, and K6.5 represents that the chest compression depth of the adult is 6.5cm;

the BRI value is calculated as follows:

wherein BRI represents a composite index of beneficial risk.

3. The integrated CPR-defibrillation machine control system according to claim 1, wherein the integrated CPR-defibrillation machine control system comprises: the pressing judgment module comprises an impedance control submodule used for establishing an impedance model; calculating to obtain a reaction elasticity error value of the chest compression, wherein the calculation formula of the reaction elasticity error value is as follows:

Fd(t)＝F _d (t)-F(t)

wherein, delta F (t) is the error value of the reaction elastic force, F _d (t) is a constant representing the desired force, and F (t) is the reaction spring force, the calculation of which includes the following three formations:

wherein M is _d As a quality parameter, B _d As a damping parameter, K _d As a parameter of the stiffness, it is,

is the actual acceleration of the end of the pressing head,

the actual speed of the pressing head end, X (t) is the position of the pressing head end, and>

for a desired acceleration of the pressing head end>

To a desired speed of the tip of the pressing headDegree, X _d (t) is the desired position of the pressing head end; />

Inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain; and filtering the reaction elastic force signal according to the frequency domain, and converting the filtered reaction elastic force signal into a corresponding motion signal of the tail end position.

4. The integrated CPR-defibrillation machine control system based on the human physiological model according to claim 1, wherein: the main control module comprises an electrocardiosignal preprocessing submodule, the electrocardiosignal preprocessing submodule is used for positioning QRS group waves by setting an adaptive threshold, and the calculation formula of the adaptive threshold is as follows:

wherein THR1 is a high threshold; mean (peak _ buffer) is the peak mean; peak is the peak of the signal detected at this moment; THR1_ lim is an empirical constant, represents the upper bound of threshold variation, and takes the value of 0.33; THR2 is low threshold; peak _ buffer is a buffer value that stores 8 consecutive peaks before the peak at that moment; THR2_ lim is also an empirical constant, representing the lower bound of threshold variation, with a value of 0.23.

5. The integrated CPR-defibrillation machine control system based on the human physiological model of claim 4, wherein: the main control module further comprises an arrhythmia identification submodule, and the arrhythmia identification submodule is used for extracting electrocardiosignal characteristics; selecting and dividing a data set to obtain a training set and a testing set; carrying out characteristic normalization processing on the data of the training set and the test set; training and testing a support vector machine model, and inputting the preprocessed electrocardiosignals into the support vector machine model for classification.

6. A control method of a cardio-pulmonary resuscitation and defibrillation integrated machine based on a human physiological model is characterized by comprising the following steps: use of the integrated cpr-defibrillation machine control system according to any one of claims 1 to 5, based on the physiological model of the human body, comprising the steps of:

step S1: acquiring electrocardiogram characteristic data and compression characteristic data of a patient;

step S2: calculating to obtain a cardiopulmonary feature result and grading levels corresponding to the BI value, the RI value and the BRI value according to the electrocardiogram feature data and the compression feature data;

and step S3: judging the cardio-pulmonary resuscitation effect and the fracture risk of the patient in the compression process according to the grading level, if BRI is less than 5, continuing to judge BI and RI, if BI is less than 4.5, gradually increasing the compression force until BI is greater than 4.5, and if RI is greater than 3.4, gradually decreasing the compression force until RI is less than 3.4; meanwhile, judging whether ventricular fibrillation occurs or not according to the cardio-pulmonary characteristic result, if yes, stopping pressing, and performing defibrillation;

and step S4: judging whether ventricular fibrillation still occurs according to the cardiopulmonary characteristic result in the defibrillation process, and if yes, continuing defibrillation.

7. The cardio-pulmonary resuscitation and defibrillation integrated machine control method based on the human physiological model according to claim 6, wherein: in step S3, the impedance control sub-step is specifically included:

step S31: establishing an impedance model;

step S32: calculating to obtain a reaction elasticity error value of the chest compression, wherein the calculation formula of the reaction elasticity error value is as follows:

ΔF(t)＝F _d (t)-F(t)

wherein, delta F (t) is the error value of the reaction elastic force, F _d (t) is a constant representing the desired force, F (t) is the reaction spring force, and the calculation of the reaction spring force includes the following three forms:

wherein, M _d As a quality parameter, B _d As a damping parameter, K _d In order to be a parameter of the stiffness,

is the actual acceleration of the end of the pressing head,

the actual speed of the pressing head end, X (t) is the position of the pressing head end, and>

for a desired acceleration of the pressing head end>

Desired speed of the pressing head end, X _d (t) is the desired position of the pressing head end;

step S33: inputting the error value of the reaction elasticity into an impedance model to obtain a frequency domain;

step S34: and filtering the reaction elastic force signal according to the frequency domain, and converting the filtered reaction elastic force signal into a corresponding motion signal of the tail end position.

8. The cardio-pulmonary resuscitation and defibrillation integrated machine control method based on the human physiological model according to claim 6, wherein: in step S2, the method specifically includes an electrocardiographic signal preprocessing substep: positioning QRS group waves by setting an adaptive threshold, wherein the adaptive threshold is calculated according to the following formula:

wherein THR1 is a high threshold; mean (peak _ buffer) is the peak mean; peak is the peak of the signal detected at this moment; THR1_ lim is an empirical constant, represents the upper bound of threshold variation and takes the value of 0.33; THR2 is low threshold; peak _ buffer is a buffer value for storing 8 continuous peak values before the peak value at the moment; TRH2_ lim is also an empirical constant, represents the lower bound of threshold variation, and takes the value of 0.23;

further comprising an arrhythmia identification sub-step: extracting electrocardiosignal characteristics; selecting and dividing a data set to obtain a training set and a testing set; carrying out characteristic normalization processing on the data of the training set and the test set; training and testing a support vector machine model, and inputting the preprocessed electrocardiosignals into the support vector machine model for classification.

Images