A group electrocardiogram arrhythmia diagnosis method

Abstract

The present invention discloses a method for diagnosing arrhythmia in a group electrocardiogram, comprising: obtaining an electrocardiogram signal; identifying an R(i) wave in the electrocardiogram signal and determining the position of the R(i) wave; calculating the RR(i) interval time; identifying the time limit characteristic parameters of the QRS(i) wave group; determining the frequency characteristic parameter F(i) of the R(i) wave; identifying the P(i) wave and the position of the P(i) wave; identifying the regularity characteristic parameters of the P wave of the electrocardiogram signal according to the position of the P(i) wave; determining the rhythm characteristic parameters of the electrocardiogram signal; and outputting the arrhythmia diagnosis result. The present invention realizes accurate identification of the R(i) wave and the P(i) wave based on the phasor transformation method, and determines whether the time limit of the QRS(i) wave group is wide and deformed through a reliable area integration method, and combines the identification of other pathological characteristics of the electrocardiogram signal to realize reliable and practical diagnosis of arrhythmia in a group electrocardiogram.

Description

Technical Field

[0001] The invention belongs to the technical field of arrhythmia diagnosis, and in particular relates to a group electrocardiogram arrhythmia diagnosis method. Background Art

[0002] In physical training or physical fitness testing scenarios for groups such as soldiers and students, MLII equivalent limb leads are often used for group ECG monitoring. MLII equivalent limb leads are improved limb leads II, which are obtained by placing electrodes on the chest of the subject. They have the advantages of easy sticking, wireless transmission, and batch collection, and are particularly suitable for group screening of cardiac function.

[0003] Unlike clinical ECG, group ECG usually has only one signal channel, the MLII equivalent limb lead, and is easily affected by many factors such as the measurement environment and improper operation, resulting in a lot of power frequency interference and myoelectric interference in the ECG signal data. In addition, most of the mainstream arrhythmia analysis methods based on deep learning technology are based on clinical ECG data, relying on a large amount of clinical ECG data as training and testing when building deep learning algorithm models. In addition, the existing deep learning network models for arrhythmia analysis are immature and imperfect, so they are not suitable for the diagnosis of group ECG arrhythmias. Therefore, in the field of group ECG monitoring and arrhythmia diagnosis, practical and reliable group ECG arrhythmia analysis methods are urgently needed to be developed for screening and diagnosis of group arrhythmias. Summary of the invention

[0004] The object of the present invention is to overcome one or more deficiencies of the prior art and to provide a method for diagnosing arrhythmia using electrocardiograms of a group of people.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] A method for diagnosing arrhythmia by electrocardiogram of a group comprises the following steps:

[0007] Obtain ECG signals;

[0008] Identifying an R(i) wave in an electrocardiogram signal and determining a position of the R(i) wave;

[0009] Calculate the RR(i) interval time according to the position of the R(i) wave, wherein the RR(i) interval time is the distance between the current R(i) wave position and the R(i-1) wave position, where i≥2;

[0010] Perform time domain integration on each R(i) wave according to the preset integration limit to obtain the integral area AUC(i) corresponding to each R(i) wave;

[0011] Calculate the mean med(AUCs) of all integrated areas AUC(i);

[0012] Determine the time characteristic parameters of the QRS(i) complex where the R(i) wave is located according to the average value med(AUCs) and the integrated area AUC(i) of the R(i) wave;

[0013] Determine the frequency characteristic parameter F(i) of the R(i) wave;

[0014] If the time characteristic parameter of the QRS(i) complex is that the time of the QRS(i) complex is normal, the position of the theoretical P(i) wave is calculated according to the position of the R(i) wave, the position of the R(i-1) wave and the RR(i) interval in the QRS(i) complex, the position of the theoretical P(1) wave is calculated according to the position of the R(1) wave, and the position of the theoretical T(i-1) wave is calculated according to the position of the R(i-1) wave and the RR(i-1) interval, where i≥2;

[0015] If the position of the theoretical P(i) wave does not overlap with the position of the theoretical T(i-1) wave, then the phasor transformation of the theoretical P(i) wave in the ECG signal is performed;

[0016] Find the maximum value U in the theoretical P(i) wave after phase transformation P (i), and according to the maximum value U P (i) and the measured current R(i) amplitude U R (i) Identify the true P(i) wave and the position of the true P(i) wave;

[0017] Identify the regular characteristic parameters of the P wave of the ECG signal according to the position of the real P(i) wave;

[0018] Determine the rhythm characteristic parameters of the ECG signal;

[0019] According to the time characteristic parameters, frequency characteristic parameters F(i), P wave regularity characteristic parameters and rhythm characteristic parameters of the QRS(i) wave group, it is judged whether the electrocardiogram signal has the first type of arrhythmia. If so, a first arrhythmia diagnosis result containing the first type of arrhythmia is output. If not, a second arrhythmia diagnosis result of a normal electrocardiogram is output. The first type of arrhythmia includes one or more of sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, supraventricular tachycardia and ventricular tachycardia.

[0020] Preferably, the identifying of the R(i) wave in the electrocardiogram signal and determining the position of the R(i) wave comprises the following sub-steps:

[0021] The ECG signal is digitally filtered using a Hamming window-based FIR bandpass filter;

[0022] Perform phasor transformation on the digitally filtered electrocardiogram (ECG) signal to obtain the phasor signal PT(n) corresponding to the ECG signal, where , is the instantaneous value of the ECG signal, 0 < Rv < 1, Rv represents the waveform enhancement degree of the ECG signal, and the complex domain transformation formula of the ECG signal is ;

[0023] Detect the maximum value Max(i) in the phasor signal PT(n) through a sliding window with a first preset time width. If the maximum value Max(i) is greater than the adaptive threshold Thr(i), then define the position of this Max(i) as the position of the R(i) wave;

[0024] where the adaptive threshold Thr(i) is twice the standard deviation of the phasor signal PT(n) calculated in a moving window with a second preset time width;

[0025] After calculating the RR(i) interval time according to the position of the R(i) wave, it further includes:

[0026] If the ratio of the RR(i) interval time to the RR(i - 1) interval time is greater than a first preset value, then detect the maximum value Max(i - 1) in the phasor signal PT(n) backward through a sliding window with a first preset time width from the position of the current R(i) wave. If the maximum value Max(i - 1) is greater than the adjusted adaptive threshold Thr(i), then define the position of this Max(i - 1) as the position of the R(i - 1) wave;

[0027] where the ratio of the adjusted adaptive threshold Thr(i) to the amplitude U R (i) of the current R(i) wave is a second preset value.

[0028] Preferably, determining the frequency characteristic parameter F(i) of the R(i) wave includes the following sub - steps:

[0029] Calculate the frequency characteristic parameter F(i) of the R(i) wave according to the RR(i) interval time, the RR(i - 1) interval time, and the RR(i + 1) interval time, where

[0030] , i ≥ 2, and the unit of F(i) is bpm;

[0031] Determining the rhythm characteristic parameter of the ECG signal includes the following sub - steps:

[0032] Calculate the difference between the RR(i) interval time and the RR(i - 1) interval time. If at least one of the differences is greater than a third preset value, then determine that the ECG signal has arrhythmia. If all the differences are less than or equal to the third preset value, then determine that the ECG signal has regular rhythm;

[0033] Determining the time characteristic parameter of the QRS(i) wave group where the R(i) wave is located according to the average value med(AUCs) and the integrated area AUC(i) of the R(i) wave comprises the following sub-steps:

[0034] Determine whether the ratio of the integrated area AUC(i) of the R(i) wave to the average value med(AUCs) is greater than a fourth preset value, if so, determine that the duration of the QRS(i) wave group to which the R(i) wave belongs is wide, if not, determine that the duration of the QRS(i) wave group to which the R(i) wave belongs is normal;

[0035] The method of identifying the P wave occurrence regularity characteristic parameters of the electrocardiogram signal according to the position of the real P(i) wave comprises the following sub-steps:

[0036] According to the position of the real P(i) wave, identify whether there are three consecutive real P(i) waves disappearing in the ECG signal. If so, determine

[0037] The P wave of the ECG signal disappears continuously. If not, it is determined that the P wave of the ECG signal is normal.

[0038] Preferably, judging whether the electrocardiogram signal has a first type of arrhythmia according to the time characteristic parameter, frequency characteristic parameter F(i), P wave regularity characteristic parameter and rhythm characteristic parameter of the QRS(i) wave group, and if so, outputting a first arrhythmia diagnosis result containing the first type of arrhythmia, and if not, outputting a second arrhythmia diagnosis result of a normal electrocardiogram, comprises the following sub-steps:

[0039] If the duration of all QRS(i) complexes is normal, the P waves are normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is less than 60 bpm, it is determined that the ECG signal has sinus bradycardia in the first type of arrhythmia;

[0040] If the duration of all QRS(i) wave groups is normal, the P wave is normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is greater than 100 bpm, it is determined that the ECG signal has sinus tachycardia in the first type of arrhythmia;

[0041] If the duration of all QRS(i) complexes is normal, the P waves disappear continuously and the rhythm is irregular, it is determined that the ECG signal has atrial fibrillation in the first type of arrhythmia;

[0042] If the duration of all QRS(i) complexes is normal, the P waves disappear continuously, the rhythm is normalized, and the frequency characteristic parameter F(i) of at least one R(i) wave is less than 150bpm, it is determined that the ECG signal has atrial flutter in the first type of arrhythmia;

[0043] If the duration of all QRS(i) complexes is normal, the P waves disappear continuously, the rhythm is normalized, and the frequency characteristic parameter F(i) of at least one R(i) wave is greater than or equal to 150bpm, it is determined that the ECG signal has supraventricular tachycardia in the first type of arrhythmia;

[0044] If the duration of the QRS(i) complex is wide and the frequency characteristic parameter F(i) of the R(i) wave in the QRS(i) complex is greater than 140 bpm, it is determined that the ECG signal has ventricular tachycardia of the first type of arrhythmia;

[0045] Outputting a first arrhythmia diagnosis result including a first type of arrhythmia, wherein the first type of arrhythmia includes one or more of sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, supraventricular tachycardia and ventricular tachycardia;

[0046] If the duration of all QRS(i) wave groups is normal, the P wave is normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is greater than or equal to 60bpm and less than or equal to 100bpm, the ECG signal is determined to be a normal ECG signal, and the output result is the second arrhythmia diagnosis result of the normal ECG.

[0047] Preferably, the arrhythmia diagnosis method further comprises:

[0048] Determine the RR interval prolongation characteristic parameters of the ECG signal;

[0049] Determine the premature beat characteristic parameters of the QRS(i) complex and the normal R wave characteristic parameters of the QRS(i) complex.

[0050] Preferably, the arrhythmia diagnosis method further comprises:

[0051] According to the RR interval prolongation characteristic parameters and the time limit characteristic parameters of the QRS(i) wave group, the premature beat characteristic parameters, and the normal R wave characteristic parameters, it is judged whether the electrocardiogram signal has the second type of arrhythmia. If so, a third arrhythmia diagnosis result including the first type of arrhythmia and the second type of arrhythmia is output. If not, the first arrhythmia diagnosis result is output as the final arrhythmia diagnosis result; the second type of arrhythmia includes one or more of supraventricular premature beats, paired supraventricular premature beats, supraventricular premature beats bigeminy, supraventricular premature beats trigeminy, ventricular premature beats, paired ventricular premature beats, bigeminy of ventricular premature beats, ventricular premature beats trigeminy and long intervals.

[0052] Preferably, the step of determining the RR interval prolongation characteristic parameter of the electrocardiogram signal comprises the following sub-steps:

[0053] If at least one RR(i) interval time is greater than or equal to a fifth preset value, it is determined that the ECG signal has a prolonged RR interval;

[0054] The method of determining the premature beat characteristic parameters of the QRS(i) complex and the normal R wave characteristic parameters of the QRS(i) complex comprises the following sub-steps:

[0055] step:

[0056] Determine whether the ratio of the RR(i) interval time to the RR(iN) interval time of the QRS(i) complex is less than a sixth preset value, if so, determine that the QRS(i) complex belongs to a premature wave, if not, determine that the QRS(i) complex does not belong to a premature wave;

[0057] The selection criteria of the RR(iN) interval time are as follows: the ratio of the RR(iN) interval time to the RR(iN-1) interval time is greater than or equal to the sixth preset value, and the RR(iN) interval time is less than the fifth preset value, and the value of N is the smallest;

[0058] If the duration of the QRS(i) complex is normal and does not belong to a premature beat, then the R(i) wave in the QRS(i) complex is determined to be a normal R wave.

[0059] Preferably, judging whether the electrocardiogram signal has the second type of arrhythmia according to the RR interval prolongation characteristic parameter and the time limit characteristic parameter of the QRS(i) wave group, the premature beat characteristic parameter, and the normal R wave characteristic parameter, and if so, outputting a third arrhythmia diagnosis result including the first type of arrhythmia and the second type of arrhythmia, and if not, outputting the first arrhythmia diagnosis result as the final arrhythmia diagnosis result, comprises the following sub-steps:

[0060] If the duration of the QRS(i) complex is normal and the QRS(i) complex is a premature beat, it is determined that the ECG signal contains a supraventricular premature beat in the second type of arrhythmia;

[0061] If two supraventricular premature beats appear in succession, it is determined that the ECG signal contains a pair of supraventricular premature beats in the second type of arrhythmia;

[0062] If each normal R wave is followed by a supraventricular premature beat, and the number of consecutive supraventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has a supraventricular premature beat bigeminy in the second type of arrhythmia;

[0063] If every two consecutive normal R waves are followed by supraventricular premature beats, and the number of consecutive supraventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has a supraventricular premature beat trigeminy in the second type of arrhythmia;

[0064] If the duration of the QRS(i) complex is wide and the QRS(i) complex is a premature beat, it is determined that the ECG signal contains a ventricular premature beat in the second type of arrhythmia;

[0065] If two ventricular premature beats appear in succession, it is determined that the ECG signal contains ventricular premature beat pairs in the second type of arrhythmia;

[0066] If each normal R wave is followed by a ventricular premature beat, and the number of consecutive ventricular premature beats is greater than or equal to the seventh preset value, it is determined that the ECG signal has ventricular premature beat bigeminy in the second type of arrhythmia;

[0067] If every two consecutive normal R waves are followed by ventricular premature beats, and the number of consecutive ventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has a ventricular premature beat trigeminy in the second type of arrhythmia;

[0068] If the ECG signal has a prolonged RR interval, it is determined that the ECG signal has a long pause in the second type of arrhythmia;

[0069] Outputting a third arrhythmia diagnosis result including the first arrhythmia type and the second arrhythmia type, or outputting the first arrhythmia diagnosis result as the final arrhythmia diagnosis result;

[0070] Among them, the second type of arrhythmia includes one or more of supraventricular premature beats, paired supraventricular premature beats, supraventricular premature beats bigeminy, supraventricular premature beats trigeminy, ventricular premature beats, paired ventricular premature beats, bigeminy, ventricular premature beats trigeminy and long pauses.

[0071] Preferably, the step of calculating the theoretical position of the P(i) wave according to the position of the R(i) wave, the position of the R(i-1) wave and the RR(i) interval time in the QRS(i) wave group, calculating the theoretical position of the P(1) wave according to the position of the R(1) wave, and calculating the theoretical position of the T(i-1) wave according to the position of the R(i-1) wave and the RR(i-1) interval time comprises the following sub-steps:

[0072] Calculate the position of the theoretical P(i) wave , where R(i) is the position of the R(i) wave, R(i-1) is the position of the R(i-1) wave, RR(i) is the RR(i) interval time, i≥2;

[0073] Calculate the position of the theoretical P(1) wave , where R(1) is the position of the R(1) wave;

[0074] Calculate the position of the theoretical T(i-1) wave , where RR(i-1) is the RR(i-1) interval time;

[0075] According to the maximum value U P (i) and the measured current R(i) amplitude U R(i) confirming the true P(i) wave and the position of the true P(i) wave, comprising the following sub-steps:

[0076] Calculate the maximum value U P (i) and the current R(i) amplitude U R Whether the ratio of (i) is greater than the eighth preset value, if so, the theoretical P(i) wave is determined to be the real P(i) wave, and the position of the theoretical P(i) wave is taken as the position of the real P(i) wave, if not, it is determined that the theoretical P(i) wave does not exist.

[0077] Preferably, the step of identifying the R(i) wave in the electrocardiogram signal further includes the following steps:

[0078] Performing wavelet transform on the electrocardiogram signal;

[0079] The ECG signal after wavelet transformation is input into the Butterworth band-stop filter for the first stage of hardware filtering;

[0080] The ECG signal after the first stage hardware filtering is input into a high-pass filter for the second stage hardware filtering to obtain the ECG signal after the second stage hardware filtering.

[0081] The present invention has the following beneficial technical effects:

[0082] (1) Based on phase transformation, accurate identification of R(i) and P(i) waves is performed, and the reliable area integration method is used to determine whether the duration of the QRS(i) complex is wide and deformed. In addition, the system combines the identification of eight pathological characteristics of ECG signals, including continuous disappearance of P waves, normal P waves, prolonged RR intervals, frequency characteristic parameters F(i), premature beats, normal R waves, irregular rhythms, and rhythm normalization. As many as fifteen types of arrhythmia pathological types can be diagnosed. This system is suitable for the diagnosis of arrhythmias in group ECGs and is accurate and practical.

[0083] (2) This method is based on software analysis thinking. Compared with the arrhythmia analysis method based on deep learning algorithm model in the prior art, it does not need to obtain a large amount of ECG sample data, thus avoiding the problems of poor detection effect caused by difficulty in obtaining sample data when building a deep learning algorithm model and long development cycle caused by large training samples.

[0084] (3) The arrhythmia diagnosis method implemented by this method is consistent with clinical diagnostic thinking and the application scenarios of most electrocardiogram arrhythmia diagnosis, which is convenient for the subsequent expansion of functions and subsequent maintenance.

[0085] (4) Because the collection of group ECG signals is generally single-channel and the ECG signals have a lot of noise, this method uses wavelet transform to filter out power frequency interference, Butterworth band-stop filter to filter out myoelectric interference, and high-pass filter to filter out baseline drift. After the above three stages of filtering processing, the subsequent identification of R(i) waves and P(i) waves in the ECG signals is more accurate, which improves the accuracy of arrhythmia diagnosis results.

[0086] (5) The ECG signal is filtered by a FIR bandpass filter based on a Hamming window to enhance the QRS complex and suppress the P wave and T wave, making the subsequent identification of the R(i) wave in the ECG signal more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 A schematic diagram of a flow chart of the arrhythmia diagnosis method corresponding to the first embodiment;

[0088] Figure 2 A flowchart of the arrhythmia diagnosis method corresponding to the second embodiment. DETAILED DESCRIPTION

[0089] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0090] Embodiment 1

[0091] like Figure 1 As shown, this embodiment provides a group electrocardiogram arrhythmia diagnosis method, comprising the following steps:

[0092] S01. Obtaining electrocardiogram signals. Specifically, reading in electrocardiogram signals collected by MLII equivalent limb leads.

[0093] S02, filtering the ECG signal. Specifically, the process includes: performing wavelet transform on the ECG signal. Inputting the ECG signal after wavelet transform into a Butterworth band-stop filter for first-stage hardware filtering. Inputting the ECG signal after the first-stage hardware filtering into a high-pass filter for second-stage hardware filtering to obtain the ECG signal after the second-stage hardware filtering. The cutoff frequency of the high-pass filter is preferably 0.67 Hz.

[0094] S03, identifying the R(i) wave in the electrocardiogram signal after the hardware filtering in the second stage, and determining the position of the R(i) wave. Specifically including:

[0095] S031. Digital filter the ECG signal after the second-stage hardware filtering through a Hamming window-based FIR band-pass filter. Enhance the QRS complex through this digital filtering while suppressing the P wave and T wave. The passband frequency range of the FIR band-pass filter is set between 12 - 19 Hz.

[0096] S032. Perform a phasor transform on the digitally filtered ECG signal to obtain the phasor signal PT(n) corresponding to the ECG signal, where , is the instantaneous value of the ECG signal, 0 < Rv < 1, Rv represents the waveform enhancement degree of the ECG signal, and the complex-domain transform formula of the ECG signal is . The preferred value of Rv is 0.001.

[0097] S033. Detect the maximum value Max(i) in the phasor signal PT(n) through a sliding window with the first preset time width. If the maximum value Max(i) is greater than the adaptive threshold Thr(i), then define the position of this Max(i) as the position of the R(i) wave. Among them, the adaptive threshold Thr(i) is twice the standard deviation of the phasor signal PT(n) calculated in a moving window with the second preset time width. The adaptive threshold , where X j is the phasor signal PT(n) data in the moving window with the second preset time width. The phasor signal PT(n) data is the data obtained after discretely sampling the phasor signal PT(n). N is the number of all phasor signal PT(n) data in the moving window with the second preset time width, is the average value of all phasor signal PT(n) data. The first preset time width is preferably 300 ms, and the second preset time width is preferably 2 s.

[0098] S04. Calculate the RR(i) interval time based on the position of the R(i) wave. The RR(i) interval time is the distance between the current R(i) wave position and the R(i - 1) wave position, where i ≥ 2.

[0099] Optionally, after calculating the RR(i) interval time, further perform the following judgment to find the position of the R(i - 1) wave that may be missed in S03:

[0100] If the ratio of the RR(i) interval time to the RR(i - 1) interval time is greater than the first preset value, then detect the maximum value Max(i - 1) in the phasor signal PT(n) backward from the current R(i) wave position through a sliding window with the first preset time width. If the maximum value Max(i - 1) is greater than the adjusted adaptive threshold Thr(i), then define the position of this Max(i - 1) as the position of the R(i - 1) wave. Among them, the adjusted adaptive threshold Thr(i) is related to the current R(i) wave amplitude UR The ratio of R(i) to the second preset value. The first preset value is preferably 1.75, and the second preset value is preferably 30%. In a common embodiment, the amplitude U of R(i) is R (i) is also called the height of the QRS(i) complex, which represents the distance between the peak (point R) of the R(i) wave and the horizontal line.

[0101] S05. Determine the time characteristic parameters of the QRS(i) complex where each R(i) wave is located. The time characteristic parameters of the QRS(i) complex include QRS(i) complex time width and QRS(i) complex time normal. In a common embodiment, QRS(i) complex time width is also called QRS(i) complex time width abnormality. S05 specifically includes:

[0102] S051. Perform time domain integration on each R(i) wave according to a preset integration limit to obtain an integration area AUC(i) corresponding to each R(i) wave. The preset integration limit is preferably: the upper integration limit is 150ms after the position of the R(i) wave, and the lower integration limit is 150ms before the position of the R(i) wave.

[0103] S052. Calculate the mean value med(AUCs) of all integrated areas AUC(i).

[0104] S053, determining the time characteristic parameters of the QRS(i) complex where the R(i) wave is located according to the average value med(AUCs) and the integrated area AUC(i) of the R(i) wave. Preferably, S053 specifically includes the following sub-steps:

[0105] It is determined whether the ratio of the integrated area AUC(i) of the R(i) wave to the average value med(AUCs) is greater than a fourth preset value. If so, it is determined that the duration of the QRS(i) wave group to which the R(i) wave belongs is wide. If not, it is determined that the duration of the QRS(i) wave group to which the R(i) wave belongs is normal. The fourth preset value is preferably 1.3.

[0106] S06, determining the frequency characteristic parameter F(i) of the R(i) wave. Preferably, S06 specifically includes the following sub-steps:

[0107] The frequency F(i) of the R(i) wave is calculated based on the RR(i) interval time, the RR(i-1) interval time, and the RR(i+1) interval time.

[0108] in , i≥2, the unit of F(i) is bpm (beats per minute). The frequency characteristic parameter F(i) is the local frequency, not the average frequency of the whole electrocardiogram, which is more conducive to the diagnosis of two kinds of paroxysmal arrhythmias, supraventricular tachycardia and ventricular tachycardia. From a clinical perspective, it is also conducive to the expansion of the diagnosis function of other paroxysmal arrhythmias in the later stage.

[0109] S07, identifying the P(i) wave in the electrocardiogram signal, and determining the position of the P(i) wave. Specifically including:

[0110] S071. If the time limit characteristic parameter of the QRS(i) wave group is that the time limit of the QRS(i) wave group is normal, the position of the theoretical P(i) wave is calculated according to the position of the R(i) wave, the position of the R(i-1) wave and the RR(i) interval time in the QRS(i) wave group, the position of the theoretical P(1) wave is calculated according to the position of the R(1) wave, and the position of the theoretical T(i-1) wave is calculated according to the position of the R(i-1) wave and the RR(i-1) interval time, where i≥2. When the time limit characteristic parameter of the QRS(i) wave group is that the time limit of the QRS(i) wave group is wide, it is directly determined that there is no P(i) wave in front of the QRS(i) wave group, and the P(i) wave is not identified for these QRS(i) wave groups.

[0111] Optionally, in S071, the position of the theoretical P(i) wave is calculated according to the position of the R(i) wave, the position of the R(i-1) wave and the RR(i) interval time in the QRS(i) wave group, the position of the theoretical P(1) wave is calculated according to the position of the R(1) wave, and the position of the theoretical T(i-1) wave is calculated according to the position of the R(i-1) wave and the RR(i-1) interval time, including the following sub-steps:

[0112] S0711. Calculate the position of the theoretical P(i) wave , where R(i) is the position of the R(i) wave, R(i-1) is the position of the R(i-1) wave, RR(i) is the RR(i) interval time, and i≥2.

[0113] S0712. Calculate the position of the theoretical P(1) wave , where R(1) is the position of the R(1) wave.

[0114] S0713. Calculate the position of the theoretical T(i-1) wave , where RR(i-1) is the RR(i-1) interval time.

[0115] S072. If the position of the theoretical P(i) wave does not overlap with the position of the theoretical T(i-1) wave, the theoretical P(i) wave in the ECG signal is transformed in phase. The same method as the aforementioned R(i) wave identification is used for phase transformation, and the preferred value of Rv is 0.05. When the position of the theoretical P(i) wave overlaps with the position of the theoretical T(i-1) wave, the P(i) wave may exist, but is hidden in the T(i-1) wave and cannot be determined, or it may not exist. Therefore, these theoretical P(i) waves are directly judged as non-existent.

[0116] S073. Find the maximum value U in the theoretical P(i) wave after the phase transformationP (i), and according to the maximum value U P (i) and the measured current R(i) amplitude U R (i) The real P(i) wave and the position of the real P(i) wave are confirmed. The position of the real P(i) wave corresponds to the position of the theoretical P(i) wave.

[0117] Optionally, S073 includes the following sub-steps:

[0118] S0731. Find the maximum value U in the theoretical P(i) wave after the phase transformation P (i).

[0119] S0732. Calculate the maximum value U P (i) and the current R(i) amplitude U R Whether the ratio of (i) is greater than the eighth preset value, if so, the theoretical P(i) wave is determined to be the real P(i) wave, and the position of the theoretical P(i) wave is used as the position of the real P(i) wave, if not, it is determined that the theoretical P(i) wave does not exist. The eighth preset value is preferably 0.05.

[0120] S08, identifying the P wave occurrence regularity characteristic parameters of the electrocardiogram signal according to the position of the real P(i) wave. The P wave occurrence regularity characteristic parameters include continuous disappearance of the P wave and normal P wave.

[0121] Optionally, S08 includes the following sub-steps:

[0122] According to the position of the real P(i) wave, identify whether there are three consecutive real P(i) waves disappearing in the ECG signal. If so, determine

[0123] The P wave of the ECG signal disappears continuously. If not, it is determined that the P wave of the ECG signal is normal.

[0124] Considering that the P(i) wave determination may have a large error, especially when individual ECG waveforms are atypical and affected by environmental movements, such as when soldiers, students and other groups are undergoing physical training or physical fitness tests. Therefore, if the P(i) wave disappears or is missed occasionally, it does not affect the determination of arrhythmia. Therefore, in the determination of arrhythmia, only the disappearance of continuous P waves is more meaningful, so the disappearance of three consecutive P waves is determined as "continuous disappearance of P waves", otherwise it is "normal P waves".

[0125] S09: determining rhythm characteristic parameters of the electrocardiogram signal. The rhythm characteristic parameters include irregular rhythm and regular rhythm.

[0126] Optionally, S09 includes the following sub-steps: calculating the difference between the RR(i) interval time and the RR(i-1) interval time, and if at least one of the above differences is greater than a third preset value, determining that the ECG signal has an irregular rhythm, and if all the differences are less than or equal to the third preset value, determining that the ECG signal rhythm is normalized. The third preset value is preferably 0.12 seconds.

[0127] S10, outputting the arrhythmia diagnosis result. Specifically including:

[0128] According to the time characteristic parameter of the QRS(i) wave group, the frequency characteristic parameter F(i), the P wave regularity characteristic parameter and the rhythm characteristic parameter, it is judged whether the electrocardiogram signal has the first type of arrhythmia. If so, the first arrhythmia diagnosis result containing the first type of arrhythmia is output. If not, the second arrhythmia diagnosis result of the normal electrocardiogram is output. The first type of arrhythmia includes one or more of sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, supraventricular tachycardia and ventricular tachycardia.

[0129] Optionally, S10 includes the following sub-steps:

[0130] If the duration of all QRS(i) complexes is normal, the P waves are normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is less than 60bpm, it is determined that the ECG signal has sinus bradycardia in the first type of arrhythmia.

[0131] If the duration of all QRS(i) complexes is normal, the P waves are normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is greater than 100bpm, it is determined that the ECG signal has sinus tachycardia in the first type of arrhythmia.

[0132] If the duration of all QRS(i) complexes is normal, P waves disappear continuously and the rhythm is irregular, it is determined that the ECG signal contains atrial fibrillation, which is a type of first arrhythmia.

[0133] If the duration of all QRS(i) complexes is normal, P waves disappear continuously, the rhythm is normalized, and the frequency characteristic parameter F(i) of at least one R(i) wave is less than 150bpm, it is determined that atrial flutter, which is a type of first arrhythmia, exists in the ECG signal.

[0134] If the duration of all QRS(i) complexes is normal, P waves disappear continuously, the rhythm is normalized, and the frequency characteristic parameter F(i) of at least one R(i) wave is greater than or equal to 150bpm, it is determined that the ECG signal has supraventricular tachycardia of the first type of arrhythmia.

[0135] If the duration of the QRS(i) complex is wide and the frequency characteristic parameter F(i) of the R(i) wave in the QRS(i) complex is greater than 140 bpm, it is determined that the ECG signal has ventricular tachycardia of the first type of arrhythmia.

[0136] The output includes a first arrhythmia diagnosis result of a first type of arrhythmia, wherein the first type of arrhythmia includes one or more of sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, supraventricular tachycardia and ventricular tachycardia.

[0137] If the duration of all QRS(i) wave groups is normal, the P wave is normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is greater than or equal to 60bpm and less than or equal to 100bpm, the ECG signal is determined to be a normal ECG signal, and the output result is the second arrhythmia diagnosis result of the normal ECG.

[0138] This embodiment has the following significant advantages:

[0139] a) The normal QRS(i) wave group duration is generally no more than 0.11 seconds, and can present rS type, Rs type or RS type according to the position difference of the lead placement. In clinical practice, a QRS(i) wave group duration greater than 0.12 seconds is often used as the recognition standard for a wide QRS(i) wave group duration. In the identification of pathological features of ECG signals, since the identification of the positions of Q and S points is prone to deviations, especially when facing various arrhythmia ECGs or human factors, environmental factors, etc. that affect the quality of ECG signals, such as when soldiers, students and other groups conduct physical training or physical fitness tests, if the QRS(i) wave group duration is detected based on the identification of the positions of Q and S points, there will be very large errors. In this embodiment, step S05 realizes a reliable judgment of the wide QRS(i) wave group duration through the area integration method.

[0140] b) In the identification of pathological features of ECG signals, the identification of P(i) waves is a difficult problem in the industry. The specific reasons are as follows:

[0141] 1) Low signal makes it easy to be confused with the background; 2) P(i) wave does not have a fixed time and frequency of occurrence; 3) The morphology of P(i) wave varies greatly from person to person; 4) In some cases of atrioventricular detachment, the location of P(i) wave is uncertain; 5) In supraventricular premature beats, P(i) wave can overlap with T wave; 6) In atrial flutter and atrial fibrillation, P(i) wave disappears and is replaced by high-frequency F wave; 7) P(i) wave does not exist in ventricular premature beats.

[0142] In step S07 of this embodiment, the theoretical P(i) wave position is first determined by the distribution law of the QRS complex, P wave and T wave of the ECG signal, and then the phasor transformation of the theoretical P(i) wave position area is performed based on the phasor transformation method. The waveform is enhanced to a certain extent by Rv. After the phasor transformation, it is easier to find the maximum value U P (i) By taking the maximum value U P (i) and R(i) amplitude U R(i) is compared to identify whether the theoretical P(i) wave is the real P(i) wave, thereby achieving accurate identification of the P(i) wave.

[0143] Embodiment 2

[0144] like Figure 2 As shown, the difference between this embodiment and the first embodiment is that after realizing the diagnosis of abnormal sinus rhythm, atrial arrhythmia, supraventricular tachycardia and ventricular tachycardia of the ECG signal, the diagnosis of ventricular arrhythmia and long pause is added. Therefore, the second embodiment adds the recognition of the following pathological features of the ECG signal: premature beats, normal R waves and prolonged RR interval.

[0145] S04 also includes: determining the RR interval prolongation characteristic parameter of the electrocardiogram signal.

[0146] Optionally, determining the RR interval prolongation characteristic parameter of the ECG signal includes the following sub-steps: if at least one RR(i) interval time is greater than or equal to a fifth preset value, determining that the ECG signal has RR interval prolongation. The fifth preset value is preferably two seconds.

[0147] The following steps are also included between S06 and S07:

[0148] Determine the premature beat characteristic parameters of the QRS(i) complex and the normal R wave characteristic parameters of the QRS(i) complex.

[0149] Optionally, determining the premature beat characteristic parameters of the QRS(i) complex and the normal R wave characteristic parameters of the QRS(i) complex comprises the following sub-steps:

[0150] It is determined whether the ratio of the RR(i) interval time of the QRS(i) wave group to the RR(iN) interval time is less than the sixth preset value. If so, it is determined that the QRS(i) wave group belongs to a premature wave. If not, it is determined that the QRS(i) wave group does not belong to a premature wave. The sixth preset value is preferably 0.85. Wherein, the selection criteria of the RR(iN) interval time are: the ratio of the RR(iN) interval time to the RR(iN-1) interval time is greater than or equal to the sixth preset value, and the RR(iN) interval time is less than the fifth preset value, and the value of N is the smallest. Because in the determination of whether the QRS(i) wave group is premature, R(1) wave, R(2) wave, R(3) wave, etc. are determined in sequence according to the timing. If the R(i-1) wave is not a premature wave and the RR interval is normal, then when determining whether the QRS(i) complex is a premature wave, the RR(i) interval time is compared with the RR(i-1) interval time; if the R(i-1) wave has been determined to be a premature wave, or has been determined to have a prolonged RR interval, then when determining whether the QRS(i) complex is a premature wave, the RR(i) interval time is compared with the RR(i-2) interval time; if the R(i-2) wave has been determined to be a premature wave, or has been determined to have a prolonged RR interval, then when determining whether the QRS(i) complex is a premature wave, the RR(i) interval time is compared with the RR(i-3) interval time, and so on.

[0151] If the duration of the QRS(i) complex is normal and does not belong to a premature beat, then the R(i) wave in the QRS(i) complex is determined to be a normal R wave.

[0152] In S10, judging whether the electrocardiogram signal has a first type of arrhythmia according to the time characteristic parameter, frequency characteristic parameter F(i) of the QRS(i) wave group, the regularity characteristic parameter of the P wave and the rhythm characteristic parameter, and if so, outputting the first arrhythmia diagnosis result including the first type of arrhythmia also includes:

[0153] According to the RR interval extension characteristic parameter and the time limit characteristic parameter of the QRS(i) wave group, the premature beat characteristic parameter, and the normal R wave characteristic parameter, it is judged whether the electrocardiogram signal has the second type of arrhythmia. If so, a third arrhythmia diagnosis result including the first type of arrhythmia and the second type of arrhythmia is output. If not, the first arrhythmia diagnosis result is output as the final arrhythmia diagnosis result. The second type of arrhythmia includes one or more of supraventricular premature beats, supraventricular premature beats in pairs, supraventricular premature beats bigeminy, supraventricular premature beats trigeminy, ventricular premature beats, ventricular premature beats in pairs, ventricular premature beats bigeminy, ventricular premature beats trigeminy and long interval.

[0154] Optionally, judging whether the electrocardiogram signal has a second type of arrhythmia according to the RR interval prolongation characteristic parameter and the time limit characteristic parameter of the QRS(i) wave group, the premature beat characteristic parameter, and the normal R wave characteristic parameter, and if so, outputting a third arrhythmia diagnosis result including the first type of arrhythmia and the second type of arrhythmia, and if not, outputting the first arrhythmia diagnosis result as the final arrhythmia diagnosis result, comprises the following sub-steps:

[0155] If the duration of the QRS(i) complex is normal and the QRS(i) complex is a premature beat, it is determined that the electrocardiogram signal contains supraventricular premature beats of the second type of arrhythmia.

[0156] If two supraventricular premature beats appear consecutively, it is determined that the ECG signal contains a pair of supraventricular premature beats of the second type of arrhythmia.

[0157] If each normal R wave is followed by a supraventricular premature beat, and the number of consecutive supraventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has supraventricular premature beat bigeminy in the second type of arrhythmia. The seventh preset value is preferably 2.

[0158] If every two consecutive normal R waves are followed by supraventricular premature beats, and the number of consecutive supraventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal contains supraventricular premature beat trigeminy in the second type of arrhythmia.

[0159] If the duration of the QRS(i) complex is wide and the QRS(i) complex is a premature beat, it is determined that the ECG signal contains ventricular premature beats of the second type of arrhythmia.

[0160] If two ventricular premature beats appear consecutively, it is determined that the ECG signal contains ventricular premature beat pairs of the second type of arrhythmia.

[0161] If each normal R wave is followed by a ventricular premature beat, and the number of consecutive ventricular premature beats is greater than or equal to a seventh preset value, it is determined that the electrocardiogram signal contains ventricular premature beat bigeminy of the second type of arrhythmia.

[0162] If every two consecutive normal R waves are followed by ventricular premature beats, and the number of consecutive ventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal contains ventricular premature beat trigeminy in the second type of arrhythmia.

[0163] If the RR interval is prolonged in the ECG signal, it is determined that the ECG signal has a long pause in the second type of arrhythmia.

[0164] If the ECG signal contains one or more of the second arrhythmia types, a third arrhythmia diagnosis result including the first arrhythmia type and the second arrhythmia type is output; otherwise, the first arrhythmia diagnosis result is output as the final arrhythmia diagnosis result.

[0165] Through the second embodiment, the method realizes the diagnosis of up to fifteen arrhythmia pathological types, is suitable for the diagnosis of arrhythmia in group electrocardiograms, and is stable and reliable. At the same time, based on clinical diagnosis thinking, it is convenient for later extension and maintenance.

[0166] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for diagnosing arrhythmia by electrocardiogram, characterized in that: The following steps are involved: Obtain ECG signals; Identifying an R(i) wave in an electrocardiogram signal and determining a position of the R(i) wave; Calculate the RR(i) interval time according to the position of the R(i) wave, wherein the RR(i) interval time is the distance between the current R(i) wave position and the R(i-1) wave position, where i≥2; Perform time domain integration on each R(i) wave according to the preset integration limit to obtain the integral area AUC(i) corresponding to each R(i) wave; Calculate the mean med(AUCs) of all integrated areas AUC(i); Determine the time characteristic parameters of the QRS(i) complex where the R(i) wave is located according to the average value med(AUCs) and the integrated area AUC(i) of the R(i) wave; Determine the frequency characteristic parameter F(i) of the R(i) wave; If the time characteristic parameter of the QRS(i) complex is that the time of the QRS(i) complex is normal, the position of the theoretical P(i) wave is calculated according to the position of the R(i) wave, the position of the R(i-1) wave and the RR(i) interval in the QRS(i) complex, the position of the theoretical P(1) wave is calculated according to the position of the R(1) wave, and the position of the theoretical T(i-1) wave is calculated according to the position of the R(i-1) wave and the RR(i-1) interval, where i≥2; If the position of the theoretical P(i) wave does not overlap with the position of the theoretical T(i-1) wave, then the phasor transformation of the theoretical P(i) wave in the ECG signal is performed; Find the maximum value U in the theoretical P(i) wave after phase transformation P (i), and according to the maximum value U P (i) and the measured current R(i) amplitude U R (i) Identify the true P(i) wave and the position of the true P(i) wave; Identify the regular characteristic parameters of the P wave of the ECG signal according to the position of the real P(i) wave; Determine the rhythm characteristic parameters of the ECG signal; According to the time characteristic parameters, frequency characteristic parameters F(i), P wave regularity characteristic parameters and rhythm characteristic parameters of the QRS(i) wave group, it is judged whether the electrocardiogram signal has the first type of arrhythmia. If so, a first arrhythmia diagnosis result containing the first type of arrhythmia is output. If not, a second arrhythmia diagnosis result of a normal electrocardiogram is output. The first type of arrhythmia includes one or more of sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, supraventricular tachycardia and ventricular tachycardia.

2. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 1, characterized in that: The step of identifying the R(i) wave in the electrocardiogram signal and determining the position of the R(i) wave comprises the following sub-steps: The ECG signal is digitally filtered using a Hamming window-based FIR bandpass filter; Perform phasor transformation on the digitally filtered electrocardiogram (ECG) signal to obtain the phasor signal PT(n) corresponding to the ECG signal, where , is the instantaneous value of the ECG signal, 0 < Rv < 1, Rv represents the waveform enhancement degree of the ECG signal, and the complex domain transformation formula of the ECG signal is ; Detecting the maximum value Max(i) in the phasor signal PT(n) through a sliding window of a first preset time width, if the maximum value Max(i) is greater than the adaptive threshold Thr(i), defining the position of Max(i) as the position of the R(i) wave; wherein the adaptive threshold Thr(i) is twice the standard deviation of the phasor signal PT(n) calculated in the second preset time width moving window; After calculating the RR(i) interval time according to the position of the R(i) wave, the method further comprises: If the ratio of the RR(i) interval time to the RR(i-1) interval time is greater than a first preset value, the maximum value Max(i-1) in the phasor signal PT(n) is detected from the current R(i) wave position through a sliding window of the first preset time width, and if the maximum value Max(i-1) is greater than the adjusted adaptive threshold Thr(i), the Max(i-1) position is defined as the R(i-1) wave position; Among them, the adjusted adaptive threshold Thr(i) and the current R(i) amplitude U R The ratio of (i) is a second preset value.

3. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 1, characterized in that: Determining the frequency characteristic parameter F(i) of the R(i) wave comprises the following sub-steps: The frequency characteristic parameters of the R(i) wave are calculated based on the RR(i) interval time, the RR(i-1) interval time, and the RR(i+1) interval time. Number F(i), where , i ≥ 2, the unit of F(i) is bpm; The step of determining the rhythm characteristic parameters of the electrocardiogram signal comprises the following sub-steps: Calculate the difference between the RR(i) interval time and the RR(i-1) interval time, if at least one of the difference is greater than a third preset value, determine that the ECG signal has an irregular rhythm, if all of the difference are less than or equal to the third preset value, determine that the ECG signal has a regular rhythm; Determining the time characteristic parameter of the QRS(i) wave group where the R(i) wave is located according to the average value med(AUCs) and the integrated area AUC(i) of the R(i) wave comprises the following sub-steps: Determine whether the ratio of the integrated area AUC(i) of the R(i) wave to the average value med(AUCs) is greater than a fourth preset value, if so, determine that the duration of the QRS(i) wave group to which the R(i) wave belongs is wide, if not, determine that the duration of the QRS(i) wave group to which the R(i) wave belongs is normal; The method of identifying the P wave occurrence regularity characteristic parameters of the electrocardiogram signal according to the position of the real P(i) wave comprises the following sub-steps: According to the position of the real P(i) wave, it is identified whether there are three consecutive disappearances of the real P(i) wave in the ECG signal. If so, it is determined that the ECG signal has continuous disappearance of the P wave. If not, it is determined that the P wave of the ECG signal is normal.

4. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 3, characterized in that: The method of judging whether the electrocardiogram signal has a first type of arrhythmia according to the time characteristic parameter of the QRS(i) wave group, the frequency characteristic parameter F(i), the P wave regularity characteristic parameter and the rhythm characteristic parameter, and if so, outputting a first arrhythmia diagnosis result containing the first type of arrhythmia, and if not, outputting a second arrhythmia diagnosis result of a normal electrocardiogram, comprises the following sub-steps: If the duration of all QRS(i) complexes is normal, the P waves are normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is less than 60 bpm, it is determined that the ECG signal has sinus bradycardia in the first type of arrhythmia; If the duration of all QRS(i) wave groups is normal, the P wave is normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is greater than 100 bpm, it is determined that the ECG signal has sinus tachycardia in the first type of arrhythmia; If the duration of all QRS(i) complexes is normal, the P waves disappear continuously and the rhythm is irregular, it is determined that the ECG signal has atrial fibrillation in the first type of arrhythmia; If the duration of all QRS(i) complexes is normal, the P waves disappear continuously, the rhythm is normalized, and the frequency characteristic parameter F(i) of at least one R(i) wave is less than 150bpm, it is determined that the ECG signal has atrial flutter in the first type of arrhythmia; If the duration of all QRS(i) complexes is normal, the P waves disappear continuously, the rhythm is normalized, and the frequency characteristic parameter F(i) of at least one R(i) wave is greater than or equal to 150bpm, it is determined that the ECG signal has supraventricular tachycardia in the first type of arrhythmia; If the duration of the QRS(i) complex is wide and the frequency characteristic parameter F(i) of the R(i) wave in the QRS(i) complex is greater than 140 bpm, it is determined that the ECG signal has ventricular tachycardia of the first type of arrhythmia; Outputting a first arrhythmia diagnosis result including a first type of arrhythmia, wherein the first type of arrhythmia includes one or more of sinus bradycardia, sinus tachycardia, atrial fibrillation, atrial flutter, supraventricular tachycardia and ventricular tachycardia; If the duration of all QRS(i) wave groups is normal, the P wave is normal, and the average value of the frequency characteristic parameter F(i) of all R(i) waves is greater than or equal to 60bpm and less than or equal to 100bpm, the ECG signal is determined to be a normal ECG signal, and the output result is the second arrhythmia diagnosis result of the normal ECG.

5. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 3, characterized in that: The arrhythmia diagnosis method further comprises: Determine the RR interval prolongation characteristic parameters of the ECG signal; Determine the premature beat characteristic parameters of the QRS(i) complex and the normal R wave characteristic parameters of the QRS(i) complex.

6. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 5, characterized in that: The arrhythmia diagnosis method further comprises: According to the RR interval prolongation characteristic parameters and the time limit characteristic parameters of the QRS(i) wave group, the premature beat characteristic parameters, and the normal R wave characteristic parameters, it is judged whether the electrocardiogram signal has the second type of arrhythmia. If so, a third arrhythmia diagnosis result including the first type of arrhythmia and the second type of arrhythmia is output. If not, the first arrhythmia diagnosis result is output as the final arrhythmia diagnosis result; the second type of arrhythmia includes one or more of supraventricular premature beats, paired supraventricular premature beats, supraventricular premature beats bigeminy, supraventricular premature beats trigeminy, ventricular premature beats, paired ventricular premature beats, bigeminy of ventricular premature beats, ventricular premature beats trigeminy and long intervals.

7. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 6, characterized in that: The step of determining the RR interval prolongation characteristic parameter of the electrocardiogram signal comprises the following sub-steps: If at least one RR(i) interval time is greater than or equal to a fifth preset value, it is determined that the ECG signal has a prolonged RR interval; The method of determining the premature beat characteristic parameters of the QRS(i) complex and the normal R wave characteristic parameters of the QRS(i) complex comprises the following sub-steps: step: Determine whether the ratio of the RR(i) interval time to the RR(iN) interval time of the QRS(i) complex is less than a sixth preset value, if so, determine that the QRS(i) complex belongs to a premature wave, if not, determine that the QRS(i) complex does not belong to a premature wave; The selection criteria of the RR(iN) interval time are as follows: the ratio of the RR(iN) interval time to the RR(iN-1) interval time is greater than or equal to the sixth preset value, and the RR(iN) interval time is less than the fifth preset value, and the value of N is the smallest; If the duration of the QRS(i) complex is normal and does not belong to a premature beat, then the R(i) wave in the QRS(i) complex is determined to be a normal R wave.

8. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 7, characterized in that: The method of judging whether the electrocardiogram signal has a second type of arrhythmia according to the RR interval prolongation characteristic parameter and the time limit characteristic parameter of the QRS(i) wave group, the premature beat characteristic parameter, and the normal R wave characteristic parameter, and if so, outputting a third arrhythmia diagnosis result including the first type of arrhythmia and the second type of arrhythmia, and if not, outputting the first arrhythmia diagnosis result as the final arrhythmia diagnosis result, comprises the following sub-steps: If the duration of the QRS(i) complex is normal and the QRS(i) complex is a premature beat, it is determined that the ECG signal contains a supraventricular premature beat in the second type of arrhythmia; If two supraventricular premature beats appear in succession, it is determined that the ECG signal contains a pair of supraventricular premature beats in the second type of arrhythmia; If each normal R wave is followed by a supraventricular premature beat, and the number of consecutive supraventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has a supraventricular premature beat bigeminy in the second type of arrhythmia; If every two consecutive normal R waves are followed by supraventricular premature beats, and the number of consecutive supraventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has a supraventricular premature beat trigeminy in the second type of arrhythmia; If the duration of the QRS(i) complex is wide and the QRS(i) complex is a premature beat, it is determined that the ECG signal contains a ventricular premature beat in the second type of arrhythmia; If two ventricular premature beats appear in succession, it is determined that the ECG signal contains ventricular premature beat pairs in the second type of arrhythmia; If each normal R wave is followed by a ventricular premature beat, and the number of consecutive ventricular premature beats is greater than or equal to the seventh preset value, it is determined that the ECG signal has ventricular premature beat bigeminy in the second type of arrhythmia; If every two consecutive normal R waves are followed by ventricular premature beats, and the number of consecutive ventricular premature beats is greater than or equal to the seventh preset value, it is determined that the electrocardiogram signal has a ventricular premature beat trigeminy in the second type of arrhythmia; If the ECG signal has a prolonged RR interval, it is determined that the ECG signal has a long pause in the second type of arrhythmia; Outputting a third arrhythmia diagnosis result including the first arrhythmia type and the second arrhythmia type, or outputting the first arrhythmia diagnosis result as the final arrhythmia diagnosis result; Among them, the second type of arrhythmia includes one or more of supraventricular premature beats, paired supraventricular premature beats, supraventricular premature beats bigeminy, supraventricular premature beats trigeminy, ventricular premature beats, paired ventricular premature beats, bigeminy, ventricular premature beats trigeminy and long pauses.

9. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 1, characterized in that: The method of calculating the theoretical position of the P(i) wave according to the position of the R(i) wave, the position of the R(i-1) wave and the RR(i) interval time in the QRS(i) wave group, calculating the theoretical position of the P(1) wave according to the position of the R(1) wave, and calculating the theoretical position of the T(i-1) wave according to the position of the R(i-1) wave and the RR(i-1) interval time, comprises the following sub-steps: Calculate the position of the theoretical P(i) wave , where R(i) is the position of the R(i) wave, R(i-1) is the position of the R(i-1) wave, RR(i) is the RR(i) interval time, i≥2; Calculate the position of the theoretical P(1) wave , where R(1) is the position of the R(1) wave; Calculate the position of the theoretical T(i-1) wave , where RR(i-1) is the RR(i-1) interval time; According to the maximum value U P (i) and the measured current R(i) amplitude U R (i) confirming the true P(i) wave and the position of the true P(i) wave, comprising the following sub-steps: Calculate the maximum value U P (i) and the current R(i) amplitude U R Whether the ratio of (i) is greater than the eighth preset value, if so, the theoretical P(i) wave is determined to be the real P(i) wave, and the position of the theoretical P(i) wave is taken as the position of the real P(i) wave, if not, it is determined that the theoretical P(i) wave does not exist.

10. A method for diagnosing arrhythmia using electrocardiogram of a group according to claim 1, characterized in that: The step of identifying the R(i) wave in the electrocardiogram signal also includes the following steps: Performing wavelet transform on the electrocardiogram signal; The ECG signal after wavelet transformation is input into the Butterworth band-stop filter for the first stage of hardware filtering; The ECG signal after the first stage hardware filtering is input into a high-pass filter for the second stage hardware filtering to obtain the ECG signal after the second stage hardware filtering.

Images