A method and device for identifying abnormal changes in electrocardiogram ST-T

Abstract

The present application relates to the field of signal recognition technology, and more specifically, the present application relates to a method and device for identifying abnormal changes in ECG ST-T. The method includes: acquiring an ECG signal; preprocessing the acquired ECG signal; performing heartbeat detection on the preprocessed ECG signal to obtain first heartbeat information; evaluating the first heartbeat information to determine second heartbeat information that meets preset conditions; constructing a feature vector of the second heartbeat information; inputting the second heartbeat information and the feature vector into a trained neural network to obtain the confidence of the type of abnormal change in ECG ST-T. The present application can detect abnormal changes in ECG ST-T in a more targeted manner. In particular, the constructed seven-dimensional feature vector participates in the processing process of the neural network, which can more accurately detect abnormal changes in ECG ST-T and thus improve the accuracy of ECG ST-T analysis.

Description

Technical Field

[0001] The present application relates to the technical field of signal recognition, and more specifically, to a method and device for recognizing abnormal changes in electrocardiogram ST-T. Background Art

[0002] ECG signals reflect the electrophysiological process of cardiac activity. Normal ECG signals are generally composed of P waves, QRS waves, and T waves, and sometimes U waves. The P wave represents the electrical activity of atrial contraction, while the QRS wave and T wave represent the electrical activity of ventricular contraction. The shape, amplitude, and duration of each ECG waveform can be used to analyze and evaluate cardiovascular health.

[0003] According to clinical statistics, ST-T abnormalities account for the highest proportion of all abnormal electrocardiograms. According to the morphological characteristics of the electrocardiogram, ST-T changes are divided into specific changes and non-specific changes. Specific changes refer to the morphological characteristics of ST-T that can be used to indicate the cause and assist in diagnosis, including ischemic ST segment depression, damaged ST elevation, ischemic T wave changes, coronary T waves, ST-T hook-shaped changes, etc. Non-specific changes refer to ST-T changes that are beyond the normal range, but their morphological changes are not specific and cannot be used to diagnose the disease. For secondary changes, in addition to ST-T changes, their manifestations are also accompanied by abnormal changes in the QRS complex.

[0004] In traditional ECG analysis methods, the process for ST-T (referring to the end point of the QRS wave to the end point of the T wave) changes generally first detects the heartbeat and reference points, then extracts the ST segment and T wave based on the detected QRS wave end point (J point) and T wave end point, calculates the ST segment offset value and morphology, T wave amplitude and other features used to represent ST-T morphological information, and finally makes a judgment based on pre-set rules. Due to the diverse morphology of ST-T changes and the low amplitude of the ECG signal, it is easy to be disturbed by external noise when analyzing the morphological information of the ST segment and T wave, making it difficult to accurately identify the type of ST-T change. Summary of the invention

[0005] Based on the above technical problems, the present invention aims to analyze the waveform using a neural network, and analyze the QRS wave, ST segment and T wave respectively after extracting the reference point to determine the abnormal type of ST-T. By inputting signal segments, mask segments and constructed feature vectors into the trained neural network, the abnormal change type of ST-T can be accurately identified.

[0006] The first aspect of the present invention provides a method for identifying abnormal changes in electrocardiogram ST-T, comprising:

[0007] Obtain ECG signals;

[0008] Preprocessing the acquired ECG signal;

[0009] Performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information;

[0010] Evaluate the first heartbeat information to determine second heartbeat information that meets a preset condition;

[0011] constructing a feature vector of the second heartbeat information;

[0012] The second heartbeat information and the feature vector are input into a trained neural network to obtain the confidence level of the abnormal change type of the electrocardiogram ST-T.

[0013] Specifically, the preprocessing of the acquired electrocardiogram signal includes:

[0014] Resample the acquired ECG signal;

[0015] Filter out the high-frequency noise of the resampled ECG signal;

[0016] Extract the ECG baseline from the filtered ECG signal.

[0017] Specifically, performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information includes:

[0018] The preprocessed ECG signal is used as a signal to be detected;

[0019] Perform heartbeat detection on the signal to be detected to obtain first heartbeat information, wherein the first heartbeat information includes the position, type and reference point of the heartbeat, wherein the reference point includes the P wave starting point, the P wave end point, the QRS wave starting point, the QRS wave end point and the T wave end point.

[0020] More specifically, the evaluating the first heartbeat information to determine the second heartbeat information that meets a preset condition includes:

[0021] determining a preset number of heartbeats to be evaluated from the first heartbeat information;

[0022] For the preset number of heartbeats to be evaluated, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate a heartbeat signal segment and a baseline segment respectively;

[0023] Counting the energy proportion of the high-frequency part and the peak-to-peak value of the baseline segment on the heartbeat signal segment, wherein the peak-to-peak value represents the difference between the maximum value and the minimum value;

[0024] The segments whose high-frequency energy ratio exceeds the threshold and whose baseline segment peak-to-peak value exceeds the threshold are removed to obtain the second heart beat information.

[0025] Furthermore, the second heartbeat information is processed to obtain a mask segment, and the mask segment, the second heartbeat information and the feature vector are input into a trained neural network.

[0026] The step of processing the second heartbeat information to obtain the mask segment includes:

[0027] Taking the heartbeat position as the center, the second heartbeat information is extended to 2s in length by adding 0s to both ends of the heartbeat signal;

[0028] A mask segment having the same length as the second heartbeat information is generated, wherein the value of the mask segment from the QRS wave end point to the T wave end point is 1, and the value of the rest of the mask segment is 0.

[0029] Furthermore, constructing the feature vector of the second heartbeat information includes:

[0030] The midpoint between the end point of the P wave and the start point of the QRS wave in the second heartbeat information is used as the ST segment reference point, and the voltage difference relative to the reference point is calculated at 20ms, 40ms, 60ms and 80ms after the end point of the QRS wave;

[0031] The T wave segment was defined as the period from 80 ms after the end of the QRS wave to the end of the T wave, and the maximum, minimum, and standard deviation of the T wave segment were calculated.

[0032] A seven-dimensional feature vector is constructed based on the voltage difference relative to the reference point, the maximum value, the minimum value and the standard deviation of the T wave segment.

[0033] A second aspect of the present invention provides a device for identifying abnormal changes in electrocardiogram ST-T, the device comprising:

[0034] An acquisition module, used for acquiring electrocardiogram signals;

[0035] A preprocessing module, used for preprocessing the acquired ECG signal;

[0036] A heartbeat detection module, used for performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information;

[0037] A heartbeat evaluation module, used to evaluate the first heartbeat information and determine second heartbeat information that meets a preset condition;

[0038] A feature vector construction module, used to construct a feature vector of the second heartbeat information;

[0039] The type recognition module is used to input the second heartbeat information and the feature vector into the trained neural network to obtain the confidence of the type of abnormal change of the electrocardiogram ST-T.

[0040] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0041] Obtain ECG signals;

[0042] Preprocessing the acquired ECG signal;

[0043] Performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information;

[0044] Evaluate the first heartbeat information to determine second heartbeat information that meets a preset condition;

[0045] constructing a feature vector of the second heartbeat information;

[0046] The second heartbeat information and the feature vector are input into a trained neural network to obtain the confidence level of the abnormal change type of the electrocardiogram ST-T.

[0047] A fourth aspect of the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the following steps:

[0048] Obtain ECG signals;

[0049] Preprocessing the acquired ECG signal;

[0050] Performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information;

[0051] Evaluate the first heartbeat information to determine second heartbeat information that meets a preset condition;

[0052] constructing a feature vector of the second heartbeat information;

[0053] The second heartbeat information and the feature vector are input into a trained neural network to obtain the confidence level of the abnormal change type of the electrocardiogram ST-T.

[0054] The beneficial effects of the present application are as follows: the method described in the present application performs heartbeat detection on the preprocessed ECG signal to obtain the first heartbeat information, evaluates the first heartbeat information, determines the second heartbeat information that meets the preset conditions, constructs the feature vector of the second heartbeat information, inputs the second heartbeat information and the feature vector into the trained neural network, obtains the confidence of the abnormal change type of the ECG ST-T, and can detect the abnormal change of the ECG ST-T more specifically. In particular, the constructed seven-dimensional feature vector participates in the processing process of the neural network, which can more accurately detect the abnormal change of the ECG ST-T; and the second heartbeat information is processed to obtain the mask fragment, and the mask fragment is also input into the trained neural network, so that the output result of the neural network is more accurate, and an objective evaluation of the cardiovascular health status can be given, thereby improving the accuracy of the ECG ST-T analysis and promoting the accuracy of the diagnosis of myocardial related diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings, which constitute a part of the specification, illustrate embodiments of the present application and, together with the description, serve to explain the principles of the present application.

[0056] The present application can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:

[0057] Figure 1 A schematic diagram of the steps of a method for identifying abnormal changes in ECG ST-T in an exemplary embodiment of the present application is shown;

[0058] Figure 2 A schematic diagram of a neural network structure and its processing process in an exemplary embodiment of the present application is shown;

[0059] Figure 3 A typical electrocardiogram signal diagram in an exemplary embodiment of the present application is shown;

[0060] Figure 4 An electrocardiogram showing abnormal ST-T changes;

[0061] Figure 5 A schematic diagram of the device structure of an exemplary embodiment of the present application is shown;

[0062] Figure 6 A schematic diagram of the structure of an electronic device provided by an exemplary embodiment of the present application is shown;

[0063] Figure 7 A schematic diagram of a storage medium provided by an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0064] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present application. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present application. It is obvious to those skilled in the art that the present application can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present application, some technical features known in the art are not described.

[0065] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of the features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0066] Now, exemplary embodiments according to the present application will be described in more detail with reference to the accompanying drawings. However, these exemplary embodiments may be implemented in a variety of different forms and should not be construed as being limited to the embodiments described herein. The accompanying drawings are not drawn to scale, and certain details may be magnified and certain details may be omitted for the purpose of clear expression. The shapes of the various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are merely exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0067] The following is attached to the instruction manual Figure 1-7 Several embodiments are given to describe exemplary implementations according to the present application. It should be noted that the following application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementations of the present application are not limited in this regard. On the contrary, the implementations of the present application can be applied to any applicable scenario.

[0068] Embodiment 1:

[0069] This embodiment implements a method for identifying abnormal changes in ECG ST-T. Figure 1 As shown, including:

[0070] S1, obtaining ECG signals;

[0071] S2, preprocessing the acquired ECG signal;

[0072] S3, performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information;

[0073] S4, evaluating the first heartbeat information to determine second heartbeat information that meets a preset condition;

[0074] S5, constructing a feature vector of the second heartbeat information;

[0075] S6. Input the second heartbeat information and the feature vector into the trained neural network to obtain the confidence level of the abnormal change type of the electrocardiogram ST-T.

[0076] In a specific implementation, the acquired ECG signal is preprocessed, including: resampling the acquired ECG signal; filtering out the high-frequency noise of the resampled ECG signal; and extracting the ECG baseline from the filtered ECG signal. Preferably, the signal is resampled to 250 Hz, and then high-frequency noise with a frequency greater than 40 Hz such as electromyographic signals and power frequency interference is filtered out, and the ECG baseline is extracted for subsequent signal quality evaluation.

[0077] Specifically, performing heartbeat detection on the preprocessed ECG signal to obtain first heartbeat information includes: using the preprocessed ECG signal as a signal to be detected; performing heartbeat detection on the signal to be detected to obtain first heartbeat information, the first heartbeat information including the position, type and reference point of the heartbeat, wherein the reference point includes the P wave starting point, the P wave end point, the QRS wave starting point, the QRS wave end point and the T wave end point. The heartbeat types here include sinus, supraventricular, ventricular and other types.

[0078] More specifically, the first heartbeat information is evaluated to determine the second heartbeat information that meets the preset conditions, including: determining a preset number of heartbeats to be evaluated from the first heartbeat information; for the preset number of heartbeats to be evaluated, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate heartbeat signal segments and baseline segments respectively; counting the energy proportion of the high-frequency part and the peak-to-peak value of the baseline segment on the heartbeat signal segment, wherein the peak-to-peak value represents the difference between the maximum value and the minimum value; removing the segments where the energy proportion of the high-frequency part exceeds the threshold and the peak-to-peak value of the baseline segment exceeds the threshold, to obtain the second heartbeat information. Here, for the preset number of heartbeats to be evaluated, if the number of heartbeats is greater than or equal to 3, for the second heartbeat to the second to last heartbeat, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate heartbeat signal segments and baseline segments respectively. The energy proportion of the high-frequency part (>40Hz) and the peak-to-peak value of the baseline segment (i.e., the difference between the maximum value and the minimum value). Remove high-frequency energy that exceeds a threshold, such as 30%, and baseline segments whose peak-to-peak value exceeds a threshold, such as 0.2mV.

[0079] Furthermore, the second heartbeat information is processed to obtain a mask segment, and the mask segment, the second heartbeat information and the feature vector are input into the trained neural network, such as Figure 2 As shown, the second heartbeat information here is the signal segment. Among them, a preferred method of processing the second heartbeat information to obtain the mask segment includes: taking the heartbeat position as the center, extending the second heartbeat information to a length of 2s by adding 0 at both ends of the heartbeat signal; generating a mask segment of the same length as the second heartbeat information, wherein the value of the mask segment from the QRS wave end point to the T wave end point is 1, and the value of the rest is 0.

[0080] In a possible specific implementation, constructing a feature vector of the second heartbeat information includes: taking the midpoint from the end of the P wave to the start of the QRS wave in the second heartbeat information as the ST segment reference point, calculating the voltage difference relative to the reference point at 20ms, 40ms, 60ms and 80ms after the end of the QRS wave; taking 80ms after the end of the QRS wave to the end of the T wave as the T wave segment, and counting the maximum, minimum and standard deviation of the T wave segment; constructing a seven-dimensional feature vector based on the voltage difference relative to the reference point, the maximum, minimum and standard deviation of the T wave segment. The formula for constructing a seven-dimensional feature vector is as follows:

[0081] fea param =[st 20 , st 40 , st 60 , st 80 , t + , t - , t std ]

[0082] st 20 =stj 20 -st ref

[0083] st 40 =stj 40 -st ref

[0084] st 60 =stj 60 -st ref

[0085] st 80 =stj 80 -st ref

[0086]

[0087]

[0088] st 20st 40 st 60 st 80 They represent the voltage difference relative to the reference point at 20ms, 40ms, 60ms and 80ms after the end point of the QRS wave, respectively. + t - t std They represent the maximum, minimum and standard deviation of the T wave segment respectively.

[0089] In a specific implementation, the signal segment, the mask segment and the feature vector are input into a trained neural network to obtain the confidence level of the abnormal change type of the ECG ST-T.

[0090] Embodiment 2:

[0091] This embodiment implements a method for identifying abnormal changes in ECG ST-T, and the steps include: acquiring an ECG signal; preprocessing the acquired ECG signal; performing heartbeat detection on the preprocessed ECG signal to obtain first heartbeat information; evaluating the first heartbeat information to determine second heartbeat information that meets preset conditions; processing the second heartbeat information to obtain a mask fragment; constructing a feature vector of the second heartbeat information; inputting the second heartbeat information, the mask fragment and the feature vector into a trained neural network to obtain the confidence of the type of abnormal change in ECG ST-T.

[0092] Figure 3 A typical ECG waveform is given, such as Figure 3 As shown in the figure, a normal ECG signal is generally composed of P wave, QRS complex wave and T wave, and occasionally U wave. Among them, P wave represents the electrical activity of atrial contraction, QRS wave and T wave represent the electrical activity of ventricular contraction. The shape, amplitude and duration of each ECG waveform can be used to analyze and evaluate the health of the cardiovascular system. Figure 4An electrocardiogram with ST-T abnormality is given (the figure is only for illustration, and the characters in the figure are not limited). According to clinical statistics, ST-T abnormality accounts for the highest proportion of all abnormal electrocardiograms, about 50%. ST-T abnormality is a manifestation of abnormal myocardial repolarization. According to the mechanism of occurrence, ST-T abnormal changes can be divided into primary changes and secondary changes. Primary changes refer to abnormal ventricular repolarization caused by the myocardium itself. Myocardial ischemia is the most common and important cause of ST-T changes. Secondary changes are mainly caused by abnormal ventricular depolarization and subsequent abnormal ventricular repolarization, which are mainly seen in diseases such as left ventricular hypertrophy and left bundle branch block. According to the morphological characteristics of the electrocardiogram, ST-T abnormal changes are divided into specific changes and non-specific changes. Specific changes refer to the morphological characteristics of ST-T, which can be used to indicate the cause and assist in diagnosis, including ischemic ST segment depression, injury-type ST elevation, ischemic T wave changes, coronary T waves, ST-T hook-shaped changes, etc. Nonspecific changes refer to changes in ST-T that are beyond the normal range, but their morphological changes are not specific and cannot be used to diagnose the disease. For secondary changes, in addition to ST-T changes, their manifestations are also accompanied by abnormal changes in the QRS complex.

[0093] In a specific implementation, the acquired ECG signal is preprocessed, including: resampling the acquired ECG signal; filtering out the high-frequency noise of the resampled ECG signal; and extracting the ECG baseline from the filtered ECG signal. Preferably, the signal is resampled to 250 Hz, and then high-frequency noise with a frequency greater than 40 Hz such as electromyographic signals and power frequency interference is filtered out, and the ECG baseline is extracted for subsequent signal quality evaluation.

[0094] Specifically, performing heartbeat detection on the preprocessed ECG signal to obtain first heartbeat information includes: using the preprocessed ECG signal as a signal to be detected; performing heartbeat detection on the signal to be detected to obtain first heartbeat information, the first heartbeat information including the position, type and reference point of the heartbeat, wherein the reference point includes the P wave starting point, the P wave end point, the QRS wave starting point, the QRS wave end point and the T wave end point. The heartbeat types here include sinus, supraventricular, ventricular and other types.

[0095] More specifically, the first heartbeat information is evaluated to determine the second heartbeat information that meets the preset conditions, including: determining a preset number of heartbeats to be evaluated from the first heartbeat information; for the preset number of heartbeats to be evaluated, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate heartbeat signal segments and baseline segments respectively; counting the energy proportion of the high-frequency part and the peak-to-peak value of the baseline segment on the heartbeat signal segment, wherein the peak-to-peak value represents the difference between the maximum value and the minimum value; removing the segments where the energy proportion of the high-frequency part exceeds the threshold and the peak-to-peak value of the baseline segment exceeds the threshold, to obtain the second heartbeat information. Here, for the preset number of heartbeats to be evaluated, if the number of heartbeats is greater than or equal to 3, for the second heartbeat to the second to last heartbeat, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate heartbeat signal segments and baseline segments respectively. The energy proportion of the high-frequency part (>40Hz) and the peak-to-peak value of the baseline segment (i.e., the difference between the maximum value and the minimum value). Remove high-frequency energy that exceeds a threshold, such as 30%, and baseline segments whose peak-to-peak value exceeds a threshold, such as 0.2mV.

[0096] Furthermore, the second heartbeat information is processed to obtain a mask segment, and the mask segment, the second heartbeat information and the feature vector are input into the trained neural network, such as Figure 2 As shown, the second heartbeat information here is the signal segment. Among them, a preferred method of processing the second heartbeat information to obtain the mask segment includes: taking the heartbeat position as the center, extending the second heartbeat information to a length of 2s by adding 0 at both ends of the heartbeat signal; generating a mask segment of the same length as the second heartbeat information, wherein the value of the mask segment from the QRS wave end point to the T wave end point is 1, and the value of the rest is 0.

[0097] In a possible specific implementation, constructing a feature vector of the second heartbeat information includes: taking the midpoint from the end of the P wave to the start of the QRS wave in the second heartbeat information as the ST segment reference point, calculating the voltage difference relative to the reference point at 20ms, 40ms, 60ms and 80ms after the end of the QRS wave; taking 80ms after the end of the QRS wave to the end of the T wave as the T wave segment, and counting the maximum, minimum and standard deviation of the T wave segment; constructing a seven-dimensional feature vector based on the voltage difference relative to the reference point, the maximum, minimum and standard deviation of the T wave segment. The formula for constructing a seven-dimensional feature vector is as follows:

[0098] fea param =[st 20 , st 40 , st 60 , st 80 , t + , t - , t std ]

[0099] st20 =stj 20 -st ref

[0100] st 40 =stj 40 -st ref

[0101] st 60 =stj 60 -st ref

[0102] st 80 =stj 80 -st ref

[0103]

[0104]

[0105] st 20 、st 40 、st 60 、st 80 They represent the voltage difference relative to the reference point at 20ms, 40ms, 60ms and 80ms after the end point of the QRS wave, respectively. + ,t - ,t std They represent the maximum, minimum and standard deviation of the T wave segment respectively.

[0106] For example Figure 2As shown in the figure, the structure of the deep neural network consists of a convolution block (C0), a residual block (C1, C2 and C3), a downsampling layer, a global pooling layer and a fully connected layer. The structures of C1, C2 and C3 are the same, and only the number of channels (c) is different. After the heartbeat signal fragment passes through the residual block C3, a deep feature matrix with a length of 1 / 4 of the original signal and a channel number of 128 is obtained. At this time, the mask fragment is resampled, the length of the mask fragment is adjusted to be consistent with the deep feature matrix, and then multiplied point by point with each channel of the deep feature matrix. After that, the deep feature matrix passes through the global pooling layer to generate a 128-dimensional deep feature vector, which is then spliced ​​with the input heartbeat feature parameter vector (i.e., 7-dimensional feature vector) into a 135-dimensional feature vector. After passing through the fully connected layer, a 3-dimensional vector is output. The values ​​in the vector represent the confidence (i.e., probability) of the three types of ST-T abnormal changes, no abnormality, specific changes or non-specific changes, and finally the type with the highest probability is output. In the convolution layer, the size of the convolution kernel is 19 sample points, the padding length is 9 sample points, and the step length is 1 sample point. The number of convolution kernels is 32 (C0 and C1), 64 (C3), and 128 (C4). Of course, the neural network needs to be trained in advance, and it can only be considered trained after iterating to a preset number of times. The specific training process belongs to the general technology in the relevant technical field and is not limited here.

[0107] The method described in the present application performs heartbeat detection on the pre-processed ECG signal to obtain the first heartbeat information, evaluates the first heartbeat information, determines the second heartbeat information that meets the preset conditions, constructs the feature vector of the second heartbeat information, inputs the second heartbeat information and the feature vector into the trained neural network, obtains the confidence of the abnormal change type of the ECG ST-T, and can detect the abnormal change of the ECG ST-T more specifically. In particular, the constructed seven-dimensional feature vector participates in the processing process of the neural network, which can more accurately detect the abnormal change of the ECG ST-T; and the second heartbeat information is processed to obtain the mask fragment, and the mask fragment is also input into the trained neural network, so that the output result of the neural network is more accurate, and an objective evaluation of the cardiovascular health status can be given, thereby improving the accuracy of the ECG ST-T analysis and promoting the accuracy of the diagnosis of myocardial related diseases.

[0108] Embodiment 3:

[0109] This embodiment provides a device for identifying abnormal changes in ECG ST-T. Figure 5 As shown, including:

[0110] An acquisition module 501 is used to acquire an electrocardiogram signal;

[0111] A preprocessing module 502, used for preprocessing the acquired ECG signal;

[0112] A heartbeat detection module 503 is used to perform heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information;

[0113] A heartbeat evaluation module 504, configured to evaluate the first heartbeat information and determine second heartbeat information that meets a preset condition;

[0114] A feature vector construction module 505, used to construct a feature vector of the second heartbeat information;

[0115] The type identification module 506 is used to input the second heartbeat information and the feature vector into the trained neural network to obtain the confidence of the type of abnormal change of the electrocardiogram ST-T.

[0116] The identification device also includes a mask module, which is used to process the second heartbeat information to obtain a mask segment. The mask module specifically performs the following steps in the process of obtaining the mask segment: taking the heartbeat position as the center, extending the second heartbeat information to a length of 2s by adding 0s to both ends of the heartbeat signal; generating a mask segment of the same length as the second heartbeat information, wherein the value of the mask segment from the QRS wave end point to the T wave end point is 1, and the value of the rest is 0.

[0117] The type recognition module will input the mask fragment together with the second heartbeat information and the feature vector into the trained neural network. Of course, the neural network needs to be trained in advance and can only be considered trained after iterating to a preset number of times.

[0118] Please refer to the following Figure 6 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 6 As shown, the electronic device 2 includes: a processor 200, a memory 201, a bus 202 and a communication interface 203, and the processor 200, the communication interface 203 and the memory 201 are connected via the bus 202; the memory 201 stores a computer program that can be run on the processor 200, and when the processor 200 runs the computer program, the method for identifying abnormal changes in the electrocardiogram ST-T provided in any of the aforementioned embodiments of the present application is executed, and the electronic device can be an electronic device with a touch-sensitive display.

[0119] The memory 201 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 203 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0120] The bus 202 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 201 is used to store a program, and the processor 200 executes the program after receiving an execution instruction. The method for identifying abnormal changes in the electrocardiogram ST-T disclosed in any of the embodiments of the present application may be applied to the processor 200, or implemented by the processor 200.

[0121] The processor 200 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 200. The above processor 200 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 201, and the processor 200 reads the information in the memory 201 and completes the steps of the above method in combination with its hardware.

[0122] The electronic device provided in the embodiment of the present application and the method for identifying abnormal changes in ECG ST-T provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0123] The present application also provides a computer-readable storage medium corresponding to the method for identifying abnormal changes in ECG ST-T provided in the above embodiment. Figure 7 , Figure 7 The computer-readable storage medium shown is a CD 30 on which a computer program (ie, a program product) is stored. When the computer program is executed by a processor, the method for identifying abnormal changes in electrocardiogram ST-T provided in any of the aforementioned embodiments is executed.

[0124] In addition, examples of the computer-readable storage medium may also 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 optical or magnetic storage media, which are not listed here one by one.

[0125] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the method for allocating channels for quantum key distribution in a space-division multiplexed optical network provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0126] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method for identifying abnormal changes in the electrocardiogram ST-T provided in any of the aforementioned embodiments, the steps of the method including: acquiring an electrocardiogram signal; preprocessing the acquired electrocardiogram signal; performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information; evaluating the first heartbeat information to determine second heartbeat information that meets preset conditions; constructing a feature vector of the second heartbeat information; and inputting the second heartbeat information and the feature vector into a trained neural network to obtain the confidence of the type of abnormal change in the electrocardiogram ST-T.

[0127] It should be noted that the algorithms and displays provided herein are not inherently related to any particular computer, virtual device or other device. Various general devices can also be used together with the teachings based thereon. According to the above description, it is obvious that the structure required for constructing such devices is constructed. In addition, the present application is not directed to any specific programming language. It should be understood that the content of the present application described herein can be implemented using various programming languages, and the above description of specific languages ​​is to disclose the best implementation mode of the present application. In the specification provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this specification.

[0128] Similarly, it should be understood that in order to streamline the present application and aid in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present application, the various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: the claimed application requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, the inventive aspects lie in less than all the features of the single embodiment disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present application.

[0129] Those skilled in the art will appreciate that the modules in the devices in the embodiments can be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition they can be divided into multiple submodules or subunits or subassemblies. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification and all processes or units of any method or device disclosed in this manner can be combined in any combination. Unless otherwise clearly stated, each feature disclosed in this specification can be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0130] The various component embodiments of the present application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all functions of some or all components in the creation device of the virtual machine according to the embodiment of the present application. The present application can also be implemented as a device or device program for executing part or all of the methods described herein. The program implementing the present application can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0131] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for identifying abnormal changes in electrocardiogram ST-T, characterized in that: include: Obtain ECG signals; Preprocessing the acquired ECG signal; Performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information, wherein the first heartbeat information includes the position, type and reference point of the heartbeat, wherein the reference point includes a P wave starting point, a P wave end point, a QRS wave starting point, a QRS wave end point and a T wave end point; Evaluate the first heartbeat information to determine second heartbeat information that meets a preset condition; constructing a feature vector of the second heartbeat information; Processing the second heartbeat information to obtain a mask segment, inputting the mask segment together with the second heartbeat information and the feature vector into a trained neural network to obtain a confidence level of the type of abnormal change in the electrocardiogram ST-T; The step of evaluating the first heartbeat information to determine second heartbeat information that meets a preset condition includes: determining a preset number of heartbeats to be evaluated from the first heartbeat information; For the preset number of heartbeats to be evaluated, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate a heartbeat signal segment and a baseline segment respectively; Counting the energy proportion of the high-frequency part and the peak-to-peak value of the baseline segment on the heartbeat signal segment, wherein the peak-to-peak value represents the difference between the maximum value and the minimum value; The segments whose high-frequency energy ratio exceeds the threshold and whose baseline segment peak-to-peak value exceeds the threshold are removed to obtain the second heart beat information; The step of processing the second heartbeat information to obtain a mask segment includes: Taking the heartbeat position as the center, the second heartbeat information is extended to 2s in length by adding 0s to both ends of the heartbeat signal; A mask segment having the same length as the second heartbeat information is generated, wherein the value of the mask segment from the QRS wave end point to the T wave end point is 1, and the value of the rest of the mask segment is 0.

2. The method for identifying abnormal changes in electrocardiogram ST-T according to claim 1, characterized in that: The preprocessing of the acquired electrocardiogram signal comprises: Resample the acquired ECG signal; Filter out the high-frequency noise of the resampled ECG signal; Extract the ECG baseline from the filtered ECG signal.

3. The method for identifying abnormal changes in electrocardiogram ST-T according to claim 1, characterized in that: The constructing the feature vector of the second heartbeat information includes: The midpoint between the end point of the P wave and the start point of the QRS wave in the second heartbeat information is used as the ST segment reference point, and the voltage difference relative to the reference point is calculated at 20ms, 40ms, 60ms and 80ms after the end point of the QRS wave; The T wave segment was defined as the period from 80 ms after the end of the QRS wave to the end of the T wave, and the maximum, minimum, and standard deviation of the T wave segment were calculated. A seven-dimensional feature vector is constructed based on the voltage difference relative to the reference point, the maximum value, the minimum value and the standard deviation of the T wave segment.

4. A device for identifying abnormal changes in electrocardiogram ST-T, characterized in that: The device comprises: An acquisition module, used for acquiring electrocardiogram signals; A preprocessing module, used for preprocessing the acquired ECG signal; A heartbeat detection module, used for performing heartbeat detection on the preprocessed electrocardiogram signal to obtain first heartbeat information, wherein the first heartbeat information includes the position, type and reference point of the heartbeat, wherein the reference point includes the P wave starting point, the P wave end point, the QRS wave starting point, the QRS wave end point and the T wave end point; A heartbeat evaluation module, used to evaluate the first heartbeat information and determine second heartbeat information that meets a preset condition; A feature vector construction module, used to construct a feature vector of the second heartbeat information; a type recognition module, configured to process the second heartbeat information to obtain a mask segment, input the mask segment together with the second heartbeat information and the feature vector into a trained neural network, and obtain a confidence level of the type of abnormal ST-T change in the electrocardiogram; The step of evaluating the first heartbeat information to determine second heartbeat information that meets a preset condition includes: determining a preset number of heartbeats to be evaluated from the first heartbeat information; For the preset number of heartbeats to be evaluated, using the electrocardiogram signal and the electrocardiogram baseline between the P wave start point and the T wave end point to generate a heartbeat signal segment and a baseline segment respectively; Counting the energy proportion of the high-frequency part and the peak-to-peak value of the baseline segment on the heartbeat signal segment, wherein the peak-to-peak value represents the difference between the maximum value and the minimum value; The segments whose high-frequency energy ratio exceeds the threshold and whose baseline segment peak-to-peak value exceeds the threshold are removed to obtain the second heart beat information; The step of processing the second heartbeat information to obtain a mask segment includes: Taking the heartbeat position as the center, the second heartbeat information is extended to 2s in length by adding 0s to both ends of the heartbeat signal; A mask segment having the same length as the second heartbeat information is generated, wherein the value of the mask segment from the QRS wave end point to the T wave end point is 1, and the value of the rest of the mask segment is 0.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

6. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

Images