Defibrillation rhythm discrimination device

Abstract

The application relates to the technical field of medical equipment, in particular to a defibrillation rhythm identification method and device. The method comprises the following steps: preprocessing a collected first electrocardiosignal to obtain a second electrocardiosignal; judging whether the second electrocardiosignal is an interference signal based on a preset rule; if it is judged that the second electrocardiosignal is not an interference signal, then comprehensively judging whether the second electrocardiosignal is a non-defibrillation rhythm through a characteristic value weighting mode based on a first characteristic parameter of the second electrocardiosignal; if it is judged that the second electrocardiosignal is not a non-defibrillation rhythm, then comprehensively judging whether the second electrocardiosignal is a VF rhythm / VT rhythm through the characteristic value weighting mode based on a second characteristic parameter of the second electrocardiosignal, if yes, outputting a result of suggesting defibrillation; otherwise, outputting a result of not suggesting defibrillation.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to a defibrillation rhythm recognition device. Background Technology

[0002] Sudden cardiac death is mainly caused by ventricular fibrillation (VF) and ventricular tachycardia (VT). VF often occurs without warning; during an attack, the electrical activity of the ventricles loses synchrony, and the heart's pumping function is lost. If measures are not taken promptly to restore rhythm, sudden death will occur within minutes. Defibrillation, delivered as quickly as possible, is the only reliable and widely used method for ventricular fibrillation cardioversion in clinical practice.

[0003] Defibrillation is an important part of cardiopulmonary resuscitation (CPR). It involves using an AED (Automated External Defibrillator) to deliver a specific electrical current to the heart to restore a patient with ventricular fibrillation to a normal sinus rhythm.

[0004] Studies have shown that unnecessary defibrillation can adversely affect myocardial function, while failure to defibrillate in a timely manner can lead to missed opportunities to save a patient's life. Therefore, in practical applications, it is necessary to accurately determine the timing of defibrillation. However, currently, there is a lack of effective analytical mechanisms to grasp the timing of defibrillation, resulting in relatively blind defibrillation decisions and low defibrillation effectiveness. Summary of the Invention

[0005] This application provides a defibrillation rhythm recognition device to solve the problem of inaccurate defibrillation timing decision-making in the prior art.

[0006] In a first aspect, this application provides a method for identifying defibrillated rhythms, the method comprising:

[0007] The first electrocardiogram (ECG) signal was preprocessed to obtain the second ECG signal.

[0008] Based on preset rules, determine whether the second electrocardiogram signal is an interference signal;

[0009] If it is determined that the second ECG signal is not an interference signal, then based on the first characteristic parameter of the second ECG signal, a comprehensive judgment is made on whether the second ECG signal is a non-defibrillation rhythm by weighting the characteristic values.

[0010] If the second ECG signal is determined not to be a non-defibrillation rhythm, then based on the second characteristic parameter of the second ECG signal, a comprehensive judgment is made on whether the second ECG signal is a VF rhythm / VT rhythm by weighting the characteristic values. If it is, then a defibrillation recommendation is output; otherwise, a defibrillation recommendation is output.

[0011] Optionally, the method further includes:

[0012] If the second ECG signal is determined to be a non-defibrillation rhythm, then the output will indicate that defibrillation is not recommended.

[0013] Optionally, the step of determining whether the second electrocardiogram signal is an interference signal within a preset time period based on preset rules includes:

[0014] Set and start a timer with a period of T;

[0015] The magnitude of the second ECG signal within cycle T, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode pads are all detected in real time to ensure they meet preset requirements.

[0016] If the magnitude of the signal amplitude, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode all meet the preset requirements within the period T, then the second ECG signal is determined to be an interference signal; otherwise, the second ECG signal is determined to be an interference signal, the timer is reset, and the step of real-time detection of whether the magnitude of the signal amplitude, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode all meet the preset requirements within the period T is executed.

[0017] Optionally, the step of comprehensively determining whether the second electrocardiogram signal is a non-defibrillation rhythm based on the first characteristic parameter of the second electrocardiogram signal through a weighted eigenvalue method includes:

[0018] The second electrocardiogram signal is subjected to amplitude normalization processing to obtain the third electrocardiogram signal;

[0019] Extract the first feature parameters of the third electrocardiogram signal, and calculate the mean and variance of the first feature parameters respectively. The first feature parameters include the number of QRS waves, fluctuation frequency, peak amplitude, heart rate, and signal effective value.

[0020] Based on the first feature parameter and its corresponding mean and variance, and the preset weight values ​​for the first feature parameter and its corresponding mean and variance, a comprehensive calculation result is obtained by weighting the feature values. Based on the comprehensive calculation result, it is determined whether the third electrocardiogram signal is a non-defibrillation rhythm. The preset weight value for the first feature parameter used to characterize the defibrillation signal is negative, and the preset weight value for the first feature parameter used to characterize the non-defibrillation signal is set to positive.

[0021] Optionally, the step of comprehensively determining whether the second electrocardiogram signal is a VF rhythm / VT rhythm by weighting the feature values ​​based on the second feature parameters of the second electrocardiogram signal includes:

[0022] The third ECG signal is subjected to VF signal feature enhancement filtering to obtain the fourth ECG signal;

[0023] Extract the fluctuation frequency, fluctuation amplitude, and VF signal effective value of the fourth electrocardiogram signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VF signal effective value respectively;

[0024] Based on the fluctuation frequency, fluctuation amplitude, and effective value of the VF signal, along with their corresponding mean and variance, and pre-set weight values ​​for each of these parameters, a comprehensive calculation result is obtained through eigenvalue weighting. Based on this comprehensive calculation result, it is determined whether the fourth ECG signal represents a VF rhythm.

[0025] The third ECG signal is subjected to VT signal feature enhancement filtering to obtain the fifth ECG signal;

[0026] Extract the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters of the fifth ECG signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters respectively;

[0027] Based on the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters and their corresponding mean and variance, as well as the preset weight values ​​for the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters and their corresponding mean and variance, a comprehensive calculation result is obtained by weighting by feature values, and the fifth electrocardiogram signal is determined as a VT rhythm based on the comprehensive calculation result.

[0028] Optionally, if it is determined that the fourth ECG signal is not a VF rhythm and the fifth ECG signal is not a VT rhythm, then the method further includes:

[0029] Based on the fluctuation frequency, fluctuation amplitude, VF signal RMS value, and VT template periodic autocorrelation parameter and their corresponding mean and variance, and at least one set of weight values ​​preset for the fluctuation frequency, fluctuation amplitude, VF signal RMS value, and VT template periodic autocorrelation parameter and their corresponding mean and variance, a comprehensive calculation result is obtained by eigenvalue weighting, and a result of recommending defibrillation / not recommending defibrillation is output based on the comprehensive calculation result.

[0030] Secondly, this application provides a defibrillation rhythm recognition device, the device comprising:

[0031] The preprocessing unit is used to preprocess the acquired first electrocardiogram signal to obtain the second electrocardiogram signal;

[0032] The first judgment unit is used to determine whether the second electrocardiogram signal is an interference signal based on preset rules;

[0033] If the first judgment unit determines that the second electrocardiogram signal is not an interference signal, the second judgment unit is used to comprehensively determine whether the second electrocardiogram signal is a non-defibrillation rhythm based on the first characteristic parameter of the second electrocardiogram signal by weighting the characteristic values.

[0034] If the second judgment unit determines that the second electrocardiogram signal is not a non-defibrillation rhythm, the third judgment unit is used to comprehensively determine whether the second electrocardiogram signal is a VF rhythm / VT rhythm based on the second characteristic parameter of the second electrocardiogram signal by weighting the characteristic values.

[0035] If the third judgment unit determines that the second electrocardiogram signal is a VF rhythm / VT rhythm, the output unit is used to output a result suggesting defibrillation; otherwise, the output unit is used to output a result suggesting not to defibrillate.

[0036] Optionally, the device further includes:

[0037] If the third judgment unit determines that the second electrocardiogram signal is a non-defibrillation rhythm, the output unit is used to output a result that defibrillation is not recommended.

[0038] Optionally, when determining whether the second electrocardiogram signal is an interference signal within a preset time period based on preset rules, the first determination unit is specifically used for:

[0039] Set and start a timer with a period of T;

[0040] The magnitude of the second ECG signal within cycle T, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode pads are all detected in real time to ensure they meet preset requirements.

[0041] If the magnitude of the signal amplitude, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode all meet the preset requirements within the period T, then the second ECG signal is determined to be an interference signal; otherwise, the second ECG signal is determined to be an interference signal, the timer is reset, and the step of real-time detection of whether the magnitude of the signal amplitude, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode all meet the preset requirements within the period T is executed.

[0042] Optionally, when determining whether the second electrocardiogram (ECG) signal is a non-defibrillation rhythm based on the first characteristic parameter of the second ECG signal and by weighting the characteristic values, the second determination unit is specifically used for:

[0043] The second electrocardiogram signal is subjected to amplitude normalization processing to obtain the third electrocardiogram signal;

[0044] Extract the first feature parameters of the third electrocardiogram signal, and calculate the mean and variance of the first feature parameters respectively. The first feature parameters include the number of QRS waves, fluctuation frequency, peak amplitude, heart rate, and signal effective value.

[0045] Based on the first feature parameter and its corresponding mean and variance, and the preset weight values ​​for the first feature parameter and its corresponding mean and variance, a comprehensive calculation result is obtained by weighting the feature values. Based on the comprehensive calculation result, it is determined whether the third electrocardiogram signal is a non-defibrillation rhythm. The preset weight value for the first feature parameter used to characterize the defibrillation signal is negative, and the preset weight value for the first feature parameter used to characterize the non-defibrillation signal is set to positive.

[0046] Optionally, when determining whether the second electrocardiogram (ECG) signal is a VF / VT rhythm based on the second characteristic parameter of the second ECG signal using a weighted eigenvalue method, the third determination unit is specifically used for:

[0047] The third ECG signal is subjected to VF signal feature enhancement filtering to obtain the fourth ECG signal;

[0048] Extract the fluctuation frequency, fluctuation amplitude, and VF signal effective value of the fourth electrocardiogram signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VF signal effective value respectively;

[0049] Based on the fluctuation frequency, fluctuation amplitude, and effective value of the VF signal, along with their corresponding mean and variance, and pre-set weight values ​​for each of these parameters, a comprehensive calculation result is obtained through eigenvalue weighting. Based on this comprehensive calculation result, it is determined whether the fourth ECG signal represents a VF rhythm.

[0050] The third ECG signal is subjected to VT signal feature enhancement filtering to obtain the fifth ECG signal;

[0051] Extract the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters of the fifth ECG signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters respectively;

[0052] Based on the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters and their corresponding mean and variance, as well as the preset weight values ​​for the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters and their corresponding mean and variance, a comprehensive calculation result is obtained by weighting by feature values, and the fifth electrocardiogram signal is determined as a VT rhythm based on the comprehensive calculation result.

[0053] Optionally, if the second determination unit determines that the fourth ECG signal is not a VF rhythm, and the third determination unit determines that the fifth ECG signal is not a VT rhythm, then the device further includes:

[0054] The calculation unit is used to calculate a comprehensive calculation result by means of eigenvalue weighting based on the fluctuation frequency, fluctuation amplitude, VF signal RMS value and VT template periodic autocorrelation parameter and their respective mean and variance, and at least one set of weight values ​​preset for the fluctuation frequency, fluctuation amplitude, VF signal RMS value and VT template periodic autocorrelation parameter and their respective mean and variance, and output a result of recommending defibrillation / not recommending defibrillation based on the comprehensive calculation result.

[0055] Thirdly, embodiments of this application provide a defibrillation rhythm identification device, which includes:

[0056] Memory, used to store program instructions;

[0057] A processor is configured to invoke program instructions stored in the memory and execute the steps of the method as described in any one of the first aspects above, according to the obtained program instructions.

[0058] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the steps of the method as described in any of the first aspects above. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings of the embodiments of this application.

[0060] Figure 1 A detailed flowchart of a defibrillation rhythm identification method provided in this application embodiment;

[0061] Figure 2 A flowchart illustrating an interference signal identification method provided in this application embodiment;

[0062] Figure 3 A detailed flowchart of a non-defibrillation rhythm identification method provided for embodiments of this application;

[0063] Figure 4 A detailed flowchart of a VF / VT rhythm recognition method provided for embodiments of this application;

[0064] Figure 5 A schematic diagram of the structure of a defibrillation rhythm recognition device provided in an embodiment of this application;

[0065] Figure 6 This is a schematic diagram of the hardware architecture of a defibrillation rhythm recognition device provided in an embodiment of this application. Detailed Implementation

[0066] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “the,” and “the” as used in this application and claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to any and all possible combinations comprising one or more of the associated listed items.

[0067] It should be understood that although the terms first, second, third, etc., may be used to describe various information in embodiments of this application, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" may also be interpreted as "when," "when," or "in response to a determination."

[0068] For example, see Figure 1 The diagram shown is a detailed flowchart of a network configuration distribution method provided in an embodiment of this application. This method is applied to network nodes running a Neutron agent component and includes the following steps:

[0069] Step 100: Preprocess the acquired first electrocardiogram signal to obtain the second electrocardiogram signal.

[0070] In this embodiment of the application, preprocessing the first electrocardiogram (ECG) signal of the patient means filtering the collected ECG signal of the patient to obtain the second ECG signal.

[0071] Specifically, a preferred implementation involves a preprocessing stage for the ECG signal that includes three steps: a 50Hz filter, a 0.5Hz high-pass filter, and a 35Hz low-pass filter. Of course, to ensure the integrity of the ECG signal, the filter order should ideally not exceed 3.

[0072] Step 110: Based on preset rules, determine whether the second ECG signal is an interference signal.

[0073] In this embodiment of the application, when determining whether the second electrocardiogram (ECG) signal is an interference signal based on preset rules, a preferred implementation is to set and start a timer with a period of T; to detect in real time whether the signal amplitude, the amplitude change / duration, the baseline drift amplitude / duration, and the contact impedance of the electrode pads all meet preset requirements within the period T; wherein, if the signal amplitude, the amplitude change / duration, the baseline drift amplitude / duration, and the contact impedance of the electrode pads all meet the preset requirements within the period T, then the second ECG signal is determined not to be an interference signal; otherwise, the second ECG signal is determined to be an interference signal, the timer is reset, and the step of detecting in real time whether the signal amplitude, the amplitude change / duration, the baseline drift amplitude / duration, and the contact impedance of the electrode pads all meet the preset requirements is executed.

[0074] Specifically, in this embodiment, the difference between the real-time impedance and the average impedance can be calculated after low-pass filtering to obtain the contact impedance fluctuation of the electrode sheet. If the contact impedance fluctuation of the electrode sheet is greater than the normal range fluctuation, it is considered that a sudden situation has occurred, and it is determined that the contact impedance of the electrode sheet does not meet the preset requirements.

[0075] For example, if the contact resistance of the electrode pad is greater than the set first threshold when the electrode pad is touched or the patient's body position is moved, the electrode pad is considered to have poor contact or fall off; if the contact resistance of the electrode pad is less than the set second threshold, the electrode pad is considered to be short-circuited.

[0076] In this embodiment of the application, it can be determined whether the amplitude of the electrocardiogram signal is within a preset range, that is, whether the maximum amplitude of the electrocardiogram signal is greater than the maximum standard value and whether the minimum amplitude is less than the minimum standard value. If the maximum amplitude is greater than the maximum standard value and / or the minimum amplitude is less than the minimum standard value, it is determined that the signal amplitude does not meet the preset requirements.

[0077] For example, regarding the amplitude range of an electrocardiogram (ECG): the peak-to-peak value of the ECG signal is calculated every n seconds, and the maximum value among N sets of peak-to-peak values ​​within M seconds is calculated. The maximum peak-to-peak value is updated every n seconds. If the peak-to-peak value is greater than the normal ECG signal range of ±5mV, it is considered that a strong interference signal exists; if the peak-to-peak value is less than 0.1mV, it is less than the recognizable signal range required by national standards, and can be considered as cardiac arrest, ventricular fibrillation, etc., without the need for defibrillation treatment.

[0078] In this embodiment, based on the preprocessed signal, a 2-3 Hz low-pass filter is used to filter out the normal ECG signal to obtain the baseline, and then a high-pass filter is used to filter out the DC offset voltage to obtain the baseline drift signal. Similarly, if the amplitude of the baseline drift signal is greater than the set threshold, it can be considered that there is interference (e.g., attaching electrode pads, moving the patient).

[0079] The interference signal identification process provided in this application embodiment will be described in detail below with reference to specific application scenarios. For example, see [link to relevant documentation]. Figure 2 The diagram shown is a detailed flowchart of an interference signal identification method provided in an embodiment of this application. The method includes the following implementation process:

[0080] Power on, signal preprocessing; real-time calculation of baseline drift of ECG signal, maximum and minimum amplitude of ECG signal in 2 seconds, and low-voltage impedance of defibrillation path; comprehensive judgment: (1) Is the baseline drift greater than the threshold of the patient's resting state? (2) Is the maximum amplitude greater than the normal range? (3) Is the maximum amplitude less than the threshold of normal ECG signal? (4) Is the impedance greater than the normal high threshold? (5) Is the impedance less than the normal low threshold? If any of the above conditions are met, the ECG signal is determined to be an interference signal, and the process of real-time calculation of baseline drift of ECG signal, maximum and minimum amplitude of ECG signal in 2 seconds, and low-voltage impedance of defibrillation path is returned; otherwise, the ECG signal is used as the target ECG signal for rhythm recognition processing.

[0081] Step 120: If it is determined that the second ECG signal is not an interference signal, then based on the first characteristic parameter of the second ECG signal, a comprehensive judgment is made on whether the second ECG signal is a non-defibrillation rhythm by weighting the characteristic values.

[0082] In this embodiment of the application, when determining whether the second electrocardiogram (ECG) signal is a non-defibrillation rhythm based on the first characteristic parameters of the second ECG signal using a weighted eigenvalue method, a preferred implementation is to perform amplitude normalization processing on the second ECG signal to obtain a third ECG signal; extract the first characteristic parameters of the third ECG signal, and calculate the mean and variance of the first characteristic parameters respectively, wherein the first characteristic parameters include the number of QRS waves, fluctuation frequency, peak amplitude, heart rate, and signal effective value; based on the first characteristic parameters and their corresponding mean and variance, and the preset weight values ​​for the first characteristic parameters and their corresponding mean and variance, a comprehensive calculation result is obtained by weighting the eigenvalues, and the determination of whether the third ECG signal is a non-defibrillation rhythm is based on the comprehensive calculation result, wherein the preset weight value for the first characteristic parameter used to characterize the defibrillation signal is negative, and the preset weight value for the first characteristic parameter used to characterize the non-defibrillation signal is set to positive.

[0083] In practical applications, the normal range of electrocardiogram (ECG) signals is ±0.1mV to ±5mV. The core basis of heart rhythm recognition lies in the fact that different heart rhythm signals have different but relatively fixed characteristic values. These characteristic values ​​are mainly divided into temporal and amplitude parameters. Because the same heart rhythm signals, despite different amplitudes, exhibit significant similarities in waveform changes and shapes, normalizing ECG signals of different amplitudes to the same amplitude allows the characteristic values ​​of the same heart rhythm to possess essentially the same temporal and amplitude parameters, thus enabling recognition.

[0084] In the embodiments of the present application, the signal normalization algorithm is implemented using an automatic gain control algorithm. The gain control of the signal is based on the peak-to-peak value of the signal within the cardiac cycle. A normal electrocardiogram (ECG) signal has 5 peak feature points, namely P, Q, R, S, and T waves. Among them, the peak-to-peak amplitude of the QRS wave is the largest. The peak-to-peak value of the T wave is the largest in some leads, and the peak-to-peak values of other characteristic waveforms are relatively small. Therefore, the automatic gain control of the ECG signal is based on the maximum amplitude among numerous waveforms within one cardiac cycle. Since the period of the ECG signal is between 0.2 seconds and 2 seconds, the gain calculation period of the automatic gain control needs to satisfy the duration of 0.2 seconds to 2 seconds. To perform control in a timely manner, the peak-to-peak value of the device signal is calculated every n seconds, the maximum value among N groups of peak-to-peak values within M seconds is calculated, and the maximum peak-to-peak value is updated every n seconds, where 0.2 < n < 2 seconds. This can not only perform real-time updates for high-speed heart rhythms, such as a heart rate of 300 beats per minute (BPM), but also accurately calculate the amplitude of a heart rate with a 2-second cycle, that is, 30 BPM. In this way, the amplitude of the AGC (Automatic Gain Control) also meets the heart rate range of 30 BPM to 300 BPM.

[0085] Furthermore, to make the amplitude normalization algorithm have stability and avoid step-like mutations in the signal amplitude, the processing of N peak-to-peak values is not simply to take the maximum value. Instead, the newly obtained maximum value and the previous maximum value need to be subjected to IIR (Infinite Impulse Response) recursive filtering. In this way, it can be ensured that the gain change output by the AGC link is smooth.

[0086] It should be noted that in most cases, the QRS wave is a necessary and sufficient condition for identifying the need for defibrillation. Then, the presence of the QRS wave indicates the presence of normal ventricular contraction and relaxation. The signal of the QRS complex has obvious amplitude change characteristics within a specified time. The time width of the R wave in normal sinus rhythm is less than 0.2 seconds, and the spectral width is between 5 Hz and 25 Hz. The frequencies of other characteristic waveforms, such as the P and T waves, are much lower than these frequencies. By designing a QRS wave differential filter, the QRS wave can be simply and efficiently filtered out. The QRS wave differential filter is implemented based on a high-order differential filter with respect to duration. The differential filter is actually a high-pass filter, and the cut-off frequency is related to the sampling rate. By selecting an appropriate delay n for high-order differentiation, the QRS complex can be simply and efficiently extracted without phase distortion.

[0087] The QRS complex exhibits a certain time periodicity, and despite arrhythmias, this periodicity remains unaffected. A QRS complex window detector can be designed to eliminate noise and filter out non-QRS signals. This detector uses a weighted time window function based on the QRS complex duration to filter the signal, resulting in a relatively clean QRS complex. To enhance the QRS characteristics, an exponential transform method is used to nonlinearly amplify the output of the QRS complex window detector, strengthening the QRS amplitude and attenuating other low-frequency ECG signals and noise. By setting appropriate thresholds, QRS complexes of normal sinus rhythm and ventricular rhythm can be selected.

[0088] Signals containing periodic sinus and low-rate ventricular QRS complexes can be identified as non-defibrillation signals.

[0089] The QRS wave signal processed by nonlinear exponential transformation is used with an autoregressive peak detection algorithm. The three periodic points of the QRS wave group (start, extreme point, and end point) are used to obtain the real-time RR interval values ​​of three cardiac cycles. The real-time heart rate and the average heart rate of five cardiac cycles are calculated. The fluctuation variance of the three RR intervals is also calculated for subsequent judgment and analysis.

[0090] Due to respiratory and other factors, the amplitude of the QRS complex varies with the heartbeat in the same individual. Furthermore, the presence of premature beats and escape rhythms, as well as abnormal rhythms such as premature contractions, PVCs, and SVPs, makes it crucial to employ sophisticated algorithms to detect and evaluate accurate QRS complex detection and calculate related parameters. Therefore, based on the unique temporal and amplitude characteristics of specific arrhythmic heart rates, specialized algorithms are designed to extract these characteristic parameters to identify these specific non-defibrillation arrhythmic rhythms. These characteristic parameters are consensus-based standard ECG identification parameters, which will not be elaborated upon in this embodiment.

[0091] Furthermore, non-defibrillated rhythms exhibit consistent patterns of change in both time and amplitude. Algorithms for extracting signal fluctuation parameters and amplitude variation parameters were designed concurrently with QRS complex detection. These parameters, combined with information from the QRS complex, can accurately identify non-defibrillated rhythms.

[0092] In non-defibrillation rhythms, the changes in signal frequency and amplitude within an RR interval are relatively consistent with statistical parameters. VF, on the other hand, shows the opposite trend, while VT goes to the other extreme. Furthermore, arrhythmic rhythms also possess unique characteristics in terms of both time and amplitude.

[0093] High-rate ventricular tachycardia (VT) and complex ventricular fibrillation (VF) are often easily confused with arrhythmias, leading to incorrect identification results. Extracting time and amplitude variation information and combining it with QRS feature information can significantly improve the identification rate of complex VF and high-tachycardia hi-VT. It can also reduce the false identification rate of non-defibrillated rhythms.

[0094] In this embodiment, the acquired electrocardiogram (ECG) signal is preprocessed and then filtered through a narrow-bandwidth bandpass filter. Although the output result has phase distortion, the zero-crossing and amplitude characteristics of the signal are completely preserved. Through simple windowed zero-crossing detection, the time interval sequence of all signals crossing the baseline within the cardiac cycle is calculated, as well as the amplitude of the waveform above or below each baseline. Then, the slope, absolute height, mean, and variance of the waveform changes within the cardiac cycle or a certain time period, along with the time interval and its mean and variance, are calculated as a basis for comprehensively judging special heart rhythms.

[0095] For example, see Figure 3 The diagram shown is a detailed flowchart of a non-defibrillation rhythm identification method provided in an embodiment of this application. An automatic gain control algorithm based on the cardiac cycle is used to normalize the ECG signal. The first feature extraction method involves sequentially filtering the normalized ECG signal using a QRS complex filter, a QRS complex weighted window filter, a clutter compression and suppression converter, and a QRS detector to obtain some first feature parameters. The second first feature parameter extraction method involves bandpass filtering the normalized ECG signal and calculating frequency parameters based on a time window trigger, as well as peak-to-peak value and slope parameters based on an amplitude window, to obtain some first feature parameters. Based on a preset feature value for a specific heart rhythm, some first feature parameters are obtained. Then, based on the obtained first feature parameters and preset weight values ​​for each first feature, feature weighting is performed to determine whether the ECG signal is a conventional non-defibrillation rhythm. If so, a result indicating that defibrillation is not recommended is output; otherwise, feature weighting is performed based on the obtained first feature parameters and preset specific weight values ​​for each first feature parameter to determine whether the ECG signal is a typical non-defibrillation rhythm. If so, a result indicating that defibrillation is not recommended is output; otherwise, defibrillation rhythm identification begins.

[0096] It should be noted that in the embodiments of this application, the pre-set weight values ​​for each first / second characteristic parameter, whether large or small, positive or negative, are based on the results obtained from multiple experiments. If a characteristic parameter represents a certain judgment result to a greater degree, then when judging whether the corresponding electrocardiogram signal is the above judgment result, the pre-set weight value for that parameter is positive, and the larger the weight value, the better.

[0097] For example, identifying whether a collected electrocardiogram (ECG) signal is a non-defibrillation rhythm requires using the total multiple feature parameters calculated in the preceding steps. Here, a fuzzy decision-making method using neural networks is employed. Different feature parameters have different influencing factors on the decision.

[0098] Stable waveform changes and periodicity, along with the presence of normal QRS complexes, have the highest weighting value, Qqrs, for identifying non-defibrillated rhythms. Other signals are assigned different weighting values, such as fluctuation frequency Qf, mean frequency Qfm, and variance Qfd; peak amplitude Qa, mean Qam, and variance Qad; heart rate Qr, mean heart rate Qrm, and spurious variance Qrd; effective mean signal Qv, effective variance Qvd, etc. Different weighting values ​​are assigned to input signals for identification as normal rhythms, sinus arrhythmias, VT, and VF.

[0099] Parameters that favor defibrillation signals are weighted negatively; those that favor non-defibrillation signals are weighted positively; and irrelevant parameters are weighted 0.

[0100] Therefore, the non-defibrillation rhythm identification algorithm is as follows:

[0101] Jnd = Cqrs*Qqrs + Fzc*Qf + Fzcm*Qfm + Fzcd*Qfd + …

[0102] In the above formula,

[0103] Cqrs is the number of QRSs present. The count is incremented by 1 for each QRS group detected.

[0104] Fzc is the signal frequency for zero-crossing detection, Fzcm is the frequency mean, and Fzcd is the frequency variance.

[0105] The commands and methods for calculating amplitude and other feature information are applied similarly. If the overall calculation result is greater than 0, it is judged as a non-defibrillation rhythm. If it is less than 0, the calculation continues. The calculation formulas for other special rhythms such as PVC, SVD, BB, etc., which are prone to identification errors, are similar, but the feature parameters have different weights. If the judgment of non-defibrillation rhythm does not meet the criteria, the defibrillation rhythm identification process begins.

[0106] Arrhythmias that do not require defibrillation, especially sinus rhythms and arrhythmias with a heart rate greater than 200 BPM, are easily identified as VT due to the very short interval between the P and T waves, or because the T waves are enlarged or even reversed due to myocardial ischemia. This is because the influence (weighting) of certain characteristic values ​​of these special rhythms on the regular rhythm is opposite. Therefore, new calculation formulas need to be designed for the rhythm signals of these special states.

[0107] If the conventional judgment result is positive, it is directly identified as a non-defibrillation rhythm, and the rhythm recognition process ends; otherwise, the weighted judgment value of the typical rhythm is calculated. If it is positive, it continues to be identified as a non-defibrillation rhythm, and the rhythm recognition process ends; otherwise, the defibrillation rhythm recognition algorithm is entered.

[0108] Step 130: If it is determined that the second ECG signal is not a non-defibrillation rhythm, then based on the second characteristic parameter of the second ECG signal, a comprehensive judgment is made on whether the second ECG signal is a VF rhythm / VT rhythm by weighting the characteristic values. If it is, then a defibrillation recommendation is output; otherwise, a defibrillation recommendation is output.

[0109] In this embodiment, when determining whether the second ECG signal is a VF / VT rhythm based on the second feature parameter of the second ECG signal using a feature value weighting method, a preferred implementation is to perform VF signal feature enhancement filtering on the third ECG signal to obtain a fourth ECG signal; extract the fluctuation frequency, fluctuation amplitude, and VF signal effective value of the fourth ECG signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VF signal effective value respectively; based on the fluctuation frequency, fluctuation amplitude, and VF signal effective value and their corresponding mean and variance... The variance, and the preset weights for the fluctuation frequency, fluctuation amplitude, and VF signal effective value, and their corresponding mean and variance, are calculated using a feature-weighted method to obtain a comprehensive calculation result. Based on the comprehensive calculation result, it is determined whether the fourth ECG signal is a VF rhythm. The third ECG signal is then subjected to VT signal feature enhancement filtering to obtain a fifth ECG signal. The fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters of the fifth ECG signal are extracted, and the mean and variance of these parameters are calculated respectively. Based on the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters, and their corresponding mean and variance, and the preset weights for these parameters, a comprehensive calculation result is obtained using a feature-weighted method. Based on the comprehensive calculation result, it is determined whether the fifth ECG signal is a VT rhythm.

[0110] Furthermore, if it is determined that the fourth ECG signal is not a VF rhythm and the fifth ECG signal is not a VT rhythm, then the defibrillation rhythm identification method further includes the following steps:

[0111] Based on the fluctuation frequency, fluctuation amplitude, VF signal RMS value, and VT template periodic autocorrelation parameter and their corresponding mean and variance, and at least one set of weight values ​​preset for the fluctuation frequency, fluctuation amplitude, VF signal RMS value, and VT template periodic autocorrelation parameter and their corresponding mean and variance, a comprehensive calculation result is obtained by eigenvalue weighting, and a result of recommending defibrillation / not recommending defibrillation is output based on the comprehensive calculation result.

[0112] For example, see Figure 4 The diagram shows a detailed flowchart of a VF / VT rhythm recognition method provided in an embodiment of this application. Specifically, the recognition of the defibrillation rhythm begins. The VF rhythm recognition process is as follows: VF signal feature enhancement filtering, frequency parameter calculation based on a time window, peak-to-peak parameter calculation based on an amplitude window, calculation based on multiple parameters of the effective value, and weighted feature value determination to determine whether it is a VF rhythm. If so, a defibrillation recommendation is output; otherwise, feature value screening and recognition for special VF rhythms are performed to determine whether it is a special VF rhythm. If so, a defibrillation recommendation is output; otherwise, a defibrillation recommendation is output. The VT rhythm recognition process is as follows: VT signal feature enhancement filtering, frequency parameter calculation based on a time window, peak-to-peak parameter calculation based on an amplitude window, calculation based on periodic autocorrelation parameters, and weighted feature value determination to determine whether it is a VT rhythm. If so, a defibrillation recommendation is output; otherwise, feature value screening and recognition for special VF rhythms are performed to determine whether it is a special VF rhythm. If so, a defibrillation recommendation is output; otherwise, a defibrillation recommendation is output.

[0113] For example, in practical applications, only two heart rhythms currently require defibrillation: VF and high-rate VT. These two rhythms are very different from sinus rhythm and harmless periodic ventricular rhythms. A typical VT rhythm is a regular, high-amplitude monophasic or biphasic ventricular contraction and diastole, while a typical VF rhythm is a high-frequency, continuous signal with irregular low-to-medium amplitude fluctuations in time and amplitude. Complex VF waves are formed by interspersing regular, but lower-frequency, VT-like single or double waves within the typical VF signal, or VF ventricular fibrillation interspersed with brief ventricular flutter signals.

[0114] Therefore, filters with special bandwidths are designed based on VF and VT to enhance the characteristics of VF and VT signals. Then, the frequency and amplitude of the output of the two filters, the effective value of the VF signal, the periodic autocorrelation of the VT template, and the mean and variance of the four data are calculated respectively.

[0115] For VF signals, the frequency range is between 200 and 600 BMP, and the variance of the signal frequency and amplitude remains large. After processing with a conventional QRS duration window function low-pass filter, the calculated output values ​​of non-defibrillated rhythms and VT signals are relatively small, while the output of VF signals is large and has a large variance, thus serving as an important parameter for identifying VF. For sinus non-defibrillated rhythms, the number of fluctuations in the signal within a 2-second cycle differs greatly from that of VF signals, but the variance of the peak-to-peak amplitude is large in both cases. Therefore, frequency and the number of fluctuations are also key parameters for distinguishing VF.

[0116] VT signals have a periodic, roughly similar shape, but because they are both ventricular waves, the amplitude of the output signal after QRS feature extraction is very small. This is a key parameter for distinguishing non-defibrillated rhythms from VT. The signal after VT filtering is similar to ventricular flutter signals. Therefore, after normalizing the amplitude and period to an equal-height, equal-duration template for autocorrelation, the output data shows a very clear difference between non-defibrillated rhythms and VF.

[0117] Similar to the non-defibrillation rhythm identification methods described above, VT and VF identification also calculates the result by weighting different feature parameters and determining whether it is greater than 0. The calculation of VT and VF identification values ​​also uses weighted values ​​of some feature values ​​calculated in the non-defibrillation rhythm identification algorithm.

[0118] In addition, VF also has special and complex waveforms. Conventional recognition weighting schemes calculate positive values, causing recognition errors.

[0119] Special VF signals typically exhibit both classic low-amplitude, high-frequency ECG signals and high-amplitude, sinusoidal window-like amplitude-weighted signals, displaying alternating patterns of regularity and irregularity. Alternatively, the frequency may fluctuate between high and low amplitude, with both frequency and amplitude being disordered, followed by distinct VT signals, or even low-frequency, high-amplitude oscillations. These signals are neither standard VF nor standard VT, and certainly not a high-speed normal heart rhythm. According to the standard VF and VT calculation formulas, the results are positive, easily identifying them as non-defibrillation rhythms. Therefore, these complex defibrillation rhythms require defining additional characteristic weights for comprehensive calculation.

[0120] If the weighted average defibrillation index for conventional VT or VF is negative, the analysis and identification can end, and a defibrillation recommendation can be output. Otherwise, continue to calculate the weighted defibrillation index for special VF. If it is negative, a defibrillation recommendation conclusion can be output; otherwise, a defibrillation recommendation conclusion can be output.

[0121] Furthermore, in this embodiment, after the heart rhythm recognition algorithm starts, considering the initialization and output stability of various filters, and given that the slowest heart rate is 30 BMP (approximately 2 seconds), the algorithm can generally output a relatively reliable recognition result after 4 seconds. Considering the possibility of self-recovery in VT and VF, or the presence of sinus R waves or pacemaker stimulation waveforms, a recognition time of 6-9 seconds is deemed more reliable.

[0122] Based on the same inventive concept as the above-described method embodiments, see, for example, the following: Figure 5 The diagram shown is a structural schematic of a defibrillation rhythm recognition device provided in an embodiment of this application. The device includes:

[0123] The preprocessing unit 50 is used to preprocess the acquired first electrocardiogram signal to obtain the second electrocardiogram signal;

[0124] The first judgment unit 51 is used to determine whether the second electrocardiogram signal is an interference signal based on preset rules;

[0125] If the first judgment unit 51 determines that the second electrocardiogram signal is not an interference signal, the second judgment unit 52 is used to comprehensively determine whether the second electrocardiogram signal is a non-defibrillation rhythm based on the first characteristic parameter of the second electrocardiogram signal by weighting the characteristic values.

[0126] If the second judgment unit 52 determines that the second electrocardiogram signal is not a non-defibrillation rhythm, the third judgment unit 53 is used to comprehensively determine whether the second electrocardiogram signal is a VF rhythm / VT rhythm based on the second characteristic parameter of the second electrocardiogram signal by weighting the characteristic values.

[0127] If the third judgment unit 53 determines that the second electrocardiogram signal is a VF rhythm / VT rhythm, the output unit 54 is used to output a result suggesting defibrillation; otherwise, the output unit 54 is used to output a result suggesting not to defibrillate.

[0128] Optionally, the device further includes:

[0129] If the third judgment unit 53 determines that the second electrocardiogram signal is a non-defibrillation rhythm, then the output unit 54 is used to output the result that defibrillation is not recommended.

[0130] Optionally, when determining whether the second electrocardiogram signal is an interference signal within a preset time period based on preset rules, the first determination unit 51 is specifically used for:

[0131] Set and start a timer with a period of T;

[0132] The magnitude of the second ECG signal within cycle T, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode pads are all detected in real time to ensure they meet preset requirements.

[0133] If the magnitude of the signal amplitude, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode all meet the preset requirements within the period T, then the second ECG signal is determined to be an interference signal; otherwise, the second ECG signal is determined to be an interference signal, the timer is reset, and the step of real-time detection of whether the magnitude of the signal amplitude, the amplitude / duration of the signal amplitude change, the amplitude / duration of the baseline drift, and the contact impedance of the electrode all meet the preset requirements within the period T is executed.

[0134] Optionally, when determining whether the second electrocardiogram (ECG) signal is a non-defibrillation rhythm based on the first characteristic parameter of the second ECG signal and by weighting the characteristic values, the second determination unit 52 is specifically used for:

[0135] The second electrocardiogram signal is subjected to amplitude normalization processing to obtain the third electrocardiogram signal;

[0136] Extract the first feature parameters of the third electrocardiogram signal, and calculate the mean and variance of the first feature parameters respectively. The first feature parameters include the number of QRS waves, fluctuation frequency, peak amplitude, heart rate, and signal effective value.

[0137] Based on the first feature parameter and its corresponding mean and variance, and the preset weight values ​​for the first feature parameter and its corresponding mean and variance, a comprehensive calculation result is obtained by weighting the feature values. Based on the comprehensive calculation result, it is determined whether the third electrocardiogram signal is a non-defibrillation rhythm. The preset weight value for the first feature parameter used to characterize the defibrillation signal is negative, and the preset weight value for the first feature parameter used to characterize the non-defibrillation signal is set to positive.

[0138] Optionally, when determining whether the second electrocardiogram (ECG) signal is a VF / VT rhythm based on the second characteristic parameter of the second ECG signal using a weighted eigenvalue method, the third determination unit is specifically used for:

[0139] The third ECG signal is subjected to VF signal feature enhancement filtering to obtain the fourth ECG signal;

[0140] Extract the fluctuation frequency, fluctuation amplitude, and VF signal effective value of the fourth electrocardiogram signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VF signal effective value respectively;

[0141] Based on the fluctuation frequency, fluctuation amplitude, and effective value of the VF signal, along with their corresponding mean and variance, and pre-set weight values ​​for each of these parameters, a comprehensive calculation result is obtained through eigenvalue weighting. Based on this comprehensive calculation result, it is determined whether the fourth ECG signal represents a VF rhythm.

[0142] The third ECG signal is subjected to VT signal feature enhancement filtering to obtain the fifth ECG signal;

[0143] Extract the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters of the fifth ECG signal, and calculate the mean and variance of the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters respectively;

[0144] Based on the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters and their corresponding mean and variance, as well as the preset weight values ​​for the fluctuation frequency, fluctuation amplitude, and VT template period autocorrelation parameters and their corresponding mean and variance, a comprehensive calculation result is obtained by weighting by feature values, and the fifth electrocardiogram signal is determined as a VT rhythm based on the comprehensive calculation result.

[0145] Optionally, if the second determination unit 52 determines that the fourth ECG signal is not a VF rhythm, and the third determination unit 53 determines that the fifth ECG signal is not a VT rhythm, then the device further includes:

[0146] The calculation unit is used to calculate a comprehensive calculation result by means of eigenvalue weighting based on the fluctuation frequency, fluctuation amplitude, VF signal RMS value and VT template periodic autocorrelation parameter and their respective mean and variance, and at least one set of weight values ​​preset for the fluctuation frequency, fluctuation amplitude, VF signal RMS value and VT template periodic autocorrelation parameter and their respective mean and variance, and output a result of recommending defibrillation / not recommending defibrillation based on the comprehensive calculation result.

[0147] These units can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more digital signal processors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs). Alternatively, when one of these units is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these units can be integrated together to form a system-on-a-chip (SOC).

[0148] Furthermore, regarding the defibrillation rhythm recognition device provided in this application embodiment, from a hardware perspective, the hardware architecture schematic diagram of the defibrillation rhythm recognition device can be found in [reference needed]. Figure 6 As shown, the defibrillation rhythm recognition device may include: a memory 60 and a processor 61.

[0149] The memory 60 is used to store program instructions; the processor 61 calls the program instructions stored in the memory 60 and executes the above-described method embodiment according to the obtained program instructions. The specific implementation and technical effects are similar, and will not be described in detail here.

[0150] Optionally, this application also provides a defibrillation rhythm recognition device, including at least one processing element (or chip) for performing the above method embodiments.

[0151] Optionally, this application also provides a program product, such as a computer-readable storage medium storing computer-executable instructions for causing the computer to perform the above-described method embodiments.

[0152] Here, a machine-readable storage medium can be any electronic, magnetic, optical, or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, a machine-readable storage medium can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drives (such as hard disk drives), solid-state drives, any type of storage disk (such as optical discs, DVDs, etc.), or similar storage media, or combinations thereof.

[0153] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer, which can take the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email sending and receiving device, game console, tablet computer, wearable device, or any combination of these devices.

[0154] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0155] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0156] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0157] Furthermore, these computer program instructions can also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in the process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0158] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0159] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A defibrillation rhythm recognition device, characterized in that, The device includes: The preprocessing unit (50) is used to preprocess the acquired first electrocardiogram signal through three steps: 50Hz filtering, 0.5Hz high-pass filtering, and 35Hz low-pass filtering, to obtain the second electrocardiogram signal. The first judgment unit (51) is used to detect in real time whether the magnitude of the signal amplitude of the second ECG signal within the period T, the change amplitude / duration of the signal amplitude, the baseline drift amplitude / duration and the contact impedance of the electrode sheet all meet the preset requirements, and determine that the second ECG signal is not an interference signal. Otherwise, it determines that the second ECG signal is an interference signal, resets the timer, and repeats the above detection to see if all preset requirements are met until it is determined that the second ECG signal is not an interference signal. The second judgment unit (52) is used for: An automatic gain control algorithm based on the cardiac cycle is used to normalize the electrocardiogram signal and obtain a third electrocardiogram signal. Method 1 for extracting the first feature parameters of the third electrocardiogram signal: The normalized electrocardiogram signal is filtered sequentially by a QRS group filter, a QRS group weighted window filter, a clutter compression and suppression converter, and a QRS wave detector to obtain some of the first feature parameters. Method 2 for extracting the first feature parameters of the third ECG signal: Bandpass filtering is performed on the normalized ECG signal, and frequency parameters are calculated based on the time window trigger, as well as peak-to-peak value and slope parameters are calculated based on the amplitude window to obtain some first feature parameters; Method 3 for extracting the first feature parameters from the third electrocardiogram signal: Based on the feature value screening and identification of a preset special heart rhythm, some first feature parameters are obtained; Based on the first feature parameters obtained above and the preset conventional weight values ​​for each first feature, feature weighting is performed to determine whether the ECG signal is a conventional non-defibrillation rhythm. If so, the result of not recommending defibrillation is output; otherwise, based on the first feature parameters obtained above and the preset specific weight values ​​for each first feature parameter, feature weighting is performed to determine whether the ECG signal is a typical non-defibrillation rhythm. If so, the result of not recommending defibrillation is output; otherwise, defibrillation rhythm identification is started. The third judgment unit (53) is used to perform VF and VT signal enhancement processing on the third electrocardiogram signal respectively if the second judgment unit (52) determines that the second electrocardiogram signal is not a non-defibrillation rhythm. The third judgment unit (53) is used to extract the second feature parameter of the fourth electrocardiogram signal and comprehensively judge whether the second electrocardiogram signal is a VF rhythm by weighted summing of the second feature parameter. The third feature parameter of the fifth electrocardiogram signal is extracted and comprehensively judge whether the second electrocardiogram signal is a VT rhythm by weighted summing of the third feature parameter. The output unit (54) is used to output a defibrillation recommendation result if the third judgment unit (53) determines that the second electrocardiogram signal is a VF rhythm / VT rhythm; otherwise, the output unit (54) is used to output a defibrillation recommendation result; wherein, the first characteristic parameter includes: the number of QRS waves, the fluctuation frequency, the peak amplitude, the heart rate, and the effective value of the signal, and the mean and variance of the number of QRS waves, the fluctuation frequency, the peak amplitude, the heart rate, and the effective value of the signal are calculated respectively; The second characteristic parameters include: fluctuation frequency, fluctuation amplitude, and effective value of VF signal, and the mean and variance of the fluctuation frequency, fluctuation amplitude, and effective value of VF signal are calculated respectively. The third feature parameter extraction includes: fluctuation frequency, fluctuation amplitude, and VT template periodic autocorrelation parameter, and the mean and variance of the fluctuation frequency, fluctuation amplitude, and VT template periodic autocorrelation parameter are calculated respectively.

2. The defibrillation rhythm recognition device as described in claim 1, characterized in that: The preprocessing unit (50) performs three steps on the first electrocardiogram signal: 50Hz filtering, 0.5Hz high-pass filtering, and 35Hz low-pass filtering, and outputs the second electrocardiogram signal. The order of the filter does not exceed 3.

3. The defibrillation rhythm recognition device as described in claim 1, characterized in that: The first judgment unit (51) is also used to perform the following step-by-step condition detection on the second electrocardiogram signal within the T cycle: (1) If the contact resistance of the electrode sheet is greater than the first threshold, the electrode sheet is in poor contact and the electrode sheet falls off; if the contact resistance of the electrode sheet is less than the second threshold, the electrode sheet is considered to be short-circuited. (2) Calculate the maximum value among the N peak-to-peak values ​​within 2 seconds, and update the maximum peak-to-peak value every n seconds; if the peak-to-peak value is greater than the normal ECG signal ±5mV range, it is judged that there is a strong interference signal; if the peak-to-peak value is less than 0.1mV, it is less than the recognizable signal range required by the national standard, and it is considered that the heart stops beating or the ECG is fibrillating, and there is no need for defibrillation treatment. (3) Use a 2~3Hz low-pass filter to filter out the normal ECG signal to obtain the baseline, and then use a high-pass filter to filter out the DC offset voltage to obtain the baseline drift signal. If the amplitude of the baseline drift signal is greater than the set threshold, it is considered that there is interference. If any of the above conditions are met, the ECG signal is determined to be an interference signal. The T timer is reset, and interference detection continues until no interference occurs at the end of the T cycle.

4. The defibrillation rhythm recognition device as described in claim 1, characterized in that, The third judgment unit (53) is also used for: (1) Based on the second ECG signal, VF signal feature enhancement filtering is performed to obtain the fourth ECG signal, and the second feature parameters are extracted. Based on the second feature parameters and the corresponding weights, a comprehensive judgment is made on whether the fourth ECG signal is a VF rhythm by weighted summation. (2) Based on the third ECG signal, the VT signal feature enhancement filtering process is performed to obtain the fifth ECG signal, and the third feature parameter is extracted. Based on the third feature parameter and the corresponding weight, the weighted summation method is used to comprehensively determine whether the fifth ECG signal is a VT rhythm. (3) If the fourth ECG signal is determined not to be a VF rhythm, then VF identification is performed based on the special VF rhythm feature value.

Images