CN110367968B - Right bundle branch retardation detection method, device, equipment and storage medium - Google Patents
Right bundle branch retardation detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN110367968B CN110367968B CN201910754379.8A CN201910754379A CN110367968B CN 110367968 B CN110367968 B CN 110367968B CN 201910754379 A CN201910754379 A CN 201910754379A CN 110367968 B CN110367968 B CN 110367968B
- Authority
- CN
- China
- Prior art keywords
- bundle branch
- branch block
- right bundle
- electrocardiosignal
- classification model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7282—Event detection, e.g. detecting unique waveforms indicative of a medical condition
Abstract
The invention discloses a method, a device, equipment and a computer storage medium for detecting right bundle branch retardation, which comprise the following steps: acquiring a first electrocardiosignal of a target object, and performing signal preprocessing on the first electrocardiosignal to obtain a second electrocardiosignal for eliminating noise interference; extracting the QRS complex from the second electrocardiosignal; extracting at least one predetermined feature from the QRS complex; wherein the characteristics are selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the characteristics at least comprise one of the following: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient; and inputting the extracted features into at least one pre-established classification model to perform right bundle branch block identification. The method and the device perform characteristic analysis based on the waveform characteristics of the electrocardiosignal of the right bundle branch block, and improve the identification efficiency and accuracy of the right bundle branch block waveform.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method, a device, equipment and a storage medium for detecting right bundle branch retardation.
Background
Right Bundle Branch Block (RBBB) is a type of cardiac electrical conduction system blocking disease that is caused by the blockage of the Right Bundle Branch of the heart, which in turn prevents electrical signals from passing through this pathway to the Right ventricle and must be activated by signals from the left ventricle. Diseases that may lead to RBBB include: atrial septal defect, brucella syndrome, right ventricular hypertrophy, pulmonary embolism, ischemic heart disease, rheumatic fever, myocarditis, cardiomyopathy or hypertension, and the like.
To address this problem, identification of right bundle branch block is required, and accurate identification of right bundle branch block can aid in the prevention and treatment of associated cardiovascular disease. The basic pathophysiological defect in right Bundle branch block is mainly due to the failure of the electrical impulse to travel from the HIS Bundle (HIS Bundle) to the right Bundle branch. In the case of normal left bundle branch conduction, the right ventricular depolarization does not coincide significantly with the left ventricle. This ventricular depolarization mismatch gives an Electrocardiogram (ECG) change. At present, the identification of the right bundle branch block can be realized based on the change of electrocardiosignals, but the accuracy and the stability of the identification cannot meet the actual requirements.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a method, an apparatus, a device and a storage medium for detecting right branch block, which are used for performing feature analysis based on the waveform characteristics of an electrocardiographic signal of right branch block, so as to improve the recognition efficiency and accuracy of the right branch block waveform.
The embodiment of the invention provides a right bundle branch block detection method, which comprises the following steps:
acquiring a first electrocardiosignal of a target object, and performing signal preprocessing on the first electrocardiosignal to obtain a second electrocardiosignal for eliminating noise interference;
extracting a QRS complex from the second electrocardiosignal;
extracting at least one predetermined feature from the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the characteristics include at least one of the following: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient;
and inputting the extracted features into at least one pre-established classification model to perform right bundle branch block identification.
Preferably, extracting the QRS complex in the second cardiac signal specifically includes:
performing discrete wavelet decomposition on the second cardiac signal to obtain a predetermined number of wavelet decomposition coefficients;
acquiring a wavelet decomposition coefficient containing larger information and larger energy;
carrying out QRS complex detection according to the obtained wavelet decomposition coefficient containing larger information and larger energy so as to extract QRS complexes; wherein the QRS complex includes a Q wave, an R wave, and an S wave.
Preferably, said extracting at least one predetermined feature according to said QRS complex is:
obtaining a comprehensive threshold value Z according to the wavelet decomposition coefficient;
obtaining QRS complex wave duration according to the sequence difference of the Q wave and the S wave and the sampling frequency;
and acquiring the skewness value of the R wave according to the sequence length of the R wave, the kernel density estimation of the sample observation vector x, the kernel density estimation value vector and operators of skewness and kurtosis.
Preferably, the wavelet decomposition level is 7, and the corresponding wavelet decomposition coefficients are D1, D2, D3, D4, D5, D6, D7 and a 7;
then, obtaining a comprehensive threshold value Z according to the wavelet decomposition coefficient, specifically:
obtaining a threshold coefficient X according to the sum of the wavelet decomposition coefficients D3, D4, D5 and D6;
obtaining a threshold coefficient Y according to the product of the wavelet decomposition coefficients D2, D6, A7 and D3;
obtaining a comprehensive threshold value Z according to the positive value of the product of the threshold value coefficient X and the threshold value coefficient Y; wherein Z ═ X × Y |; x ═ D3+ D4+ D5+ D6;
preferably, before inputting the extracted features into at least one pre-established classification model for right bundle branch block identification, the method further comprises:
establishing a first classification model corresponding to each feature;
establishing a second classification model corresponding to all the characteristics;
inputting the extracted features into at least one pre-established classification model to perform right bundle branch block identification specifically as follows:
inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature;
inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type;
and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
Preferably, the establishing of the second classification model corresponding to all the features specifically includes:
initializing a second classification model; wherein the second classification model is a logistic regression classification model;
reading at least one pair of electrocardiosignal sample pairs to be trained, wherein the electrocardiosignal sample pairs comprise electrocardiosignals and right bundle branch block marks corresponding to the electrocardiosignals; the right bundle branch block identifier is used for identifying whether a right bundle branch block exists in the corresponding electrocardiosignal;
extracting the characteristics of the electrocardiosignals of the electrocardiosignal sample pair to obtain at least one preset characteristic, and forming a training sample pair according to the at least one preset characteristic and the corresponding right bundle branch block identifier;
and taking the at least one preset feature as the input of a second classification model, and taking the corresponding right bundle branch block identifier as the output of the classification model to train the classification model so as to obtain the trained classification model.
Preferably, the method further comprises the following steps:
the first classification model comprises an input processing layer, a comparison layer and an output layer, wherein the input processing layer can obtain an input value of a current sampling point according to a characteristic value input by the current sampling point and a historical characteristic value, the comparison layer can compare the input value of the current sampling point with a threshold value of the current sampling point to generate a comparison result, and meanwhile, the comparison layer generates a threshold value of a next sampling point according to the threshold value of the current sampling point and a threshold value of a previous sampling point; and the output layer can generate a right bundle branch block identifier according to the comparison result.
The embodiment of the invention also provides a right bundle branch block detection device, which comprises:
the device comprises a first electrocardiosignal acquisition unit, a second electrocardiosignal acquisition unit and a first processing unit, wherein the first electrocardiosignal acquisition unit is used for acquiring a first electrocardiosignal of a target object and carrying out signal preprocessing on the first electrocardiosignal so as to obtain a second electrocardiosignal for eliminating noise interference;
a QRS complex extracting unit for extracting a QRS complex from the second electrocardiosignal;
a feature extraction unit, for extracting at least one predetermined feature according to the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the characteristics include at least one of the following: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient;
and the right bundle branch block identification unit is used for inputting the extracted features into at least one pre-established classification model so as to identify the right bundle branch block.
Preferably, the QRS complex extraction unit is specifically configured to:
a discrete wavelet decomposition module for performing discrete wavelet decomposition on the second cardiac signal to obtain a predetermined number of wavelet decomposition coefficients;
the wavelet decomposition coefficient acquisition module is used for acquiring a wavelet decomposition coefficient containing larger information and larger energy;
the QRS complex extraction module is used for carrying out QRS complex detection according to the acquired wavelet decomposition coefficient containing larger information and larger energy so as to extract the QRS complex; wherein the QRS complex includes a Q wave, an R wave, and an S wave.
Preferably, the feature extraction unit is specifically configured to:
the comprehensive threshold value Z generating module is used for obtaining a comprehensive threshold value Z according to the wavelet decomposition coefficient;
the QRS complex duration generation module is used for obtaining QRS complex duration according to the sequence difference of the Q wave and the S wave and the sampling frequency;
and the R wave skewness value generation module is used for acquiring the skewness value of the R wave according to the sequence length of the R wave, the kernel density estimation value of the sample observation vector x, the kernel density estimation value vector and the skewness and kurtosis operators.
Preferably, the wavelet decomposition level is 7, and the corresponding wavelet decomposition coefficients are D1, D2, D3, D4, D5, D6, D7 and a 7;
the integrated threshold Z generation module is specifically configured to:
obtaining a threshold coefficient X according to the sum of the wavelet decomposition coefficients D3, D4, D5 and D6;
obtaining a threshold coefficient Y according to the product of the wavelet decomposition coefficients D2, D6, A7 and D3;
obtaining a comprehensive threshold value Z according to the positive value of the product of the threshold value coefficient X and the threshold value coefficient Y; wherein Z ═ X × Y |; x ═ D3+ D4+ D5+ D6;
preferably, the method further comprises the following steps:
a first classification model establishing unit for establishing a first classification model corresponding to each feature; the second classification model establishing unit is used for establishing a second classification model corresponding to all the characteristics; the right bundle branch block identification unit is specifically configured to:
inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature; inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type; and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
Preferably, the second classification model establishing unit is specifically configured to:
the second classification model initialization module is used for initializing a second classification model; wherein the second classification model is a logistic regression classification model;
the electrocardiosignal sample pair reading module is used for reading at least one pair of electrocardiosignal sample pairs to be trained, wherein the electrocardiosignal sample pairs comprise electrocardiosignals and right bundle branch block marks corresponding to the electrocardiosignals; the right bundle branch block identifier is used for identifying whether a right bundle branch block exists in the corresponding electrocardiosignal;
the training sample pair forming module is used for extracting the characteristics of the electrocardiosignals of the electrocardiosignal sample pair to obtain at least one preset characteristic and forming a training sample pair according to the at least one preset characteristic and the corresponding right bundle branch block identifier;
and the classification model training module is used for taking the at least one preset characteristic as the input of a second classification model and taking the corresponding right bundle branch block identifier as the output of the classification model to train the classification model so as to obtain the trained classification model.
Preferably, the method further comprises the following steps:
the first classification model comprises an input processing layer, a comparison layer and an output layer, wherein the input processing layer can obtain an input value of a current sampling point according to a characteristic value input by the current sampling point and a historical characteristic value, the comparison layer can compare the input value of the current sampling point with a threshold value of the current sampling point to generate a comparison result, and meanwhile, the comparison layer generates a threshold value of a next sampling point according to the threshold value of the current sampling point and a threshold value of a previous sampling point; and the output layer can generate a right bundle branch block identifier according to the comparison result.
An embodiment of the present invention further provides a right bundle branch block detection device, which includes a processor, a memory, and a computer program stored in the memory, where the computer program can be executed by the processor to implement the right bundle branch block detection method according to the foregoing embodiment.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program, where when the computer program runs, the apparatus where the computer-readable storage medium is located is controlled to execute the method for detecting right bundle branch block according to the above embodiment.
In the above embodiment, by acquiring the electrocardiosignal of the target object, extracting the QRS complex of the electrocardiosignal with the noise background removed, extracting the features from the QRS complex according to the waveform characteristics of the electrocardiosignal with the right bundle branch block, and inputting the extracted features into at least one pre-established classification model to perform right bundle branch block identification, the identification efficiency and accuracy of the right bundle branch block waveform can be improved, and the actual application requirements can be met.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a right bundle branch block detection method according to a first embodiment of the present invention.
FIG. 2 is a block diagram of an ECG signal including a right bundle branch block according to an embodiment of the present invention.
Fig. 3 is a diagram of electrocardiographic signal conduction according to an embodiment of the present invention.
Fig. 4 is an electrocardiogram of right bundle branch block according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a single-beat signal according to an embodiment of the present invention.
Fig. 6 is a schematic structural diagram of a right bundle branch block detection apparatus according to a third embodiment of the present invention.
Detailed Description
For better understanding of the technical solutions of the present invention, the following detailed descriptions of the embodiments of the present invention are provided with reference to the accompanying drawings.
It should be understood that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the examples of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be understood that the term "and/or" as used herein is merely one type of association that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination" or "in response to a detection", depending on the context. Similarly, the phrases "if determined" or "if detected (a stated condition or event)" may be interpreted as "when determined" or "in response to a determination" or "when detected (a stated condition or event)" or "in response to a detection (a stated condition or event)", depending on the context.
In the embodiments, the references to "first \ second" are merely to distinguish similar objects and do not represent a specific ordering for the objects, and it is to be understood that "first \ second" may be interchanged with a specific order or sequence, where permitted. It should be understood that "first \ second" distinct objects may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced in sequences other than those illustrated or described herein.
Referring to fig. 1 to 5, a first embodiment of the present invention provides a right branch block detection method, which can be performed by a right branch block detection device, and in particular, by one or more processors in the right branch block detection device, and at least includes the following steps:
s101, collecting a first electrocardiosignal of a target object, and performing signal preprocessing on the first electrocardiosignal to obtain a second electrocardiosignal for eliminating noise interference.
In the present embodiment, the right bundle branch block detection apparatus includes an electrocardiograph (e.g., an analog electrocardiograph, a digital intelligent electrocardiograph, etc.) for acquiring an electrocardiographic signal, and a digital filter or the like for performing filter processing on the acquired electrocardiographic signal. The electrocardiograph may include, among other things, electrocardiographic leads and sensors. It should be noted that the target object may be any living organism, such as a human being, an animal, etc., which can generate the first cardiac electric signal.
Specifically, in this embodiment, first, a first electrocardiographic signal of a target object is acquired through the electrocardiographic lead and the sensor, then, the acquired first electrocardiographic signal is subjected to impedance matching, filtering, amplification and other processing through the analog circuit, and then, the analog signal of the human physiological parameter is converted into a digital signal through the analog-to-digital converter and stored in the memory. The first electrocardiosignal acquired by the right bundle branch block detection equipment contains various noises, and the waveform is rough and unsmooth, so that the useful information contained in the QRS wave group in the electrocardiosignal is difficult to extract. The low-pass digital filter (e.g. butterworth filter) may be used to perform low-pass filtering to filter out high-frequency noise (above 300 Hz) and obtain a filtered electrocardiographic signal, i.e. obtain a second electrocardiographic signal with noise interference removed. Referring to fig. 2, fig. 2 is a diagram of an electrocardiograph signal including a right bundle branch block according to an embodiment of the present invention.
It should be noted that the electrocardiosignal conduction process may be, in order: the atrioventricular Node 11 (AV Node), the Bundle of HIS 15(HIS Bundle), the Left Bundle Branch 12 (LBB) or the Right Bundle Branch 16(Right Bundle Branch, the Left Bundle Branch 12 may be followed by the Left Anterior Branch 13(Left Anterior Branch, LAF) and the Left Posterior Branch 14(Left temporal Branch, LPF), and finally the Purkinje fiber 17(Purkinje fibers). The basic pathophysiological defect of the Right Bundle Branch block is mainly due to the fact that the electrical pulse conducted from the Bundle of HIS (HIS Bundle) to the Right Bundle Branch is not conducted.in the case of normal conduction of the Left Bundle Branch, the Right Ventricular depolarization significantly differs from the Left ventricle, which gives the Electrocardiogram (ECG) change.A signal conduction map is provided in the first embodiment of the present invention, see FIG. 3.
Further, electrical signals from the left ventricle must pass through the myocardium and travel slower than the original his bundle-purkinje fiber path, so the QRS complex is wider on the electrocardiogram. The method comprises the following specific steps: (1) supraventricular heart rhythm. The ventricular rhythm must come from above the ventricles (i.e., the sinoatrial node, the atrium, or the atrioventricular node); (2) QRS complex time length. QRS complex times are longer than 100 milliseconds (non-complete occlusion) or 120 milliseconds (complete occlusion); (3) the V1 pole has an end R wave (e.g., R, rR ', rsR', rSR 'or qR waveform) and an RSR' pattern ('M-shaped' QRS complex). Referring to fig. 4, fig. 4 is an electrocardiogram of right bundle branch block provided by the embodiment of the present invention, in which the S-wave is wide and fuzzy, which is a typical representation of the right bundle branch block on the electrocardiogram.
And S102, extracting the QRS complex from the second electrocardiosignal.
In this embodiment, the right bundle branch block detecting apparatus may extract waveform information of the P wave, the QRS complex, and the T wave in the second cardiac signal (e.g., the cardiac signal in fig. 3), i.e., a complete heartbeat, which is also referred to as a single-heartbeat signal, by using a wavelet transform technique. Specifically, the right bundle branch retardation detection apparatus performs discrete wavelet decomposition on the second cardiac signal to obtain a predetermined number of wavelet decomposition coefficients, for example, the Daubechies4(Db4) wavelet is used to perform discrete wavelet decomposition on the second cardiac signal, the number of decomposition layers is N, and then N +1 number of wavelet decomposition coefficients are obtained, for facilitating understanding of the present invention, the following embodiments are described by taking N ═ 7 (best practical performance) as an example, and the wavelet decomposition coefficients are D1, D2, D3, D4, D5, D6, D7, and a7 as examples.
Further, the right bundle branch block detection device acquires a wavelet decomposition coefficient containing larger information and larger energy, and performs QRS complex detection according to the acquired wavelet decomposition coefficient containing larger information and larger energy to extract a QRS complex; wherein the QRS complex includes a Q wave, an R wave, and an S wave. For example, for the case of N being 7, generally speaking, D3, D4 and D5 in the wavelet decomposition coefficients contain the maximum information and the maximum energy, so the QRS complex is detected by using the wavelet decomposition coefficients D3, D4 and D5, and the detected QRS complex is as shown in fig. 5, fig. 5 is a schematic diagram of a single-beat signal provided by the embodiment of the present invention, where the abscissa is time, the second (S), the ordinate is voltage, and the V is a unit, the single-beat signal in the diagram includes a P wave, a QRS complex, a T wave and a U wave, where the P wave reflects the electrical activation process of the left atrium and the right atrium, the first half is mainly generated by the right atrium, the second half is mainly generated by the left atrium, the width of the P wave of a normal person is not more than 0.11S, the maximum amplitude is not more than 2.5mv, the P-R segment reflects the process of the activation of the heart from the atrium to the ventricle through the ventricular conduction system along the atrial muscle after the P wave appears, down to the ventricles. The potential effect produced by activation through this section of conducting tissue is extremely weak, so that after the P-wave, there is a period of time before ventricular activation where no potential effect is produced, this is the P-R section. The QRS complex reflects the electrical activation of the left and right ventricles, and the width of the entire QRS complex, called the QRS interval duration, represents the time required for the activation of all ventricular muscles, and normal persons do not exceed 0.1S at most. The S-T segment is at one end of the beginning of the T wave from the end point of the QRS complex, and the S-T segment of a normal person is close to the baseline and the passing time is not more than 0.1S. The T wave represents the potential influence generated during the recovery after the ventricular activation, and on an electrocardiogram mainly comprising the R wave, the T wave is not lower than one tenth of the R wave. The U-wave represents the process of the excited ventricles returning to the resting phase, and the U-wave is very small in normal humans, it being understood that it is not necessary to include it.
S103, extracting at least one preset feature according to the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the characteristics include at least one of the following: duration of QRS complex, skewness value of R wave, and composite threshold value Z generated according to wavelet decomposition coefficient.
In this embodiment, in order to facilitate understanding of the present invention, the following description will be made of features selected from the waveform characteristics of the electrocardiographic signal of the right bundle branch block:
duration of QRS complex: morphological features of the ECG, including duration-related features of the QRS complex (for analysis of the duration of the QRS wave), are extracted for right bundle branch blocked electrocardiographic signal (ECG) waveform features (QRS complex duration >120 ms). Let X _ P, X _ Q, X _ R, X _ S and X _ T denote the sequence of P, Q, R, S and T waves, the QRS interval can be denoted as QRS _ width, and the QRS complex duration is obtained from the sequence difference of Q and S waves and the sampling frequency; wherein the QRS interval duration calculation formula is as follows: QRS _ width ═ X _ S-X _ Q)/fs where fs represents the sampling frequency.
Second, deviation value of R wave: morphological characteristics of an Electrocardiogram (ECG) are extracted according to waveform characteristics (end-stage R waves (such as R, rR ', rsR ', rSR ' or qR and other waveforms)) of right bundle branch block, wherein the morphological characteristics comprise probability density distribution of the R waves (used for analyzing whether the R waves are monophasic or biphasic). Let X _ P, X _ Q, X _ R, X _ S and X _ T denote the sequence of P, Q, R, S and T waves, then the skewness value skew _ R of the R wave is obtained according to the sequence length of the R wave, the kernel density estimation of the sample observation vector X, the kernel density estimation value vector, and the operators of skewness and kurtosis.
Wherein ecg represents an electrocardiosignal sequence, ksdensity is an operator for calculating the nuclear density estimation of the sequence and is used for solving the nuclear density estimation of a sample observation vector x, xi is a vector formed by 100 points selected at equal intervals in the value range of x, and f1 is a nuclear density estimation value vector corresponding to xi. The kernel function used is a Gaussian kernel function; skewness and kurtosis are operators for calculating sequence length, skewness and kurtosis respectively. Then [ f1, xi ] ═ ksdensity (X _ R), kurt _ R ═ kurtosis (f1), and skew _ R ═ skew (f 1).
Thirdly, generating a comprehensive threshold value Z according to the wavelet decomposition coefficient: after spectrum analysis, the wavelet coefficient between the normal QRS complex and the right bundle branch block is found to have larger difference. Therefore, firstly, designing a wavelet coefficient threshold, determining through multiple tests, and setting a threshold coefficient X, which is called a wavelet coefficient sum, as shown in formula 1; a threshold coefficient Y, called the wavelet coefficient product, is then set, as shown in equation 2. Experiments show that when the right bundle branch block occurs, the numerical value of the two is obviously higher than that of a normal electrocardiogram waveform, and in order to amplify the change, the right bundle branch block can be detected more effectively, a brand new comprehensive threshold value Z is designed, as shown in formula 3. Thus, according to the wavelet decomposition coefficients, a threshold coefficient X generated according to the wavelet decomposition coefficients is obtained; obtaining a threshold coefficient Y generated according to the wavelet decomposition coefficient; and generating a comprehensive threshold value Z according to the threshold value system X and the threshold value coefficient Y.
X＝D3+D4+D5+D6 (1)
Z＝|X×Y|. (3)
And S104, inputting the extracted features into at least one pre-established classification model to identify the right bundle branch block.
In this embodiment, a classification model may be formed according to neural networks, mathematical programming, genetic algorithms, and machine learning training, and then the extracted features are input into the corresponding classification model to identify whether a right bundle branch block or a non-right bundle branch block.
In summary, by acquiring the first electrocardiosignal of the target object, extracting the QRS complex of the second electrocardiosignal after interference noise elimination, then extracting features from the QRS complex according to the waveform characteristics of the electrocardiosignal of the right bundle branch block, and inputting the extracted features into at least one pre-established classification model to identify the right bundle branch block, the identification efficiency and accuracy of the right bundle branch block can be improved.
On the basis of the first embodiment, in a preferred embodiment of the present invention, before inputting the extracted features into at least one pre-established classification model for right bundle branch block identification, the method further includes:
establishing a first classification model corresponding to each feature;
in this embodiment, the first classification model includes an input processing layer, a comparison layer and an output layer, the input processing layer can obtain an input value of a current sampling point according to a feature value input by the current sampling point and a historical feature value, the comparison layer can compare the input value of the current sampling point with a threshold of the current sampling point to generate a comparison result, and the comparison layer generates a threshold of a next sampling point according to the threshold of the current sampling point and a threshold of a previous sampling point; and the output layer can generate a right bundle branch block identifier according to the comparison result.
Specifically, the process of establishing the first classification model according to the features of the comprehensive threshold Z is as follows: when the number of the leaders is M, extracting M Z values, calculating an average value Zmean,
zs represents the Z value of the s-th lead, and Zmeanq is the Zmean value obtained by the q-th calculation in the continuous monitoring process; then, setting the initial threshold value of Zmean to Zmean D0 (empirical parameter obtained by a large number of experiments), Zmean Dq represents the qth threshold value, and calculating the Zmean q value according to the average value Zmean; calculating Zmeandq from the Zmeaandq; zmeandq+1＝λ1ZmeanDq+μ1ZmeanDq-1(ii) a Wherein ZmeanDq is the qth Z threshold; wherein λ and μ are parameters and satisfy λ1+μ11, preferably λ1＝0.85，μ10.15; and then when the Zmeanandq value obtained by real-time detection is judged to be larger than ZmeanandDq, judging that the acquired electrocardiosignal is the right bundle branch block.
Specifically, the process of establishing the first classification model for the characteristics of QRS complex duration (QRS _ width) is as follows: when the number of the leads is M, extracting M QRS _ width values, calculating the average value and QRS _ width mean value
Wherein QRS _ width represents the QRS _ width value of the s-th lead, and QRS _ width means q represents the QRS _ width value obtained by q-th calculation in the continuous monitoring process; then calculating a QRS _ width mean value according to the average QRS _ width mean value; finally, setting the initial QRS _ width mean threshold to QRS _ width mean 0 (here 120ms), QRS _ width mean Dq representing the qth QRS _ width mean threshold, calculating the QRS _ width mean value based on the QRS _ width mean value, QRS _ width meanq+1＝λ2QRS_widthmeanDq+μ2QRS_widthmeanDq-1(ii) a Where λ and μ are parameters, λ2+μ21. Preferably, λ2＝0.85，μ20.15; then, when judged to detect in real timeAnd when the obtained QRS _ width mean dq value is larger than the QRS _ width mean Dq, judging the electrocardiosignal as right bundle branch block.
Specifically, the process of establishing the first classification model for the skew value (skew _ R) characteristic of the R wave is as follows: when the number of the pilot connections is M, extracting M skew _ R values, and calculating an average value skew _ Rmean;
wherein the skew _ Rs represents a skew _ R value of the s-th lead, and the skew _ Rmean q represents a skew _ Rmean value obtained by the q-th calculation in the continuous monitoring process; setting an initial threshold of the skew _ Rmean to be skew _ Rmean D0 (obtained by a large number of experiments), wherein the skew _ Rmean Dq represents a q-th skew _ R threshold, calculating a skew _ Rmean Dq value according to the average value, and calculating the skew _ Rmean Dq and the skew _ Rmean D according to the skew _ Rmean D valueq+1＝λ3skew_RmeanDq+μ3skew_RmeanDq-1(ii) a Wherein the skew _ Rmean Dq represents the qth skew _ R threshold, λ and μ are parameters, λ3+μ31, preferably λ3＝0.8，μ30.2; and when the value of the skew _ Rmean dq obtained by real-time detection is judged to be larger than the value of the skew _ Rmean dq, the electrocardiosignal is a right bundle branch block.
Establishing a second classification model corresponding to all the characteristics;
specifically, a second classification model is initialized; wherein the second classification model may be a logistic regression classification model; then reading at least one pair of electrocardiosignal sample pairs to be trained, wherein the electrocardiosignal sample pairs comprise electrocardiosignals and right bundle branch block marks corresponding to the electrocardiosignals; the right bundle branch block identifier is used for identifying whether a right bundle branch block exists in the corresponding electrocardiosignal; secondly, extracting the characteristics of the electrocardiosignals of the electrocardiosignal sample pair to obtain at least one preset characteristic, and forming a training sample pair according to the at least one preset characteristic and the corresponding right bundle branch block identifier; and finally, taking the at least one preset feature as the input of a second classification model, and taking the corresponding right bundle branch block identifier as the output of the classification model to train the classification model so as to obtain the trained classification model. For example, the extracted features are first used as input samples X for training LR, and the right bundle branch block or non-right bundle branch block is labeled as output Y of LR. (X, Y) together form a training sample pair of LR for LR training. And (3) inputting the extracted features into the model as input samples X of the training LR by using the LR model obtained by training, and identifying (namely right bundle branch block or non-right bundle branch block).
Inputting the extracted features into at least one pre-established classification model to perform right bundle branch block identification specifically as follows:
inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature;
inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type;
and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
For example, for an acquired first electrocardiosignal d, three features of the acquired first electrocardiosignal d are extracted, the three features are sequentially input into a first classification model for recognition to obtain one recognition type a and two recognition types B (assuming that a is right bundle branch block recognition and B is non-right bundle branch block recognition), then the three features of the first electrocardiosignal d are input into a second classification model for recognition to obtain one recognition type B, and then the obtained maximum number of recognition types, namely B, is used as the final class of the first electrocardiosignal d.
According to the method, different characteristics of the same electrocardiosignal are respectively identified and integrally identified, and a final judgment result is obtained according to the identification result, so that the identification accuracy of the right bundle branch block waveform can be improved.
Second embodiment of the invention:
referring to fig. 6, the second embodiment of the present invention further provides a right bundle branch block detection apparatus, including:
the first electrocardiosignal acquisition unit 100 is configured to acquire a first electrocardiosignal of a target object and perform signal preprocessing on the first electrocardiosignal to obtain a second electrocardiosignal from which noise interference is eliminated;
a QRS complex extracting unit 200 for extracting a QRS complex from the second electrocardiographic signal;
a feature extraction unit 300, configured to extract at least one predetermined feature according to the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the characteristics include at least one of the following: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient;
and a right bundle branch block identification unit 400, configured to input the extracted features into at least one pre-established classification model to perform right bundle branch block identification.
Based on the second embodiment, in a preferred embodiment of the present invention, the QRS complex extracting unit 200 specifically includes:
a discrete wavelet decomposition module for performing discrete wavelet decomposition on the second cardiac signal to obtain a predetermined number of wavelet decomposition coefficients;
the wavelet decomposition coefficient acquisition module is used for acquiring a wavelet decomposition coefficient containing larger information and larger energy;
the QRS complex extraction module is used for carrying out QRS complex detection according to the acquired wavelet decomposition coefficient containing larger information and larger energy so as to extract the QRS complex; wherein the QRS complex includes a Q wave, an R wave, and an S wave.
On the basis of the second embodiment, in a preferred embodiment of the present invention, the feature extraction unit 300 specifically includes:
the comprehensive threshold value Z generating module is used for obtaining a comprehensive threshold value Z according to the wavelet decomposition coefficient;
the QRS complex duration generation module is used for obtaining QRS complex duration according to the sequence difference of the Q wave and the S wave and the sampling frequency;
and the R wave skewness value generation module is used for acquiring the skewness value of the R wave according to the sequence length of the R wave, the kernel density estimation value of the sample observation vector x, the kernel density estimation value vector and the skewness and kurtosis operators.
On the basis of the second embodiment, in a preferred embodiment of the present invention, the number of wavelet decomposition layers is 7, and the corresponding wavelet decomposition coefficients are D1, D2, D3, D4, D5, D6, D7, and a 7;
the integrated threshold Z generation module is specifically configured to:
obtaining a threshold coefficient X according to the sum of the wavelet decomposition coefficients D3, D4, D5 and D6;
obtaining a threshold coefficient Y according to the product of the wavelet decomposition coefficients D2, D6, A7 and D3;
obtaining a comprehensive threshold value Z according to the positive value of the product of the threshold value coefficient X and the threshold value coefficient Y; wherein Z ═ X × Y |; x ═ D3+ D4+ D5+ D6;
on the basis of the second embodiment, a preferred embodiment of the present invention further includes:
a first classification model establishing unit for establishing a first classification model corresponding to each feature;
the second classification model establishing unit is used for establishing a second classification model corresponding to all the characteristics;
the right bundle branch block identification unit is specifically configured to:
inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature; inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type; and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
On the basis of the second embodiment, in a preferred embodiment of the present invention, the second classification model establishing unit is specifically configured to:
the second classification model initialization module is used for initializing a classification model; wherein the second classification model is a logistic regression classification model;
the electrocardiosignal sample pair reading module is used for reading at least one pair of electrocardiosignal sample pairs to be trained, wherein the electrocardiosignal sample pairs comprise electrocardiosignals and right bundle branch block marks corresponding to the electrocardiosignals; the right bundle branch block identifier is used for identifying whether a right bundle branch block exists in the corresponding electrocardiosignal;
the training sample pair forming module is used for extracting the characteristics of the electrocardiosignals of the electrocardiosignal sample pair to obtain at least one preset characteristic and forming a training sample pair according to the at least one preset characteristic and the corresponding right bundle branch block identifier;
and the classification model training module is used for taking the at least one preset characteristic as the input of a second classification model and taking the corresponding right bundle branch block identifier as the output of the classification model to train the classification model so as to obtain the trained classification model.
On the basis of the second embodiment, in a preferred embodiment of the present invention:
the first classification model comprises an input processing layer, a comparison layer and an output layer, wherein the input processing layer can obtain an input value of a current sampling point according to a characteristic value input by the current sampling point and a historical characteristic value, the comparison layer can compare the input value of the current sampling point with a threshold value of the current sampling point to generate a comparison result, and meanwhile, the comparison layer generates a threshold value of a next sampling point according to the threshold value of the current sampling point and a threshold value of a previous sampling point; and the output layer can generate a right bundle branch block identifier according to the comparison result.
Third embodiment of the invention:
the third embodiment of the present invention also provides a right bundle branch block detection device, which includes a processor, a memory, and a computer program stored in the memory, the computer program being executable by the processor to implement the right bundle branch block detection method as described in the above embodiments.
The fourth embodiment of the present invention:
a fourth embodiment of the present invention provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program runs, the apparatus on which the computer-readable storage medium is located is controlled to execute the method for detecting right bundle branch block as described above.
Illustratively, the computer program may be divided into one or more units, which are stored in the memory and executed by the processor to accomplish the present invention. The one or more units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution of the computer program in the right bundle branch block detection device.
The right branch block detection device may include, but is not limited to, a processor, a memory. It will be understood by those skilled in the art that the schematic diagram is merely an example of a right branch block detection device, and does not constitute a limitation of the right branch block detection device, and may include more or less components than those shown, or combine some components, or different components, for example, the right branch block detection device may further include an input-output device, a network access device, a bus, etc.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the control center of the right branch block detection device connects the various parts of the entire right branch block detection device using various interfaces and lines.
The memory may be used to store the computer programs and/or modules, and the processor may implement the various functions of the right bundle branch block detection apparatus by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Wherein, the unit integrated with the right bundle branch block detection device can be stored in a computer readable storage medium if the unit is realized in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
It should be noted that the above-described device embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (8)
1. A right bundle branch block detection device, comprising:
the device comprises a first electrocardiosignal acquisition unit, a second electrocardiosignal acquisition unit and a first processing unit, wherein the first electrocardiosignal acquisition unit is used for acquiring a first electrocardiosignal of a target object and carrying out signal preprocessing on the first electrocardiosignal so as to obtain a second electrocardiosignal for eliminating noise interference;
a QRS complex extracting unit for extracting a QRS complex from the second electrocardiosignal;
a feature extraction unit, which is used for extracting at least two preset features according to the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the features include at least two of: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient;
the right bundle branch block identification unit is used for inputting the extracted features into at least one pre-established classification model so as to identify the right bundle branch block; wherein, still include:
a first classification model establishing unit for establishing a first classification model corresponding to each feature;
the second classification model establishing unit is used for establishing a second classification model corresponding to all the characteristics;
the right bundle branch block identification unit is specifically configured to:
inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature; inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type; and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
2. A right bundle branch block detection device comprising a processor, a memory and a computer program stored in the memory, the computer program being executable by the processor to perform the steps of:
acquiring a first electrocardiosignal of a target object, and performing signal preprocessing on the first electrocardiosignal to obtain a second electrocardiosignal for eliminating noise interference;
extracting a QRS complex from the second electrocardiosignal;
extracting at least two preset features according to the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the features include at least two of: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient;
establishing a first classification model corresponding to each feature;
establishing a second classification model corresponding to all the characteristics;
inputting the extracted features into at least one pre-established classification model to identify right bundle branch block; the method specifically comprises the following steps: inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature; inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type; and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
3. The right bundle branch block detecting device according to claim 2, wherein extracting the QRS complex in the second cardiac signal specifically comprises:
performing discrete wavelet decomposition on the second cardiac signal to obtain a predetermined number of wavelet decomposition coefficients;
acquiring a wavelet decomposition coefficient containing larger information and larger energy;
carrying out QRS complex detection according to the obtained wavelet decomposition coefficient containing larger information and larger energy so as to extract QRS complexes; wherein the QRS complex includes a Q wave, an R wave, and an S wave.
4. The right bundle branch block detecting device according to claim 3, wherein said extracting at least one predetermined feature from said QRS complex is specifically:
obtaining a comprehensive threshold value Z according to the wavelet decomposition coefficient;
obtaining QRS complex wave duration according to the sequence difference of the Q wave and the S wave and the sampling frequency;
and acquiring the skewness value of the R wave according to the sequence length of the R wave, the kernel density estimation of the sample observation vector x, the kernel density estimation value vector and operators of skewness and kurtosis.
5. The right bundle branch block detecting apparatus according to claim 4, wherein the number of wavelet decomposition layers is 7, and corresponding wavelet decomposition coefficients are D1, D2, D3, D4, D5, D6, D7, and a 7;
then, obtaining a comprehensive threshold value Z according to the wavelet decomposition coefficient, specifically:
obtaining a threshold coefficient X according to the sum of the wavelet decomposition coefficients D3, D4, D5 and D6;
obtaining a threshold coefficient Y according to the product of the wavelet decomposition coefficients D2, D6, A7 and D3;
obtaining a comprehensive threshold value Z according to the positive value of the product of the threshold value coefficient X and the threshold value coefficient Y; wherein Z ═ X × Y |; x ═ D3+ D4+ D5+ D6;
6. the right bundle branch block detecting apparatus according to claim 2,
the establishing of the second classification model aiming at all the characteristics specifically comprises the following steps:
initializing a second classification model; wherein the second classification model is a logistic regression classification model;
reading at least one pair of electrocardiosignal sample pairs to be trained, wherein the electrocardiosignal sample pairs comprise electrocardiosignals and right bundle branch block marks corresponding to the electrocardiosignals; the right bundle branch block identifier is used for identifying whether a right bundle branch block exists in the corresponding electrocardiosignal;
extracting the characteristics of the electrocardiosignals of the electrocardiosignal sample pair to obtain at least one preset characteristic, and forming a training sample pair according to the at least one preset characteristic and the corresponding right bundle branch block identifier;
and taking the at least one preset feature as the input of a second classification model, and taking the corresponding right bundle branch block identifier as the output of the classification model to train the classification model so as to obtain the trained second classification model.
7. The right bundle branch block detecting device according to claim 2, wherein the first classification model includes an input processing layer, a comparison layer and an output layer, the input processing layer can obtain the input value of the current sampling point according to the input characteristic value of the current sampling point and the historical characteristic value, the comparison layer can compare the input value of the current sampling point with the threshold value of the current sampling point to generate a comparison result, and the comparison layer generates the threshold value of the next sampling point according to the threshold value of the current sampling point and the threshold value of the previous sampling point; and the output layer can generate a right bundle branch block identifier according to the comparison result.
8. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the steps of:
acquiring a first electrocardiosignal of a target object, and performing signal preprocessing on the first electrocardiosignal to obtain a second electrocardiosignal for eliminating noise interference;
extracting a QRS complex from the second electrocardiosignal;
extracting at least two preset features according to the QRS complex; wherein the characteristic is selected according to the waveform characteristics of the electrocardiosignals of the right bundle branch block; the features include at least two of: duration of QRS complex wave, skewness value of R wave and comprehensive threshold value Z generated according to wavelet decomposition coefficient;
establishing a first classification model corresponding to each feature;
establishing a second classification model corresponding to all the characteristics;
inputting the extracted features into at least one pre-established classification model to identify right bundle branch block; the method specifically comprises the following steps: inputting the extracted features into the corresponding first classification models respectively to obtain a first identification type corresponding to each feature; inputting all the extracted features into a second classification model together to perform right bundle branch block identification to obtain a second identification type; and obtaining a recognition result which is used as the right bundle branch block and has a larger number according to the recognition types according to the first recognition type and the second recognition type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910754379.8A CN110367968B (en) | 2019-08-15 | 2019-08-15 | Right bundle branch retardation detection method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910754379.8A CN110367968B (en) | 2019-08-15 | 2019-08-15 | Right bundle branch retardation detection method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110367968A CN110367968A (en) | 2019-10-25 |
| CN110367968B true CN110367968B (en) | 2022-04-15 |
Family
ID=68259559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910754379.8A Active CN110367968B (en) | 2019-08-15 | 2019-08-15 | Right bundle branch retardation detection method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110367968B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111449646B (en) * | 2020-03-18 | 2023-05-02 | 广州视源电子科技股份有限公司 | Electrocardiogram identification method and device, storage medium and electronic equipment |
| CN111449647B (en) * | 2020-03-18 | 2023-05-02 | 广州视源电子科技股份有限公司 | Electrocardiogram identification method and device, storage medium and electronic equipment |
| CN111612042B (en) * | 2020-04-24 | 2023-09-08 | 广州视源电子科技股份有限公司 | Detection device for left front branch conduction block, storage medium and terminal |
| CN111887836B (en) * | 2020-07-08 | 2023-11-14 | 广州视源电子科技股份有限公司 | Electrocardiosignal processing method, device and equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108937916A (en) * | 2018-08-03 | 2018-12-07 | 西南大学 | A kind of electrocardiograph signal detection method, device and storage medium |
| CN109475741A (en) * | 2016-05-31 | 2019-03-15 | 美敦力公司 | The control based on electrogram to cardiac resynchronization therapy |
| CN109770860A (en) * | 2019-03-28 | 2019-05-21 | 广州视源电子科技股份有限公司 | A kind of electrocardiogram signal processing device and electrocardio equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10426364B2 (en) * | 2015-10-27 | 2019-10-01 | Cardiologs Technologies Sas | Automatic method to delineate or categorize an electrocardiogram |
-
2019
- 2019-08-15 CN CN201910754379.8A patent/CN110367968B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109475741A (en) * | 2016-05-31 | 2019-03-15 | 美敦力公司 | The control based on electrogram to cardiac resynchronization therapy |
| CN108937916A (en) * | 2018-08-03 | 2018-12-07 | 西南大学 | A kind of electrocardiograph signal detection method, device and storage medium |
| CN109770860A (en) * | 2019-03-28 | 2019-05-21 | 广州视源电子科技股份有限公司 | A kind of electrocardiogram signal processing device and electrocardio equipment |
Non-Patent Citations (2)
| Title |
|---|
| Automatic ECG arrhythmia classification using dual tree complexwavelet based features;Manu Thomas et.al.;《International Journal of Electronics and Communications》;20150430;第69卷(第4期);第715-721页 * |
| ECG beat classification using PCA, LDA, ICA and Discrete Wavelet Transform;Roshan Joy Martis et.al.;《Biomedical Signal Processing and Control》;20130930;第8卷(第5期);第437-448页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110367968A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110367968B (en) | Right bundle branch retardation detection method, device, equipment and storage medium | |
| Javadi et al. | Classification of ECG arrhythmia by a modular neural network based on mixture of experts and negatively correlated learning | |
| Beritelli et al. | A novel training method to preserve generalization of RBPNN classifiers applied to ECG signals diagnosis | |
| CN109770860B (en) | Electrocardiosignal processing device and electrocardio equipment | |
| Prasad et al. | Classification of ECG arrhythmias using multi-resolution analysis and neural networks | |
| CN110367936B (en) | Electrocardiosignal detection method and device | |
| Zhang et al. | ECG signal classification with deep learning for heart disease identification | |
| CN109077720B (en) | Signal processing method, device, equipment and storage medium | |
| Karimifard et al. | Morphological heart arrhythmia detection using hermitian basis functions and kNN classifier | |
| CN110717416B (en) | Neural network training method for ST segment classification recognition based on feature selection | |
| CN110226919B (en) | Electrocardiosignal type detection method and device, computer equipment and storage medium | |
| Sumathi et al. | A wavelet transform based feature extraction and classification of cardiac disorder | |
| CN109770861B (en) | Method, device, equipment and storage medium for training and detecting cardioelectric rhythm model | |
| CN110717415A (en) | ST-segment classification convolutional neural network based on feature selection and using method thereof | |
| Wu et al. | A novel features learning method for ECG arrhythmias using deep belief networks | |
| Perez et al. | The application of supervised learning through feed-forward neural networks for ECG signal classification | |
| Butt et al. | Classifying normal sinus rhythm and cardiac arrhythmias in ECG signals using statistical features in temporal domain | |
| KR102216047B1 (en) | Method and apparatus for classifying cardiac arrhythmia using an auto associative neural network | |
| CN111956202B (en) | Training method and model of convolutional neural network based on sample distribution improvement | |
| Chakroborty | Accurate Arrhythmia classification using auto-associative neural network | |
| Rao et al. | Performance identification of different heart diseases based on neural network classification | |
| Zubair et al. | Removal of motion artifacts from ECG signals by combination of recurrent neural networks and deep neural networks | |
| Gliner et al. | Identification of features for machine learning analysis for automatic arrhythmogenic event classification | |
| Ali et al. | Robust arrhythmia classifier using hybrid multilayered perceptron network | |
| Rodrigues et al. | The issue of automatic classification of heartbeats |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |