Heart global electrical heterogeneity parameter combination for identifying OCAD and CMD and application

Abstract

The invention provides a heart global electrical heterogeneity parameter combination for identifying OCAD and CMD and application, particularly three heart global electrical heterogeneity parameters including an SVG area elevation angle (Area SVG Event), an area of VM under a QT interval curve (AUC of Vector Magnitude) and a QT interval (QT interval), information of multiple dimensions of a QRS wave group, a T wave, the SVG and the like is covered, rich features are provided for evaluating heterogeneity of heart electrical activity, and the heart electrical heterogeneity is evaluated. The superiority of the three parameters in distinguishing CMD and OCAD is proved, and the accurate variable identification of the OCAD and CMD is realized by means of the three parameter combinations.

Description

Technical Field

[0001] The present application relates to the field of electrocardiogram (ECG) signal processing, and in particular to a combination and application of cardiac global electrical inhomogeneity parameters for identifying OCAD and CMD. Background Art

[0002] Coronary artery disease (CAD) is a global health challenge. Symptomatic patients with evidence of myocardial ischemia during routine stress tests are further diagnosed with obstructive coronary artery disease (OCAD) by invasive coronary angiography (CAG), defined as severe stenosis of epicardial coronary arteries (diameter severity ≥50% on CAG) or abnormal fractional flow reserve (FFR<0.8). However, up to 70% of patients with CAG have non-obstructive coronary artery ischemia (INOCA). In symptomatic INOCA cases, the main cause of myocardial ischemia is coronary microvascular dysfunction (CMD), which is caused by abnormalities in the structure and / or function of coronary microvessels. CMD prevents coronary microvessels from generating maximal coronary blood flow (CBF) to meet the increased metabolic demands of the myocardium, leading to myocardial ischemia and increasing the risk of poor prognosis, rehospitalization, adverse cardiovascular events, and death.

[0003] Obstructive coronary artery disease (OCAD) and coronary microvascular dysfunction (CMD) are different manifestations of coronary artery disease (CAD). Patients may show similar angina symptoms because both will eventually lead to myocardial ischemia. However, there are significant differences between CMD and OCAD in terms of pathogenesis, treatment strategies, and long-term prognosis. In clinical practice, it is crucial to use non-invasive methods to distinguish CMD from OCAD early.

[0004] Existing technologies have explored some biomarkers to distinguish CMD from OCAD, such as low-density lipoprotein-cholesterol, heart rate, and the expression of different molecular biomarkers, which are significantly different between CMD and OCAD. In addition, in the pre-hospital and emergency room, the vast majority of patients with angina pectoris receive electrocardiogram (ECG) as the first-line treatment. Changes in the ECG can reflect changes in cardiac depolarization and repolarization caused by OCAD or CMD. ST segment changes, heart rate variability (HRV) parameters, and ST segment drop time to 1mm are significantly different between CMD and OCAD, suggesting that CMD and OCAD can be distinguished by ECG characteristics. However, CMD and OCAD are highly similar in clinical manifestations, laboratory biomarkers, and general ECG characteristics, which makes early and accurate detection of CMD difficult.

[0005] In other words, the current methods targeting ECG characteristics mainly focus on studying myocardial ischemia caused by obstructive coronary artery disease, or distinguishing between CMD and healthy people, but there is no method that uses ECG characteristics to distinguish between obstructive coronary artery disease OCAD and CMD. The non-invasive identification of ECG characteristics of OCAD and CMD is still a major technical gap in this field. Summary of the invention

[0006] The embodiment of the present application provides a combination and application of cardiac global electrical heterogeneity parameters for identifying OCAD and CMD. OCAD and CMD are effectively distinguished by using global electrical heterogeneity parameters extracted from the electrocardiogram vector graph (VCG), and an early non-invasive detection algorithm for CMD based on the characteristics of cardiac global electrical heterogeneity parameters is established.

[0007] In a first aspect, an embodiment of the present application provides a combination of cardiac global electrical heterogeneity parameters for identifying OCAD and CMD, including: SVG area elevation angle, VM area under the QT interval curve, and QT interval of cardiac global electrical heterogeneity parameters extracted based on 12-lead ECG signals.

[0008] In a second aspect, an embodiment of the present application provides a screening method for identifying a combination of cardiac global electrical inhomogeneity parameters for OCAD and CMD, comprising: collecting 12-lead ECG signals from different OCAD patients and different CMD patients, and converting the 12-lead ECG signals into 3-lead VCG signals; selecting a data segment beat from the VCG signal of each lead, selecting a single-lead median beat from the data segment beat using a median algorithm, and constructing a time-coherent global XYZ median beat based on the single-lead median beat of each lead, detecting the VCG origin on the global XYZ median beat and calculating the vector amplitude, marking key points based on the vector amplitude, obtaining QRS vectors, T vectors, and SVG vectors based on the key points, and calculating Obtain the cardiac global electrical heterogeneity parameters; take each cardiac global electrical heterogeneity parameter as the independent variable and CMD or OCAD as the dependent variable for binary univariate logistic regression analysis, and screen the cardiac global electrical heterogeneity parameters with a P value less than the first threshold as the first undetermined parameter; take the combination of the first undetermined parameters as the independent variable and CMD or OCAD as the dependent variable for multicollinearity analysis, retain the combination in which the variance inflation factor of each first undetermined parameter in the combination is less than the second threshold and take the first undetermined parameter in the combination as the second undetermined parameter; take the combination of the second undetermined parameters as the independent variable and CMD or OCAD as the dependent variable for binary multivariate logistic regression analysis, and screen the second undetermined parameters with a P value less than the third threshold as the parameter combination.

[0009] In the third aspect, an embodiment of the present application provides an identification model for identifying OCAD and CMD based on global cardiac electrical heterogeneity parameters, which is a regression model that records the regression relationship between SVG area elevation angle, VM area under the QT interval curve and QT interval of global cardiac electrical heterogeneity parameters extracted based on 12-lead ECG signals and OCAD or CMD, or a model trained by SVG area elevation angle, VM area under the QT interval curve and QT interval of global cardiac electrical heterogeneity parameters extracted based on 12-lead ECG signals that marks OCAD or CMD.

[0010] In a fourth aspect, an embodiment of the present application provides an application method for identifying a combination of cardiac global electrical inhomogeneity parameters for OCAD and CMD, the method comprising:

[0011] Collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal;

[0012] A data segment beat is selected from the VCG signal of each lead, a single lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval are calculated;

[0013] The SVG area elevation, the area under the QT curve of VM, and the QT interval were input into a recognition model to identify the subject as having OCAD or CMD.

[0014] In a fifth aspect, an embodiment of the present application provides an identification device for identifying OCAD and CMD based on global cardiac electrical heterogeneity parameters, comprising: a 12-lead electrocardiogram device, an electronic data processing device and a display component, wherein the electronic data processing device performs the following steps: collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal; selecting a data segment beat from the VCG signal of each lead, selecting a single-lead median beat from the data segment beat using a median algorithm, and constructing a time-coherent global XYZ median beat based on the single-lead median beat of each lead, detecting the VCG origin on the global XYZ median beat and calculating the vector amplitude, marking key points based on the vector amplitude, obtaining QRS vectors, T vectors and SVG vectors based on the key points and calculating the global cardiac electrical heterogeneity parameters; inputting the global cardiac electrical heterogeneity parameters into a recognition model to identify whether the subject suffers from OCAD or CMD; and the display component is configured to display the recognition result of the electronic data processing device.

[0015] In a sixth aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any application method of identifying any combination of cardiac global electrical heterogeneity parameters of OCAD and CMD.

[0016] The main contributions and innovations of the present invention are as follows:

[0017] The present invention provides an application of a combination of cardiac global electrical inhomogeneity parameters for identifying OCAD and CMD. Compared with traditional methods such as TTDE, MCE, PET and CMR, this method has lower cost, is non-invasive and does not bring the risk of ionizing radiation. The 12-lead ECG signal is converted into a 3-lead VCG signal, and the median filter and wavelet filter are used to remove noise, thereby improving the signal quality. At the same time, the influence of noise on signal analysis is reduced by selecting the data segment beat and calculating the maximum absolute value point of the first-order derivative, and the median algorithm is used to select the single-lead median beat, thereby improving the accuracy and representativeness of feature extraction.

[0018] The present invention particularly includes three cardiac global electrical heterogeneity parameters, namely, Area SVG Elevation, Area under the QT interval curve (AUC of Vector Magnitude) and QT interval, which cover information of multiple dimensions such as QRS complex, T wave and SVG, and provide rich features for evaluating the heterogeneity of cardiac electrical activity, proving their superiority in distinguishing CMD from OCAD. These parameters are defined in detail through comprehensive calculation formulas, ensuring the scientificity and effectiveness of feature extraction.

[0019] Finally, the present invention improves the prediction performance of the model by training the recognition model to record the regression relationship between the cardiac global electrical heterogeneity parameters and CMD and OCAD. This method can realize low-cost and non-invasive CMD and OCAD recognition under the setting of conventional clinical electrocardiographs, has high clinical application value, and brings new possibilities for the diagnosis of heart diseases.

[0020] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 Figure 1 is a schematic diagram of coronary angiography in patients with OCAD and CMD. Figure 1 (A) in the figure corresponds to the schematic diagram of coronary angiography of a CMD patient, with 40% stenosis of the left anterior descending coronary artery, FFR = 0.86, and microcirculatory resistance index (IMR) = 48.4; Figure 1 (B) in the figure corresponds to a schematic diagram of coronary angiography in an OCAD patient, with a 60% stenosis of the left circumflex distal coronary artery and FFR=0.53.

[0023] Figure 2 It is a schematic diagram of the process of obtaining the cardiac global electrical inhomogeneity parameters of the present invention;

[0024] Figure 3 is the temporally coherent global XYZ median heart beat for CMD patients and OCAD patients;

[0025] Figure 4 Schematic diagram of global cardiac electrical heterogeneity parameters in CMD and OCAD patients.

[0026] Figure 5 is a schematic diagram of the area under the QT curve of VM in CMD patients and OCAD patients.

[0027] Figure 6 It is a flowchart for statistical analysis of cardiac global electrical heterogeneity parameters. Figure 7 is a schematic diagram of the ROC AUC values ​​for identifying CMD and OCAD.

[0028] Figure 8 It is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of this specification. Instead, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0030] It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0031] Embodiment 1

[0032] An embodiment of the present application provides a combination of cardiac global electrical heterogeneity parameters for identifying OCAD and CMD, including: SVG area elevation angle, VM area under the QT interval curve, and QT interval of cardiac global electrical heterogeneity parameters extracted based on 12-lead ECG signals.

[0033] Based on Wilson's ventricular gradient vector theory, the global electrical heterogeneity parameter (GEH) can be extracted from the electrocardiogram vector graph (VCG). Compared with traditional electrocardiogram indicators, GEH provides spatiotemporal characteristics of repolarization and depolarization, providing additional predictive value for a variety of heart and cardiovascular diseases, such as ischemia / infarction, post-myocardial infarction patients, atherosclerosis, sudden cardiac death (SCD), and subclinical myocardial injury. This scheme, based on the verification of the effectiveness of global electrical heterogeneity parameters in distinguishing CMD from OCAD, screened out a combination of cardiac global electrical heterogeneity parameters that can effectively identify OCAD and CMD.

[0034] It should be noted that although transthoracic Doppler echocardiography (TTDE), myocardial contrast echocardiography (MCE), positron emission tomography (PET), and cardiac magnetic resonance imaging (CMR) have been used to evaluate coronary microcirculation, these identification methods have problems with reduced spatial resolution, high cost, lack of availability, or ionizing radiation, which limits their application in clinical settings. In contrast, global electrical heterogeneity parameters are easier to obtain and less expensive, and do not pose ionizing radiation risks. Global electrical heterogeneity parameters can capture the unique patterns of depolarization and repolarization heterogeneity of CMD and OCAD, and their high repeatability of measurement supports their implementation in clinical practice and research. This protocol can achieve low-cost, noninvasive identification of CMD and OCAD under the setting of a conventional clinical electrocardiograph.

[0035] The calculation methods for the three parameters of global cardiac electrical heterogeneity, namely, Area SVG Elevation, Area of ​​VM under the QT interval curve AUC of Vector Magnitude, and QT interval QT Interval, are as follows:

[0036] Collect 12-lead ECG signals and convert them into 3-lead VCG signals. Select data segment beats from the VCG signals of each lead. Use the median algorithm to select the single-lead median beat from the data segment beats. Construct a time-coherent global XYZ median beat based on the single-lead median beat of each lead. Detect the VCG origin on the global XYZ median beat and calculate the vector amplitude. Mark key points based on the vector amplitude. Obtain QRS vector, T vector and SVG vector based on the key points and calculate the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval.

[0037] In this solution, a conventional ECG acquisition device can be used to acquire the subject's 12-lead ECG signal. In order to facilitate the subsequent processing of the ECG signal, in some embodiments, this solution uses median filtering to filter out baseline drift and uses wavelet filtering to filter out myoelectric noise.

[0038] In some embodiments, the present approach only requires collecting 10 seconds of 12-lead ECG signals from the subject.

[0039] In the step of “converting the 12-lead ECG signal into a 3-lead VCG signal”, the 12-lead ECG signal is converted into a 3-lead VCG signal using the following formula:

[0040]

[0041] Among them, I, II, V1, V2, V3, V4, V5, V6 represent the leads of ECG signals respectively; Vx, Vy, Vz represent the VCG of X lead, VCG of Y lead and VCG of Z lead respectively. X lead: left and right direction on the horizontal plane (cross section); Y lead: front and back direction on the sagittal plane; Z lead: up and down direction on the vertical plane. These three leads correspond to the three dimensions of cardiac electrical activity in space.

[0042] The calculation method of the global electrical heterogeneity parameters of the heart is as follows Figure 2As described, in the step of "selecting data segment beats from the VCG signal of each lead", the maximum absolute value point of the first-order derivative is determined based on the VCG signal of the Vx lead, the maximum absolute value point of the first-order derivative is located in the VCG signal of each lead, and the data segment beats are selected before and after the maximum absolute value point of the first-order derivative. The reason why the present scheme selects the data segment beats is to extract the most representative and stable feature parts from the complex VCG signal. By selecting the data segment beats near the maximum absolute value point of the first-order derivative, the influence of noise on signal analysis can be effectively reduced. Noise usually causes the first-order derivative of the signal to fluctuate, but near these feature points, the actual electrical activity changes are much greater than the influence of noise, making these points easier to accurately identify, and the maximum absolute value point of the first-order derivative usually marks the turning point of the heart's electrical activity. The accurate positioning of these points helps to improve the accuracy of subsequent analysis.

[0043] Specifically, this solution determines the maximum absolute value point of the first-order derivative based on the VCG signal of the Vx lead, and selects sample points of the front and rear segments based on the maximum absolute value point of the first-order derivative on the VCG signals of different leads to form a data segment beat. In some specific embodiments, 299 sample points are selected before the maximum absolute value point of the first-order derivative on the VCG signal of each lead, and 300 sample points are selected after the maximum absolute value point of the first-order derivative on the VCG signal of each lead to form a data segment beat. Correspondingly, the data segment beat includes 600 sample points including the maximum absolute value point of the first-order derivative.

[0044] The calculation formula for the maximum absolute value point of the first-order derivative of the VCG signal of the Vx lead is:

[0045]

[0046] Where Vx represents the VCG of lead X and t is the lead length.

[0047] In the step of "selecting the median beat of a single lead from the data segment beat using the median algorithm", multiple periodic beats are collected from the sub-data segment beats, the periodic beats are aligned to obtain aligned beats, the median of all aligned beats is calculated for the sample value at each time point, and the median is integrated in sequence to obtain the median beat of a single lead. The advantage of this is that the beats that can best represent these heartbeat cycles are selected from the data segment beats, reducing the influence of noise and abnormal beats.

[0048] In some embodiments, the ECG waveform of each cycle is selected as the periodic beat, multiple periodic beats are aligned to ensure that different periodic beats are consistent in time to obtain aligned beats, the median of all aligned beat values ​​at each time point is calculated, and the median beat of a single lead is obtained by integrating the medians in sequence.

[0049] In the step of "constructing a time-coherent global XYZ median beat based on the single-lead median beats of each lead", the single-lead median beats of each lead are time-aligned, and the single-lead median beats of each lead after time alignment are merged to obtain the global XYZ median beat.

[0050] A physiologically meaningful definition of the cardiac vector origin can improve the prediction accuracy of GEH. Therefore, the VCG origin needs to be detected on the time-coherent global XYZ median beat, that is, the electrostatic or isoelectric state of the heart when the cardiac vector does not move in three-dimensional space. Correspondingly, in the step of "detecting the VCG origin on the global XYZ median beat", the vector amplitude of the global XYZ median beat is calculated, and the point with the smallest vector amplitude is located as the VCG origin. If the vector amplitude value of a point is indeed close to zero or very small, the point can be considered as the VCG origin. In addition, the VM values ​​around the point can be checked to ensure that it is a local minimum.

[0051] In "Mark key points based on vector magnitude", mark the QRS start and end, R peak, and T wave end as key points.

[0052] In the step of "obtaining QRS vector, T vector and SVG vector based on key points and calculating SVG area elevation angle, area of ​​VM under QT interval curve and QT interval",

[0053] The calculation formula of SVG area elevation angle is:

[0054] Area SVG Elevation:

[0055]

[0056] Among them, |V MX (t)| is the VM signal, t is time, T-offset represents the end point of the T wave, and QRS-onset represents the start point of the QRS wave

[0057] The area under the QT interval curve of VM is calculated as:

[0058]

[0059] Where T-offset represents the end point of the T wave, QRS-onset represents the starting point of the QRS wave, |V MX (t)| is the VM signal, t is time, VM=(V MX ,V MY ,V MZ ).

[0060] The QT interval is calculated as:

[0061] QT_int erval=Time(VM(T-offset))-Time(VM(Q-onset));

[0062] T-offset represents the end point of the T wave, QRS-onset represents the start point of the QRS wave, |V MX (t)| is the VM signal.

[0063] Embodiment 2

[0064] The following introduces the selection method of SVG area elevation Area SVG Elevation, area AUC of Vector Magnitude of VM under the QT interval curve, and QT interval QT Interval of this scheme. This scheme provides a screening method for identifying a combination of cardiac global electrical heterogeneity parameters for OCAD and CMD, including the following steps:

[0065] Collect 12-lead ECG signals from different OCAD patients and different CMD patients;

[0066] A data segment beat is selected from the VCG signal of each lead, a single lead median beat is selected from the data segment beat using the median algorithm, and a time-coherent global XYZ median beat is constructed based on the single lead median beat of each lead. The VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated. Key points are marked based on the vector amplitude. QRS vectors, T vectors and SVG vectors are obtained based on the key points and the global electrical inhomogeneity parameters of the heart are calculated, where the global electrical inhomogeneity parameters of the heart are: QRS S-T peak angle, QRS-T area angle, VM area under the QT interval curve, QT interval, QRS peak azimuth, QRS peak elevation, QRS area azimuth, QRS area elevation, QRS amplitude peak, QRS area, T wave peak azimuth, T wave peak elevation, T wave area azimuth, T wave area elevation, T wave amplitude peak, T wave area, SVG peak azimuth, SVG peak elevation, SVG area azimuth, SVG area elevation, SVG amplitude peak, Wilson SVG;

[0067] Each cardiac global electrical heterogeneity parameter was used as an independent variable, and CMD or OCAD was used as the dependent variable for binary univariate logistic regression analysis. The cardiac global electrical heterogeneity parameters with P values ​​less than the first threshold were screened as the first undetermined parameters; the combination of the first undetermined parameters was taken as the independent variable, and CMD or OCAD was used as the dependent variable for multicollinearity analysis. The combination in which the variance inflation factor of each first undetermined parameter in the combination was less than the second threshold was retained, and the first undetermined parameter in the combination was taken as the second undetermined parameter; the combination of the second undetermined parameters was taken as the independent variable, and CMD or OCAD was used as the dependent variable for binary multivariate logistic regression analysis. The second undetermined parameters with P values ​​less than the third threshold were screened as parameter combinations.

[0068] Through the above steps, this scheme can systematically extract cardiac global electrical heterogeneity parameters from ECG signals, and finally determine which parameters are independent predictive factors for distinguishing CMD and OCAD patients through step-by-step screening and analysis. This scheme finally screened out SVG area elevation Area SVG Elevation, VM area under the QT interval curve AUC of VectorMagnitude, and QT interval QT Interval as the final cardiac global electrical heterogeneity parameters.

[0069] In the step of "obtaining QRS vector, T vector and SVG vector based on key points and calculating the parameters of global cardiac electrical heterogeneity", the parameters of global cardiac electrical heterogeneity are: QRS-T peak angle (Peak QRS-T Angle), QRS-T area angle (Area QRS-T Angle), VM area under the QT interval curve (AUC of Vector Magnitude), QT interval (QT Interval), QRS peak azimuth (Peak QRS Azimuth), QRS peak elevation (Peak QRS Elevation), QRS area azimuth (Area QRS Azimuth), QRS area elevation (Area QRS Elevation), QRS amplitude peak (Peak QRS Magnitude), QRS area (Area QRS), T wave peak azimuth (Peak TAzimuth), T wave peak elevation (Peak T Elevation), T wave area azimuth (Area T Azimuth), T wave area elevation (Area T Elevation), T wave amplitude peak (Peak T One or a combination of the following parameters: QRS complex, T wave, SVG peak azimuth, SVG peak elevation, SVG area azimuth, SVG area elevation, SVG peak amplitude, Wilson SVG. These parameters cover the characteristics of cardiac electrical activity in different dimensions, including information on the azimuth, elevation, amplitude and area of ​​the QRS complex, T wave and spatial ventricular gradient (SVG). These parameters are of great significance in electrocardiogram analysis and help evaluate the heterogeneity of cardiac electrical activity.

[0070] The calculation formulas for these 22 cardiac global electrical heterogeneity parameters are as follows:

[0071] Peak QRS-T Angle:

[0072]

[0073] in is the vector dot product of the peak value of the QRS wave and the peak value of the T wave, is the vector modulus length of the peak value of the QRS wave, |T peak | is the vector modulus of the peak of the T wave;

[0074] Area QRS-T Angle:

[0075]

[0076] in is the vector dot product of the area of ​​the QRS wave and the area of ​​the T wave, |QRSmean| is the vector modulus of the area of ​​the QRS wave, and |Tmean| is the vector modulus of the area of ​​the T wave.

[0077] The area under the QT interval curve of VM (AUC of Vector Magnitude):

[0078]

[0079] Where T-offset represents the end point of the T wave, QRS-onset represents the starting point of the QRS wave, |V MX (t)| is the VM signal, t is time, VM=(V MX ,V MY ,V MZ ).

[0080] QT Interval:

[0081] QT_interval=Time(VM(T-offset))-Time(VM(Q-onset));

[0082] T-offset represents the end point of the T wave, QRS-onset represents the start point of the QRS wave, |V MX (t)| is the VM signal.

[0083] The calculation formulas for the direction and magnitude parameters of the measured spatial peak QRS vector and the spatial area QRS vector related to the spatial QRS vector are as follows:

[0084] Peak QRS Azimuth:

[0085]

[0086] Peak QRS Elevation:

[0087]

[0088] Area QRS Azimuth:

[0089]

[0090] Area QRS Elevation:

[0091]

[0092] Peak QRS Magnitude:

[0093]

[0094] Area QRS:

[0095]

[0096] Spatial QRS vectors are defined as:

[0097]

[0098] QRSV x (t R ),QRSV y (t R ),QRSV z (t R ) are defined as Mx ,V My ,V Mz The corresponding spatial QRS vector on the lead, QRS-onset and QRS-offset represent the starting point and end point of the QRS wave.

[0099] The directions and magnitudes of the spatial peak T vectors and spatial area T vectors of the spatial T vectors are calculated as follows:

[0100] Peak T Azimuth:

[0101]

[0102] Peak T Elevation:

[0103]

[0104] Area T Azimuth:

[0105]

[0106] Area T Elevation:

[0107]

[0108] Peak T Magnitude:

[0109]

[0110] T wave area (Area T):

[0111]

[0112] Spatial T vectors are defined as:

[0113]

[0114] Among them TV x (t R ),TV y (t R ),TV z (t R ) are defined as Mx ,V My ,V Mz The corresponding spatial T vector on the lead, T-offset represents the end point of the T wave, and QRS-offset represents the end point of the QRS wave.

[0115] The direction and magnitude of the spatial ventricular gradient peak vector (SVG) and the spatial area SVG vector are calculated according to the following formula:

[0116] Peak SVGAzimuth:

[0117]

[0118] Peak SVG Elevation:

[0119]

[0120] Area SVGAzimuth:

[0121]

[0122] Area SVG Elevation:

[0123]

[0124] Peak SVG Magnitude:

[0125]

[0126] The spatial ventricular gradient (SVG) peak vector is defined as:

[0127]

[0128] SVGV x (t R ),SVGV y (t R ),SVGV z (t R ) are defined as Mx ,V My ,V Mz The corresponding spatial ventricular gradient on the lead, T-

[0129] in

[0130] Offset represents the end point of the T wave, and QRS-onset represents the starting point of the QRS wave.

[0131] Wilson SVG:

[0132]

[0133] Specifically, this scheme first uses SPSS (Version 25.0, IBM Corp) and Graphpad prism (Version 9.0, Graphpad prism Corp) software to screen the parameters of global cardiac electrical heterogeneity, and defines the third threshold value of 0.05 as clinically significant.

[0134] This protocol performed data preprocessing for each cardiac global electrical heterogeneity parameter: categorical data were expressed as numbers and percentages (relative frequencies), parametric quantitative variables were presented as mean ± standard deviation, nonparametric quantitative variables were presented as median [interquartile range (IQR)], and the KS test was used to test data distribution, among which the independent sample t test was used for parametric variables, the Mann-Whitney U test was used for nonparametric variables, and the 2 test was used for categorical variables.

[0135] like Figure 6As shown in the figure, each cardiac global electrical heterogeneity parameter of this scheme was selected as the input of binary univariate logistic regression analysis; then the first undetermined parameter with P<0.10 was selected for multicollinearity analysis. The parameters with variance inflation factor (VIF) less than 5 in the tolerance test were considered as non-collinearity, and the combination with variance inflation factor of each first undetermined parameter less than 5 was retained and the first undetermined parameter in the combination was taken as the second undetermined parameter; finally, binary multivariate logistic regression analysis and forward stepwise selection were used to identify independent predictors from these non-collinearity features, and the odds ratio (OR) and the corresponding 95% confidence interval (CI) were measured, and the discriminant performance of multivariate logistic regression analysis was evaluated by the area under the receiver operator characteristic curve (ROC AUC).

[0136] In order to ensure the comprehensiveness of the statistical analysis samples, in a specific embodiment, the basic information table of the subjects selected in this scheme is shown in Table 1:

[0137] Table 1 Basic information of subjects

[0138]

[0139]

[0140] The 22 global cardiac electrical heterogeneity parameters of CMD patients and OCAD patients are shown in Table 2. It can be seen from Table 2 that the QT interval, QRS peak azimuth, and SVG peak azimuth of CMD patients are significantly higher than those of OCAD patients, while the T wave area elevation and SVG peak elevation of CMD patients are significantly lower than those of OCAD patients.

[0141] Table 2 22 parameters of global cardiac electrical heterogeneity in patients with CMD and OCAD

[0142]

[0143] The correlation analysis between GEH and CMD by univariate and multivariate logistic regression analysis is shown in Table 3. In the univariate logistic regression analysis, six GEH parameters (QT interval, QRS peak azimuth, QRS area azimuth, QRS area azimuth, T wave area elevation, and SVG area elevation) were selected as independent predictors because their p < 0.5. After multivariate adjustment, the area under the QT interval curve (AUC of Vector Magnitude) of VM (OR=0.979, [95% CI 0.964-0.995], p=0.009), Area SVG Elevation (OR=0.978, [95% CI 0.959-0.998], p=0.031) and QT interval (OR=1.014, [95% CI 1.000-1.024], p=0.010) were selected, and the accuracy rate of distinguishing CMD from OCAD reached 65.6%. In addition, the ROC AUC value of multivariate logistic regression features (i.e., AUC of Vector Magnitude, Area SVGElevation, and QT Interval) for distinguishing CMD from OCAD was 0.717 (95% CI 0.636-0.798), and the ROC AUC values ​​of QT Interval and Area SVG Elevation were 0.621 (95% CI 0.532-0.710) and 0.635 (95% CI 0.547-0.724), respectively. Figure 7 As shown, Area SVG Elevation is an independent protective factor, while QT interval is an independent risk factor.

[0144] Table 3 Correlation analysis between GEH and CMD

[0145]

[0146] It is worth mentioning that this scheme specially selects the following three global cardiac electrical heterogeneity parameters: SVG area elevation Area SVG Elevation, VM area under the QT interval curve AUC of Vector Magnitude, and QT interval QTInterval. The inventor's research shows that these three global cardiac electrical heterogeneity parameters can better capture the differentiated ECG feature differences between CMD and OCAD compared with existing ECG features (such as maximum ST segment depression, ST segment depression duration, time required for ST segment depression to 1 mm, and HRV parameters extracted from Holter ECG).

[0147] Embodiment 3

[0148] The present invention provides an OCAD and CMD recognition model based on cardiac global electrical heterogeneity parameters, which is a regression model that records the regression relationship between SVG area elevation angle, VM area under the QT interval curve and QT interval of cardiac global electrical heterogeneity parameters extracted based on 12-lead ECG signals and OCAD or CMD, or a model trained by SVG area elevation angle, VM area under the QT interval curve and QT interval of cardiac global electrical heterogeneity parameters extracted based on 12-lead ECG signals that mark OCAD or CMD.

[0149] The OCAD and CMD recognition model based on cardiac global electrical heterogeneity parameters identifies OCAD and CMD based on the SVG area elevation angle of cardiac global electrical heterogeneity parameters extracted from 12-lead ECG signals, the area under the QT interval curve of VM, and the QT interval.

[0150] For the same technical contents in the second embodiment, please refer to the first embodiment, and no redundant description is given here.

[0151] Embodiment 4:

[0152] The present application embodiment provides an application method for identifying a combination of cardiac global electrical inhomogeneity parameters for OCAD and CMD. Specifically, the method includes:

[0153] Collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal;

[0154] A data segment beat is selected from the VCG signal of each lead, a single lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval are calculated;

[0155] The SVG area elevation, the area under the QT curve of VM, and the QT interval were input into a recognition model to identify the subject as having OCAD or CMD.

[0156] The recognition model of this solution records the regression relationship between the cardiac global electrical heterogeneity parameters and CMD and OCAD, and identifies OCAD and CMD based on the regression relationship. In some embodiments, the recognition model can be a regression model trained using the cardiac global electrical heterogeneity parameters and labeled OCAD and CMD, or a statistical model that records statistical relationships.

[0157] For the same technical contents in the third embodiment, please refer to the first embodiment, and no redundant description is given here.

[0158] Embodiment 5

[0159] The present invention provides an identification device for identifying OCAD and CMD, the identification device comprising a 12-lead electrocardiogram (ECG) device, an electronic data processing device and a display component, wherein the electronic data processing device performs the following steps:

[0160] Collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal;

[0161] A data segment beat is selected from the VCG signal of each lead, a single lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval are calculated;

[0162] Inputting the SVG area elevation angle, the area under the QT interval curve of VM, and the QT interval into a recognition model to identify the subject as having OCAD or CMD;

[0163] The display component is configured to display the recognition result of the electronic data processing device.

[0164] The technical contents of the method executed by the electronic data processing device are the same as those of the first embodiment, and the repeated contents will not be described here in detail.

[0165] Embodiment 6

[0166] This embodiment also provides an electronic device, referring to Figure 8 , comprising a memory 404 and a processor 402, wherein the memory 404 stores a computer program, and the processor 402 is configured to run the computer program to execute the steps in any of the above-mentioned embodiments of the application method for identifying a combination of cardiac global electrical heterogeneity parameters of OCAD and CMD.

[0167] Specifically, the processor 402 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0168] Among them, the memory 404 may include a large capacity memory 404 for data or instructions. By way of example and not limitation, the memory 404 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 404 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 404 may be inside or outside the data processing device. In a specific embodiment, the memory 404 is a non-volatile memory. In a specific embodiment, the memory 404 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, PROM for short), an erasable PROM (Erasable Programmable Read-Only Memory, EPROM for short), an electrically erasable PROM (Electrically Erasable Programmable Read-Only Memory, EEPROM for short), an electrically alterable ROM (Electrically Alterable Read-Only Memory, EAROM for short) or a flash memory (FLASH) or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (Static Random-Access Memory, abbreviated as SRAM) or a dynamic random access memory (Dynamic Random Access Memory, abbreviated as DRAM), wherein the DRAM can be a fast page mode dynamic random access memory 404 (Fast Page Mode Dynamic Random Access Memory, abbreviated as FPMDRAM), an extended data output dynamic random access memory (Extended Date Out Dynamic Random Access Memory, abbreviated as EDODRAM), a synchronous dynamic random access memory (Synchronous Dynamic Random-Access Memory, abbreviated as SDRAM), etc.

[0169] The memory 404 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 402 .

[0170] The processor 402 reads and executes the computer program instructions stored in the memory 404 to implement any one of the OCAD and CMD identification methods based on cardiac global electrical heterogeneity parameters in the above embodiments.

[0171] Optionally, the electronic device may further include a transmission device 406 and an input / output device 408 , wherein the transmission device 406 is connected to the processor 402 , and the input / output device 408 is connected to the processor 402 .

[0172] The transmission device 406 can be used to receive or send data via a network. The specific examples of the above-mentioned network may include a wired or wireless network provided by a communication provider of the electronic device. In one example, the transmission device includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 406 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0173] The input / output device 408 is used to input or output information. In this embodiment, the input information may be the collected ECG signal and VCG signal, etc., and the output information may be the recognition result, etc.

[0174] Optionally, in this embodiment, the processor 402 may be configured to perform the following steps through a computer program:

[0175] Collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal;

[0176] A data segment beat is selected from the VCG signal of each lead, a single lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval are calculated;

[0177] The SVG area elevation, the area under the QT curve of VM, and the QT interval were input into a recognition model to identify the subject as having OCAD or CMD.

[0178] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0179] In general, various embodiments may be implemented in hardware or dedicated circuits, software, logic, or any combination thereof. Some aspects of the invention may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device, but the invention is not limited thereto. Although various aspects of the invention may be shown and described as block diagrams, flow charts, or using some other graphical representation, it should be understood that, as non-limiting examples, the boxes, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0180] Embodiments of the present invention can be implemented by computer software, which is executable by a data processor of a mobile device, such as in a processor entity, or implemented by hardware, or implemented by a combination of software and hardware. Computer software or programs (also referred to as program products) including software routines, applets and / or macros can be stored in any device readable data storage medium, and they include program instructions for performing specific tasks. Computer program products can include one or more computer executable components configured to perform embodiments when the program is running. One or more computer executable components can be at least one software code or a part thereof. In addition, at this point, it should be noted that any box of the logic flow in the figure can represent a program step, or interconnected logic circuits, boxes and functions, or a combination of program steps and logic circuits, boxes and functions. Software can be stored in physical media such as memory chips or storage blocks implemented in processors, magnetic media such as hard disks or floppy disks, and optical media such as, for example, DVDs and data variants thereof, CDs. Physical media are non-transient media.

[0181] Those skilled in the art should understand that the technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0182] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A combination of cardiac global electrical inhomogeneity parameters for identifying OCAD and CMD, characterized in that: include: SVG area elevation, VM area under the QT interval curve and QT interval are the parameters of global cardiac electrical heterogeneity extracted based on 12-lead ECG signals.

2. The cardiac global electrical heterogeneity parameter combination for identifying OCAD and CMD according to claim 1, characterized in that: Collect 12-lead ECG signals and convert them into 3-lead VCG signals. Select data segment beats from the VCG signals of each lead. Use the median algorithm to select the single-lead median beat from the data segment beats. Construct a time-coherent global XYZ median beat based on the single-lead median beat of each lead. Detect the VCG origin on the global XYZ median beat and calculate the vector amplitude. Mark key points based on the vector amplitude. Obtain QRS vector, T vector and SVG vector based on the key points and calculate the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval.

3. The cardiac global electrical inhomogeneity parameter combination for identifying OCAD and CMD according to claim 1, characterized in that: The maximum absolute value point of the first-order derivative is determined based on the VCG signal of the Vx lead, the maximum absolute value point of the first-order derivative is located in the VCG signal of each lead, and data segment beats are selected before and after the maximum absolute value point of the first-order derivative.

4. The cardiac global electrical inhomogeneity parameter combination for identifying OCAD and CMD according to claim 1, characterized in that: Mark the QRS start and end, R peak, and T wave end as key points.

5. A screening method for identifying a combination of cardiac global electrical heterogeneity parameters for OCAD and CMD, characterized in that: Collect 12-lead ECG signals from different OCAD patients and different CMD patients, and convert the 12-lead ECG signals into 3-lead VCG signals; A data segment beat is selected from the VCG signal of each lead, a single-lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single-lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and global cardiac electrical inhomogeneity parameters are calculated; Binary univariate logistic regression analysis was performed using each cardiac global electrical heterogeneity parameter as an independent variable and CMD or OCAD as a dependent variable, and the cardiac global electrical heterogeneity parameter with a P value less than the first threshold was selected as the first undetermined parameter; The combination of the first undetermined parameters is taken as the independent variable, and CMD or OCAD is taken as the dependent variable for multicollinearity analysis, and the combination whose variance inflation factor of each first undetermined parameter in the combination is less than the second threshold is retained, and the first undetermined parameter in the combination is taken as the second undetermined parameter; The combination of the second undetermined parameters was taken as the independent variable, and CMD or OCAD was used as the dependent variable for binary multivariate logistic regression analysis, and the second undetermined parameters with p values ​​less than the third threshold were screened as parameter combinations.

6. The method for screening a combination of cardiac global electrical heterogeneity parameters for identifying OCAD and CMD according to claim 5, characterized in that: The parameters of global cardiac electrical heterogeneity are: QRS-T peak angle, QRS-T area angle, area of ​​VM under the QT interval curve, QT interval, QRS peak azimuth, QRS peak elevation, QRS area azimuth, QRS area elevation, QRS amplitude peak, QRS area, T wave peak azimuth, T wave peak elevation, T wave area azimuth, T wave area elevation, T wave amplitude peak, T wave area, SVG peak azimuth, SVG peak elevation, SVG area azimuth, SVG area elevation, SVG amplitude peak, and Wilson SVG.

7. A recognition model for identifying OCAD and CMD based on cardiac global electrical heterogeneity parameters, characterized in that: A regression model is used to record the regression relationship between SVG area elevation, VM area under the QT interval curve and QT interval, which are global electrical heterogeneity parameters of the heart extracted based on 12-lead ECG signals, and OCAD or CMD, or a model is trained by SVG area elevation, VM area under the QT interval curve and QT interval, which are global electrical heterogeneity parameters of the heart extracted based on 12-lead ECG signals, marking OCAD or CMD.

8. An application method for identifying a combination of cardiac global electrical inhomogeneity parameters for OCAD and CMD, characterized in that: The method comprises: Collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal; A data segment beat is selected from the VCG signal of each lead, a single lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and the SVG area elevation angle, the area of ​​VM under the QT interval curve and the QT interval are calculated; The SVG area elevation, the area under the QT curve of VM, and the QT interval were input into a recognition model to identify the subject as having OCAD or CMD.

9. A device for identifying OCAD and CMD based on cardiac global electrical heterogeneity parameters, comprising: A 12-lead electrocardiogram device, an electronic data processing device and a display component, wherein the electronic data processing device performs the following steps: collecting a 12-lead ECG signal of a subject, and converting the 12-lead ECG signal into a 3-lead VCG signal; A data segment beat is selected from the VCG signal of each lead, a single-lead median beat is selected from the data segment beat using a median algorithm, and a time-coherent global XYZ median beat is constructed based on the single-lead median beat of each lead, the VCG origin is detected on the global XYZ median beat and the vector amplitude is calculated, key points are marked based on the vector amplitude, QRS vectors, T vectors and SVG vectors are obtained based on the key points, and global cardiac electrical inhomogeneity parameters are calculated; Inputting the cardiac global electrical heterogeneity parameter into a recognition model to identify the subject as having OCAD or CMD; The display component is configured to display the recognition result of the electronic data processing device.

10. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the application method of any one of claim 8 for identifying a combination of cardiac global electrical heterogeneity parameters for OCAD and CMD.

Images