CN116135147B

CN116135147B - Electrocardiosignal processing method and device, electronic equipment and storage medium

Info

Publication number: CN116135147B
Application number: CN202111361859.1A
Authority: CN
Inventors: 赵巍; 胡静; 马云驹; 李振齐
Original assignee: Guangzhou Shiyuan Electronics Thecnology Co Ltd; Guangzhou Shiyuan Artificial Intelligence Innovation Research Institute Co Ltd
Current assignee: Guangzhou Shiyuan Electronics Thecnology Co Ltd; Guangzhou Shiyuan Artificial Intelligence Innovation Research Institute Co Ltd
Priority date: 2021-11-17
Filing date: 2021-11-17
Publication date: 2024-10-18
Anticipated expiration: 2041-11-17
Also published as: CN116135147A

Abstract

The invention relates to a method and a device for processing electrocardiosignals, electronic equipment and a storage medium. The electrocardiosignal processing method comprises the following steps: acquiring an original electrocardiosignal, carrying out noise reduction treatment on the original electrocardiosignal to obtain a sample signal, and detecting the sample signal to obtain a heart beat segment; based on preset ST-T wave abnormal characteristic parameters, carrying out ST-T wave abnormal identification on the heart shooting segments to obtain ST-T wave abnormal classification results; based on preset Q wave abnormal characteristic parameters, carrying out Q wave abnormal identification on the heart shooting segments to obtain Q wave abnormal classification results; and determining probability data of the myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result. The electrocardiosignal processing method has the advantages of being capable of accurately identifying myocardial infarction and high in reliability of identification results.

Description

Electrocardiosignal processing method and device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of signal processing technologies, and in particular, to a method and apparatus for processing an electrocardiograph signal, an electronic device, and a storage medium.

Background

The electrocardiosignals reflect the electrophysiological process of heart activity, so cardiovascular abnormality can be found as early as possible by long-time monitoring of the electrocardiosignals so as to be convenient for timely treatment. Normal electrocardiographic signals are typically composed of P-waves, QRS-complexes and T-waves, sometimes also U-waves. Wherein the P wave represents the electrical activity of atrial contraction, the QRS complex and T represent the electrical activity of ventricular contraction, and cardiovascular health can be assessed by the morphology, amplitude and duration of each of the electrocardiographic waveforms.

Myocardial infarction is a malignant disease of myocardial necrosis caused by myocardial ischemia due to thrombus formation in the coronary vessels of the heart. After myocardial infarction, the electrocardiogram of most patients may show specific changes, such as ischemia type T wave, injury type ST segment and necrosis type Q wave. In a general myocardial infarction recognition and classification method, features of each waveform are recognized through a simple neural network model, but since an electrocardiographic abnormality of myocardial infarction is likely to be similar to waveform abnormality caused by other lesions or variations, the feature recognition through the simple neural network model may have the problems that the waveform abnormality recognition is not accurate enough, the reliability of probability data output by the model is not high and the like.

Disclosure of Invention

In order to overcome the problems in the related art, the embodiment of the invention provides a processing method, a device, electronic equipment and a storage medium of electrocardiosignals, wherein the processing method of electrocardiosignals has the advantages of being capable of accurately identifying myocardial infarction and high in reliability of identification results.

According to a first aspect of an embodiment of the present invention, there is provided a processing method of an electrocardiograph signal, including the steps of:

Acquiring an original electrocardiosignal, carrying out noise reduction treatment on the original electrocardiosignal to obtain a sample signal, and detecting the sample signal to obtain a heart beat segment;

based on preset ST-T wave abnormal characteristic parameters, carrying out ST-T wave abnormal identification on the heart shooting segments to obtain ST-T wave abnormal classification results;

based on preset Q wave abnormal characteristic parameters, carrying out Q wave abnormal identification on the heart shooting segments to obtain Q wave abnormal classification results;

And determining probability data of the myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result.

According to a second aspect of an embodiment of the present invention, there is provided an electrocardiograph signal processing device including:

The acquisition module is used for acquiring an original electrocardiosignal, carrying out noise reduction treatment on the original electrocardiosignal to obtain a sample signal, and carrying out heart beat detection on the sample signal to obtain a heart beat segment;

The first classification module is used for carrying out ST-T wave abnormality identification on the heart shooting segments based on preset ST-T wave abnormality characteristic parameters to obtain ST-T wave abnormality classification results;

The second classification module is used for carrying out Q wave abnormality recognition on the heart shooting segment based on preset Q wave abnormality characteristic parameters to obtain Q wave abnormality classification results;

And the determining module is used for determining probability data of the myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result.

According to a third aspect of embodiments of the present invention, there is provided an electronic device including: at least one processor and at least one memory; wherein the memory is configured to store one or more computer programs adapted to be loaded by the processor and to perform the method for processing an electrocardiographic signal according to any one of the embodiments above.

According to a fourth aspect of embodiments of the present invention, there is provided an arithmetic machine readable storage medium, on which an arithmetic machine program is stored, characterized in that the arithmetic machine program, when executed by a processor, implements the method for processing an electrocardiographic signal according to any one of the embodiments above.

By applying the technical scheme, the original electrocardiosignal is subjected to noise reduction treatment, heart beat detection is carried out to obtain heart beat fragments, ST-T wave abnormality identification is carried out on the heart beat fragments based on preset ST-T wave abnormality characteristic parameters to obtain ST-T wave abnormality classification results, Q wave abnormality identification is carried out on the heart beat fragments based on preset Q wave abnormality characteristic parameters to obtain Q wave abnormality classification results, and finally probability data of a heart peduncles stage are determined by combining the ST-T wave abnormality classification results and the Q wave abnormality classification results.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.

In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Drawings

FIG. 1 is a schematic waveform diagram of an electrocardiographic signal according to an alternative embodiment of the present invention;

FIG. 2 is a schematic diagram showing an electrocardiogram during myocardial infarction according to an alternative embodiment of the present invention;

FIG. 3 is a flow chart of a method for processing an electrocardiograph signal according to an alternative embodiment of the present invention;

FIG. 4 is a flowchart of step S2 of a method for processing an electrocardiograph signal according to an alternative embodiment of the present invention;

FIG. 5 is a block diagram of a neural network for ST-T wave anomaly identification, according to an alternative embodiment of the present invention;

FIG. 6 is a flow chart of a method for obtaining ST-T waveform feature vectors according to one embodiment of the present invention;

FIG. 7 is a flow chart of a method for obtaining mask fragments according to one embodiment of the invention;

FIG. 8 is a flowchart of a method for acquiring ST-T wave abnormal characteristic parameters according to an embodiment of the present invention;

fig. 9 is a flowchart of step S3 of a method for processing an electrocardiograph signal according to an embodiment of the present invention;

Fig. 10 is a schematic structural diagram of an electrocardiograph signal processing device according to an embodiment of the present invention;

fig. 11 is a schematic structural diagram of an electronic device according to an alternative embodiment of the present invention.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of apparatus and methods consistent with aspects of the invention as detailed in the accompanying claims.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification 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 also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the invention. The word "if"/"if" as used herein may be interpreted as "at … …" or "at … …" or "in response to a determination", depending on the context.

As shown in fig. 1, a normal electrocardiograph signal is generally composed of a P wave, a QRS complex, and a T wave, and sometimes a U wave.

P-waves are atrial depolarization waves that represent the excitation of the left and right atria. Since the sinus node is located under the right atrium, activation is transmitted first to the right atrium and later to the left atrium. The depolarization of the right atrium is thus also completed slightly earlier than the left atrium. Clinically, for practical purposes, the front of the P-wave represents the activation of the right atrium and the back represents the activation of the left atrium.

The QRS complex reflects the changes in left and right ventricular depolarization potentials and time, the first downward wave being the Q wave, the upward wave being the R wave, then the downward wave being the S wave, the time from the start of the QRS complex to the end of the QRS complex being the QRS time limit. The process of T-wave generation is the process of ventricular repolarization. Normal T wave has long ascending branch, short descending branch and rounded peak.

T-wave abnormalities are one of the most common and common phenomena in electrocardiographic changes, and some T-wave changes may be related to temperature, emotion and nerves, and are physiological changes. However, some T-wave variations, although quite small, are important clues for diagnosing heart disease. Abnormal height of T wave is mainly seen in the super acute injury stage of myocardial infarction, hyperkalemia, increased vagal tension and the like; t-wave abnormality is found deeply in acute subendocardial myocardial infarction, acute myocardial infarction in the transformation phase, severe electrolyte disorder, etc.

The U wave is a smaller wave in an Electrocardiogram (ECG), and is a wave that appears wide and low 0.02 to 0.04 seconds after the T wave. It is believed that the negative back potential that may be generated by each part during diastole is also believed to be the result of repolarization of purkinje fibers.

Myocardial infarction is a malignant disease of myocardial necrosis caused by myocardial ischemia due to thrombus formation in the coronary vessels of the heart. The electrocardiogram of most patients may be changed with specificity after the occurrence of the myocardial infarction, as shown in fig. 2, and the electrocardiogram of the patients may be changed with ischemic T wave, damaged ST segment, necrotic Q wave, etc.

In the prior art, there is a processing method of electrocardiosignals, which identifies the characteristics of each waveform through a simple neural network model. However, since the electrocardiographic abnormality of myocardial infarction may be similar to waveform abnormality caused by other lesions or variations, feature recognition by a simple neural network model may have problems such as inaccurate waveform abnormality recognition and low reliability of probability data output by the model.

Based on the above, the embodiment of the invention provides a processing method, a device, electronic equipment and a storage medium of electrocardiosignals, wherein the processing method of electrocardiosignals has the advantages of being capable of accurately identifying myocardial infarction and high in reliability of identification results.

Various embodiments of the present invention are described below with reference to the accompanying drawings.

Example 1

Referring to fig. 3, fig. 3 is a flow chart of a processing method of an electrocardiograph signal according to an alternative embodiment of the invention. According to the first embodiment, as shown in fig. 3, the method for processing an electrocardiograph signal includes the following steps:

S1: and obtaining an original electrocardiosignal, carrying out noise reduction treatment on the original electrocardiosignal to obtain a sample signal, and detecting the sample signal to obtain a heart beat segment.

In this embodiment, the original electrocardiographic signal is a signal of an electrical activity change generated from each cardiac cycle of the heart recorded from the body surface, and is usually recorded by an Electrocardiogram (ECG) in a graphical manner. In the embodiment of the application, the multichannel synchronous data acquisition and storage of the heart signals, the background noise and the electrocardiosignals of the human body can be utilized. For example, electrocardiosignals can be acquired through electrocardiosignals and sensors, analog signals of physiological parameters of a human body are converted into digital signals through an analog-to-digital converter, and the digital signals are stored by a memory.

When the noise reduction processing is carried out on the original electrocardiosignal, the original electrocardiosignal is resampled to 250HZ, and then high-frequency noise with the frequency of more than 40HZ, such as the electromyographic signal, fair interference and the like, is filtered. When the original electrocardiosignals are subjected to noise reduction treatment, the acquired electrocardiosignals can be subjected to impedance matching, filtering, amplifying and other treatments through an analog circuit. Because the electrocardiosignals obtained by actual collection contain various noises, have rough waveforms and are not smooth, useful information contained in the QRS complex is difficult to extract, a low-pass digital filter (Butterworth filter) can be adopted to carry out low-pass filtering, high-frequency noises (more than 40 Hz) are filtered, and a filtered sample signal is obtained.

When the sample signal is detected to obtain a heart beat segment, heart beat detection is performed on the sample signal to obtain heart beat positions, heart beat types (sinus, supraventricular, ventricular and other types), and reference points (a P wave starting point and a terminal point, a QRS wave group, a starting point and a terminal point of each wavelet and a T wave terminal point of each wave band).

In an alternative embodiment, when detecting the sample signal to obtain the heart beat segment, signal quality evaluation is also performed on the sample signal, and part of the signal is filtered to obtain the heart beat segment with higher quality. Wherein the signal quality assessment comprises the steps of:

If the number of the cardiac beats contained in the sample signal is detected to be more than or equal to three, dividing a cardiac beat segment and a baseline segment from the second cardiac beat to the penultimate cardiac beat by using an electrocardiosignal and an electrocardiosignal baseline between a P wave starting point and a T wave ending point of the cardiac beat;

Counting the energy proportion of the high-frequency part of each heart beat segment, and filtering heart beat segments with the energy proportion exceeding a preset energy threshold; the preset energy threshold may be 30% or some other value.

And (3) counting the peak-peak value difference of the baseline segments, filtering the baseline segments with peak-peak values exceeding the preset peak-peak value threshold value, and obtaining heart beat segments. The preset peak-to-peak threshold may be 0.2mV, or other values.

S2: based on preset ST-T wave abnormal characteristic parameters, ST-T wave abnormal identification is carried out on the heart shooting segments, and ST-T wave abnormal classification results are obtained.

The electrocardiogram of myocardial infarction mainly shows ST-T abnormality and Q wave abnormality, but because ST-T wave abnormality can also be caused by other pathological changes, normal variation and other factors, and the ST-T change caused by different factors has no specificity, the recognition difficulty is high.

In this embodiment, in order to eliminate interference of ST-T wave abnormality caused by other lesions, ST-T wave abnormality identification is performed on the heart beat segment in combination with preset ST-T wave abnormality characteristic parameters to obtain ST-T wave abnormality classification results, which is helpful for improving accuracy of classification results of myocardial infarction.

The ST-T wave abnormal characteristic parameter is clinical priori information and is an ST-T abnormal characteristic set of electrocardiosignals with myocardial infarction lesions. ST-T wave abnormality identification is carried out on the heart shooting segments by combining ST-T wave abnormality characteristic parameters, interference caused by abnormal ST-T waves caused by other lesions can be prevented, and accurate ST-T wave abnormality classification results are obtained.

In this embodiment, referring to fig. 4 and fig. 5, fig. 4 is a flowchart illustrating a step S2 of a processing method of an electrocardiograph signal according to an alternative embodiment of the present invention; FIG. 5 is a block diagram of a neural network showing ST-T wave anomaly identification according to an alternative embodiment of the present invention.

According to the illustration of fig. 4, step S2 comprises the steps of:

s21: and sequentially inputting the heart shooting segments into an ST-T wave anomaly identification model, and obtaining an ST-T waveform characteristic vector after convolution, normalization, activation and residual block processing.

The chip fragments can be relatively pure heart beat fragments which are beneficial to abnormality detection and have relatively good quality after quality evaluation.

In this embodiment, the cardiac segment is input to the ST-T wave anomaly identification model to identify ST-T wave anomalies, that is, to extract ST-T wave anomaly characteristics, so as to obtain ST-T wave anomaly feature vectors. The ST-T wave anomaly identification model is a classifier model, and is an ST-T wave anomaly identification model trained in advance through a training set, so that ST-T wave anomaly identification of heart beat fragments can be realized.

In other embodiments, ST-T wave anomaly feature vectors may be obtained by other methods, or the heart segments may be input to other ST-T wave recognition models for ST-T wave anomaly recognition.

Referring to fig. 6, fig. 6 is a flowchart illustrating a method for obtaining ST-T waveform feature vectors according to an embodiment of the present invention.

Step S21 includes the steps of:

step S211: the heart beat segment is input to a convolution layer (C0 in fig. 4) of the ST-T wave anomaly identification classification model, and convolution operation is carried out on the heart beat segment and a convolution kernel function to obtain a first feature matrix.

The convolution layer is a convolution network and is used for extracting the characteristics of the input cardiac film segments and outputting a first characteristic matrix. Wherein the convolution kernel of the convolution layer has a dimension of 32. The convolution layers comprise a first convolution layer, a first normalization layer and a first activation layer.

Step S212: residual analysis (C1, C2, and C3 in fig. 4) is performed on the first feature matrix to obtain a second feature matrix.

The residual block is a depth residual network layer, the first feature matrix output by the convolution network is input into the depth residual network layer, and further feature extraction is carried out to obtain a second feature matrix, so that the problem of gradient disappearance or gradient explosion is avoided.

The dimension of the convolution kernel of the depth residual error network layer C1 is 32; the depth residual network layer C2 comprises a second convolution layer, a second normalization layer, a second activation layer, a third convolution layer, a third normalization layer and a third activation layer, wherein the dimension of a convolution kernel in the C2 layer is 64; the convolution kernel of the depth residual network layer C3 has a dimension of 128.

Step S213: and obtaining a mask segment, and performing product operation on the mask segment and the second feature matrix to obtain a third feature matrix so as to obtain the third feature matrix only with ST-T feature information.

In this embodiment, the mask segment is a binary signal composed of 0 and 1, wherein the region with the value of 1 is processed and the region with the value of 0 is not processed. The value of the ST-T area on the mask segment is 1, the values of other areas are 0, after the mask segment is multiplied by the second feature matrix point by point, the value of the ST-T area on the second feature matrix is reserved, the values of other areas are set to 0, in the subsequent analysis, only the ST-T area on the second feature matrix is further analyzed, and the other areas are not analyzed, namely, the features of other wave bands of the heart beat segment are filtered, so that the ST-T waveform feature vector can be finally obtained.

Referring to fig. 7, fig. 7 is a flowchart of a method for obtaining mask fragments according to an embodiment of the invention.

Optionally, the step of obtaining the mask segment includes:

S2131: and analyzing the heart shooting segment to obtain a heart shooting position, and carrying out zero padding on the heart shooting segment based on the heart shooting position so as to extend the heart shooting segment to a preset length to obtain a reference signal.

S2132: and generating an initial mask segment, and resampling the initial mask segment to obtain a mask segment with the length consistent with the length of the reference signal, thereby changing the length of the mask segment. Wherein the mask region from the QRS wave end point to the T wave end point of the mask segment has a value of 1, and the remaining regions have a value of 0.

Step S214: and inputting the third feature matrix into a global pooling layer to be pooled to obtain the ST-T waveform feature vector.

The pooling layer is used for compressing the output third feature matrix so as to obtain ST-T waveform feature vectors, so that the subsequent ST-T waveform abnormal classification is facilitated.

S22: and inputting the ST-T waveform characteristic vector and preset ST-T wave abnormal characteristic parameters into a full-connection layer output three-dimensional vector of the ST-T wave abnormal identification model.

In this embodiment, the ST-T waveform feature vector is a depth feature vector with a dimension of 128, and the ST-T wave anomaly feature parameter includes a plurality of clinical prior ST-T wave anomaly features, and the plurality of clinical prior ST-T wave anomaly features form a feature parameter vector with a dimension of 7; and (3) inputting the depth feature vector with the dimension of 128 and the feature parameter vector with the dimension of 7 into a full connection layer of the ST-T wave anomaly identification model, and outputting a three-dimensional vector with the dimension of 135 after full connection.

The three-dimensional vector is used for representing the probability of the three ST-T wave abnormal types corresponding to the ST-T wave, and the numerical value of each vector can be calculated by obtaining the three-dimensional vector.

In an alternative embodiment, please refer to fig. 8, fig. 8 is a flowchart illustrating a method for acquiring ST-T wave abnormal characteristic parameters according to an embodiment of the present invention.

The method for acquiring the ST-T wave abnormal characteristic parameters comprises the following steps:

s221: and acquiring an electrocardiosignal with abnormal ST-T waves.

The electrocardiosignal with abnormal ST-T wave can be obtained from priori clinical experience, is an electrocardiosignal with myocardial infarction lesion and abnormal ST-T wave, and extracts the abnormal ST-T wave characteristic from the electrocardiosignal with abnormal ST-T wave, so as to extract the abnormal ST-T wave characteristic from the heart beat segments of the detection set.

S222: and identifying the electrocardiosignal to obtain a P wave band, a QRS wave group and an S-T wave band, determining the midpoint between the end point of the P wave band and the start point of the QRS wave group as a reference point, and calculating a voltage value st _ref of the reference point.

S223: and selecting a plurality of sampling points after the QRS wave end point, and calculating the voltage values of the sampling points relative to the reference points so as to construct an ST-T wave abnormal characteristic vector representing the ST-T wave abnormal characteristic.

Specifically, the plurality of sampling points include 20ms, 40ms, 60ms and 80ms after the QRS wave end point, and voltage values of the sampling points relative to the reference point are calculated, and are respectively st ₂₀、st₄₀、st₆₀ and st ₈₀.

The operational formulas of st ₂₀、st₄₀、st₆₀ and st ₈₀ are:

st₂₀＝stj₂₀-st_ref

st₄₀＝stj₄₀-st_ref

st₆₀＝stj₆₀-st_ref；

st₈₀＝stj₈₀-st_ref

Wherein stj ₂₀ is a voltage value obtained by subtracting a voltage value st _ref of the reference point from a voltage value 20ms after the QRS wave end point, and the voltage value st ₂₀ of 20ms after the QRS wave end point is opposite to the voltage value st ₂₀ of the reference point;

stj ₄₀ is a voltage value obtained by subtracting a voltage value st _ref of the reference point from a voltage value 40ms after the QRS wave end point, and the voltage value st ₄₀ of the reference point is 40ms after the QRS wave end point;

stj ₆₀ is a voltage value obtained by subtracting the voltage value st _ref of the reference point from a voltage value 60ms after the QRS wave end point, and the voltage value st ₆₀ of the reference point is 60ms after the QRS wave end point;

stj ₈₀ is a voltage value obtained by subtracting the voltage value st _ref of the reference point from the voltage value 80ms after the QRS wave end point, and the voltage value st ₈₀ of 80ms after the QRS wave end point relative to the reference point is obtained.

S224: the T wave band is divided from the electrocardiosignal, and the positive amplitude T ₊, the negative amplitude T _- and the standard deviation T _std of the T wave band are determined.

The method for determining the forward amplitude T ₊ of the T wave band comprises the following steps: calculating the maximum value of the T wave band, if the maximum value is larger than a first preset threshold value, the forward wave amplitude of the T wave band is the maximum value, otherwise, the forward wave amplitude of the T wave band is 0; the first preset threshold may be 0.15mV.

The method for determining the negative amplitude T _- of the T wave band comprises the following steps: calculating the minimum value of the T wave band, if the minimum value is smaller than a second preset threshold value, the negative wave amplitude of the T wave band is the minimum value, otherwise, the negative wave amplitude of the T wave band is 0; the second preset threshold may be-0.15 mV.

When the T wave is not abnormal, the forward amplitude of the T wave is larger than a first preset threshold value; the negative amplitude of the T wave should be less than a second preset threshold. T wave characteristics with anomalies can be obtained according to the positive amplitude and the negative amplitude of the T wave band, so that ST-T wave anomaly characteristic vectors representing ST-T wave anomaly characteristics can be constructed conveniently. Wherein the T wave anomaly comprises: bi-directional T-wave, or T-wave inversion.

S225: according to the voltage values of a plurality of sampling points, the positive amplitude of the T wave band, the negative amplitude of the T wave band and the standard deviation of the T wave band, an ST-T wave abnormal characteristic vector representing ST-T wave abnormal characteristics is constructed, and the ST-T wave abnormal characteristic vector is determined to be an ST-T wave abnormal characteristic parameter.

The ST-T wave abnormal characteristic vector is as follows: FEa _param＝[st₂₀,st₄₀,st₆₀,st₈₀,t₊,t_-,t_std ].

S23: and determining a vector with the highest numerical value in the three-dimensional vectors, and determining an ST-T wave abnormality classification result, namely an ST-T abnormality type with the highest output probability, according to the category of the vector with the highest numerical value so as to classify and output the probability of myocardial infarction.

S3: and based on preset Q wave abnormal characteristic parameters, carrying out Q wave abnormal identification on the heart shooting segments to obtain Q wave abnormal classification results.

In this embodiment, by combining preset ST-T wave anomaly characteristic parameters, ST-T wave anomaly identification is performed on the cardiac segments to obtain ST-T wave anomaly classification results, so that relatively accurate Q wave anomaly classification results can be obtained, and accuracy of output probability data of myocardial infarction is improved. The Q wave abnormal characteristic parameter is a Q wave abnormal characteristic set of an electrocardiosignal with myocardial infarction lesion.

Specifically, preset Q-wave abnormal characteristic parameters and cardiac segments can be input into a Q-wave abnormal identification model to identify Q-wave abnormal characteristics.

Referring to fig. 9, fig. 9 is a flowchart illustrating a step S3 of a method for processing an electrocardiograph signal according to an embodiment of the invention.

The step S3 comprises the following steps:

S31: and analyzing the heart shooting segment to obtain each lead data, and calculating the Q wave amplitude, the Q wave time limit and the R wave amplitude of each lead data. The lead data includes 12 lead data, and an international universal lead system is adopted, namely an I lead, a II lead, a III lead, an aVL lead, an aVF lead, an aVR lead, a V1 lead, a V2 lead, a V3 lead, a V4 lead, a V5 lead and a V6 lead.

S32: combining preset Q wave abnormality characteristic parameters, determining whether the Q wave is abnormal or not according to the Q wave amplitude value, the Q wave time limit and the R wave amplitude value of each lead data, and outputting Q wave abnormality probability data; the preset Q wave abnormal characteristic parameters comprise a plurality of clinical priori Q wave abnormal characteristics.

In this embodiment, when determining whether the Q wave is abnormal according to the Q wave amplitude, the Q wave time limit, and the R wave amplitude of each lead data in combination with the preset Q wave abnormal characteristic parameter, the Q wave abnormality is determined.

On 12-lead electrocardiographic data, if there is one of the following abnormal conditions, it is judged that the Q wave is abnormal, and the abnormal conditions include:

Judging that the Q wave is abnormal when the ratio of the Q wave amplitude value to the R wave amplitude value is more than or equal to one fourth and/or when the Q wave time limit is more than or equal to 0.04 seconds on the I lead, the II lead, the III lead, the aVL lead, the V3 lead, the V4 lead, the V5 lead and the V6 lead;

On the V1 lead, the V2 lead, the V3 lead, the V4 lead, the V5 lead and the V6 lead, if the Q wave time limit of one lead is more than or equal to the Q wave time limit of the next lead, or the Q wave amplitude of one lead is more than or equal to the Q wave amplitude of the next lead;

On the V1 lead, the V2 lead, the V3 lead and the V4 lead, the R wave amplitude does not have a trend of increasing by leads;

on the V1 lead, the V2 lead, the V3 lead and the V4 lead, the difference of R wave amplitude values of two adjacent leads is more than or equal to 50 percent.

S33: and determining Q wave abnormality classification results according to the Q wave abnormality probability data and outputting the Q wave abnormality classification results.

S4: and determining probability data of the myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result.

After myocardial infarction occurs, the electrocardiogram of the patient is changed with specificity, such as ischemia type T wave, injury type ST segment, necrosis type Q wave and the like, and waveform characteristics of each stage are different, so that probability data of a myocardial infarction stage can be obtained according to waveform abnormality classification results, namely, probability data of the myocardial infarction stage can be obtained according to ST-T wave abnormality classification results and Q wave abnormality classification results.

Example two

According to a second aspect of the embodiment of the present application, an electrocardiograph signal processing device is disclosed, which may be used to execute the content of the central electrical signal processing method according to the embodiment of the present application, and has corresponding functions and beneficial effects. For details not disclosed in the embodiment of the electrocardiosignal processing device of the application, please refer to the content of the electrocardiosignal processing method of the application.

Referring to fig. 10, fig. 10 is a schematic structural diagram of an electrocardiograph signal processing device according to an embodiment of the present invention.

The electrocardiograph signal processing device 800 is characterized by comprising:

the acquisition module 801 is configured to acquire an original electrocardiograph signal, perform noise reduction processing on the original electrocardiograph signal to obtain a sample signal, and perform heartbeat detection on the sample signal to obtain a heartbeat segment;

the first classification module 802 is configured to perform ST-T wave anomaly identification on the cardiac segment based on a preset ST-T wave anomaly characteristic parameter to obtain an ST-T wave anomaly classification result;

The second classification module 803 is configured to perform Q-wave anomaly identification on the cardiac shooting segment based on a preset Q-wave anomaly characteristic parameter to obtain a Q-wave anomaly classification result;

The determining module 804 is configured to determine probability data of the myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result.

In the embodiment of the application, the original electrocardiosignal is subjected to noise reduction treatment, the heart beat detection is carried out to obtain heart beat fragments, ST-T wave abnormality identification is carried out on the heart beat fragments based on preset ST-T wave abnormality characteristic parameters to obtain ST-T wave abnormality classification results, Q wave abnormality identification is carried out on the heart beat fragments based on preset Q wave abnormality characteristic parameters to obtain Q wave abnormality classification results, and finally probability data of a myocardial infarction stage is determined by combining the ST-T wave abnormality classification results and the Q wave abnormality classification results.

In an alternative embodiment, first classification module 802 is configured to obtain ST-T wave anomaly classification results.

The first classification module 802 includes:

And the ST-T waveform characteristic vector unit is used for sequentially inputting the heart shooting segments into the ST-T waveform abnormality recognition model and obtaining the ST-T waveform characteristic vector after convolution, normalization, activation and residual block processing.

The identification output unit is used for inputting the ST-T waveform characteristic vector and preset ST-T wave abnormal characteristic parameters to the full-connection layer output three-dimensional vector of the ST-T wave abnormal identification model; the three-dimensional vector is used for representing the probability of the ST-T wave corresponding to the three ST-T wave anomaly types; the preset ST-T wave abnormal characteristic parameters comprise a plurality of ST-T wave abnormal characteristics based on clinical prior information.

And the classification output unit is used for determining the vector with the highest numerical value in the three-dimensional vectors, and determining the ST-T wave abnormal classification result according to the category of the vector with the highest numerical value.

In an alternative embodiment, the second classification module 803 is configured to obtain a Q-wave anomaly classification result.

The second classification module 803 includes:

the computing unit is used for analyzing the heart shooting segment to obtain each lead data and computing the Q wave amplitude, the Q wave time limit and the R wave amplitude of each lead data.

The probability output unit is used for combining preset Q wave abnormal characteristic parameters, determining whether the Q wave is abnormal or not according to the Q wave amplitude value, the Q wave time limit and the R wave amplitude value of each lead data, and outputting Q wave abnormal probability data; the preset Q wave abnormal characteristic parameters comprise a plurality of clinical priori Q wave abnormal characteristics;

and the classification result output unit is used for determining and outputting Q wave abnormal classification results according to the Q wave abnormal probability data.

It should be noted that, in the electrocardiograph signal processing apparatus provided in the foregoing embodiment, when the electrocardiograph signal processing method is executed, only the division of the foregoing functional modules is used as an example, and in practical application, the foregoing functional allocation may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the electrocardiosignal processing device and the electrocardiosignal processing method provided in the above embodiments belong to the same conception, and the detailed implementation process is shown in the embodiments, and will not be repeated here.

Example III

According to a third aspect of the embodiment of the present invention, an electronic device is disclosed, please refer to fig. 11, fig. 11 is a schematic structural diagram of an electronic device according to an alternative embodiment of the present invention.

The electronic device 900 includes: at least one processor 901 and at least one memory 902;

The memory 902 is configured to store one or more computer programs, where the computer programs are adapted to be loaded by a processor and execute the method for processing an electrocardiograph signal according to any one of the above embodiments.

The electronic device 900 also includes at least one network interface, user interface, memory, and at least one communication bus. Wherein the communication bus is used to enable connection communication between these components.

The user interface may include an interface for connecting to a display screen, and an interface for connecting to a camera, and the optional user interface may also include a standard wired interface, a wireless interface.

The network interface may optionally include a standard wired interface, a wireless interface (e.g., WIFI interface).

Processor 901 may include one or more processing cores, among other things. The processor 901 connects various portions of the overall electronic device 900 using various interfaces and lines, and performs various functions of the electronic device 900 and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the processor 901, and invoking data stored in the memory 902. Alternatively, the processor 901 may be implemented in at least one hardware form of digital signal Processing (DIGITAL SIGNAL Processing, DSP), field-Programmable gate array (Field-Programmable GATE ARRAY, FPGA), programmable logic array (Programmable logic arrays, PLA). The processor 901 may integrate one or a combination of several of a central processing unit (Central Processing Unit, CPU), an image processor (Graphics Processing Unit, GPU), and a modem, etc. The CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 901 and may be implemented by a single chip.

The Memory 902 may include a random access Memory (Random Access Memory, RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 902 includes a non-transitory computer readable medium (non-transitory computer-readable storage medium). Memory 902 may be used to store instructions, programs, code, sets of codes, or instruction sets. The memory 902 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-described various method embodiments, etc.; the storage data area may store data or the like referred to in the above respective method embodiments. The memory 902 may also optionally be at least one storage device located remotely from the processor 901. As shown in fig. 9, an operating system, a network communication module, a user interface module, and an operating application of the smart device may be included in the memory 902, which is one type of computer storage medium.

In the electronic device 900 shown in fig. 9, the user interface is mainly used for providing an input interface for a user, acquiring data input by the user, and providing a video input interface for a camera, and acquiring an image signal; and the processor 901 may be used to call an operating application of the smart device stored in the memory 902 and perform the related operations in the image quality adjustment method in the above-described embodiment.

The intelligent device can be used for executing the content of the electrocardiosignal processing method according to the corresponding embodiment of the application and has corresponding functions and beneficial effects.

According to a fourth aspect of the embodiments of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the relevant operations in the method for processing an electrocardiographic signal according to any one of the above embodiments, and has corresponding functions and advantageous effects. Where computer readable media includes both permanent and non-permanent, removable and non-removable media, information storage may be implemented by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.

Note that the above is only a preferred embodiment of the present application and the technical principle applied. It will be understood by those skilled in the art that the present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the application. Therefore, while the application has been described in connection with the above embodiments, the application is not limited to the embodiments, but may be embodied in many other equivalent forms without departing from the spirit or scope of the application, which is set forth in the following claims.

Claims

1. The electrocardiosignal processing method is characterized by comprising the following steps of:

Based on preset ST-T wave abnormal characteristic parameters, carrying out ST-T wave abnormal identification on the heart shooting segments to obtain ST-T wave abnormal classification results, wherein the ST-T wave abnormal classification results comprise: sequentially inputting the heart shooting segments into an ST-T wave anomaly identification model, and obtaining an ST-T wave characteristic vector after convolution, normalization, activation and residual block processing;

Inputting the heart beat segment to a convolution layer of an ST-T wave anomaly identification classification model, and carrying out convolution operation on the heart beat segment and a convolution kernel function to obtain a first feature matrix;

residual analysis is carried out on the first feature matrix to obtain a second feature matrix;

Obtaining a mask segment, and performing product operation on the mask segment and the second feature matrix to obtain a third feature matrix;

Inputting the third feature matrix into a global pooling layer to be pooled to obtain ST-T waveform feature vectors;

inputting the ST-T waveform characteristic vector and preset ST-T wave abnormality characteristic parameters into a full-connection layer of an ST-T wave abnormality identification model to output a three-dimensional vector, wherein the three-dimensional vector is used for representing the probability of three ST-T wave abnormality types corresponding to the ST-T wave; the preset ST-T wave abnormal characteristic parameters comprise a plurality of ST-T wave abnormal characteristics based on clinical prior information;

Determining a vector with the highest numerical value in the three-dimensional vectors, and determining an ST-T wave abnormal classification result according to the category of the vector with the highest numerical value;

determining probability data of a myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result;

wherein, the obtaining the mask segment includes:

Analyzing the heart shooting section to obtain a heart shooting position, and carrying out zero padding on the heart shooting section based on the heart shooting position so as to extend the heart shooting section to a preset length to obtain a reference signal;

Generating an initial mask segment, and resampling the initial mask segment to obtain a mask segment with the length consistent with the length of the reference signal, wherein the value of a mask region from a QRS wave end point to a T wave end point of the mask segment is1, and the values of other regions are 0;

Acquiring an electrocardiosignal with abnormal ST-T waves;

Identifying the electrocardiosignal to obtain a P wave band, a QRS wave group and an S-T wave band, determining the midpoint between the end point of the P wave band and the start point of the QRS wave group as a reference point, and calculating the voltage value of the reference point;

Selecting a plurality of sampling points after the QRS wave end point, and calculating the voltage values of the sampling points relative to the reference point;

Dividing a T wave band in the electrocardiosignal, and determining the positive amplitude, the negative amplitude and the standard deviation of the T wave band;

Constructing an ST-T wave abnormal feature vector representing ST-T wave abnormal features according to the voltage values of a plurality of sampling points, the positive amplitude of a T wave band, the negative amplitude of the T wave band and the standard deviation of the T wave band, and determining the ST-T wave abnormal feature vector as ST-T wave abnormal feature parameters

The method for determining the forward amplitude of the T wave band comprises the following steps: calculating the maximum value of a T wave band, wherein if the maximum value is larger than a first preset threshold value, the forward amplitude of the T wave band is the maximum value, otherwise, the forward amplitude of the T wave band is 0;

the method for determining the negative amplitude of the T wave band comprises the following steps: and calculating the minimum value of the T wave band, wherein if the minimum value is smaller than a second preset threshold value, the negative amplitude of the T wave band is the minimum value, otherwise, the negative amplitude of the T wave band is 0.

2. The method for processing an electrocardiograph signal according to claim 1, wherein the step of performing Q-wave anomaly identification on the cardiac segment based on a preset Q-wave anomaly characteristic parameter to obtain a Q-wave anomaly classification result includes:

Analyzing the heart shooting segment to obtain each lead data, and calculating the Q wave amplitude, the Q wave time limit and the R wave amplitude of each lead data;

Combining preset Q wave abnormality characteristic parameters, determining whether the Q wave is abnormal or not according to the Q wave amplitude, the Q wave time limit and the R wave amplitude of each lead data, and outputting Q wave abnormality probability data; the preset Q wave abnormal characteristic parameters comprise a plurality of clinical priori Q wave abnormal characteristics;

and determining the Q wave abnormal classification result according to the Q wave abnormal probability data and outputting the Q wave abnormal classification result.

3. The method for processing electrocardiographic signals according to claim 2, wherein when determining whether there is an abnormality in the Q wave based on the Q wave amplitude, Q wave time limit, and R wave amplitude of each lead data in combination with a preset Q wave abnormality characteristic parameter,

On 12-lead electrocardiographic data, if there is one of the following abnormal conditions, it is determined that the Q wave is abnormal, the abnormal conditions including:

Judging that the Q wave is abnormal if the ratio of the Q wave amplitude value to the R wave amplitude value is more than or equal to one fourth and/or the Q wave time limit is more than or equal to 0.04 seconds on the I lead, the II lead, the III lead, the aVL lead, the V3 lead, the V4 lead, the V5 lead and the V6 lead;

4. An electrocardiograph signal processing device, comprising:

the first classification module is configured to perform ST-T wave anomaly identification on the cardiac segment based on a preset ST-T wave anomaly characteristic parameter to obtain an ST-T wave anomaly classification result, and includes: sequentially inputting the heart shooting segments into an ST-T wave anomaly identification model, and obtaining an ST-T wave characteristic vector after convolution, normalization, activation and residual block processing;

the determining module is used for determining probability data of the myocardial infarction stage according to the ST-T wave abnormal classification result and the Q wave abnormal classification result;

wherein, the obtaining the mask segment includes:

Acquiring an electrocardiosignal with abnormal ST-T waves;

5. An electronic device, the electronic device comprising: at least one processor and at least one memory;

Wherein the memory is adapted to store one or more computer programs adapted to be loaded by the processor and to perform the method of processing an electrocardiographic signal according to any one of claims 1 to 3.

6. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the method of processing an electrocardiographic signal according to any one of claims 1 to 3.