Electrocardiogram processing method, device and equipment, medium and electrocardiogram monitoring system

Abstract

The embodiment of the invention is suitable for the technical field of medical electronics, and provides an electrocardiogram processing method, device and equipment, a medium and an electrocardiogram monitoring system.The method comprises the steps that an electrocardiogram to be processed is divided into a plurality of heart beat segments, and each heart beat segment corresponds to one heart beat in the electrocardiogram; performing data conversion on the electrocardiosignal in each heart beat segment to obtain a noise feature of each heart beat segment; and based on the noise features and a classification mode selected by the user, classifying the plurality of heart beat segments. Through the method, the noise characteristics of the electrocardiosignal can be determined without depending on priori knowledge, so that the processing process of the electrocardiogram is simplified.

Description

Technical Field

[0001] This application belongs to the field of medical electronics technology, and particularly relates to a method, device, equipment, medium and electrocardiogram monitoring system for processing electrocardiograms. Background Art

[0002] An electrocardiogram (ECG) is a comprehensive reflection of the potential changes on the body surface during the cardiac electrical activity process, and can be used to examine cardiac electrical activity and myocardial function.

[0003] However, there may be noise in the electrocardiogram output by an electrocardiograph. The noise may distort the morphological features and interval features of the electrocardiogram leads, thereby causing misdiagnosis of patients and inappropriate treatment. Based on this, when using an electrocardiogram, it is necessary to detect the noise of the electrocardiogram to avoid misdiagnosis.

[0004] Currently, the noise recognition algorithms for electrocardiograms all rely on prior knowledge and experience, and are computationally complex with unsatisfactory effects. The complex noise recognition algorithms for electrocardiograms make the processing of electrocardiograms complicated and the time consumed in the electrocardiogram processing process becomes longer. Summary of the Invention

[0005] In view of this, embodiments of this application provide a method, device, equipment, medium and electrocardiogram monitoring system for processing electrocardiograms, which can detect the noise of electrocardiogram signals without relying on prior knowledge and have a small amount of calculation, thereby improving the processing speed of electrocardiograms.

[0006] The first aspect of the embodiments of this application provides a method for processing an electrocardiogram, including:

[0007] Dividing the electrocardiogram to be processed into multiple heartbeat segments, each of the heartbeat segments corresponding to a heartbeat in the electrocardiogram;

[0008] Performing data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments;

[0009] Classifying the multiple heartbeat segments based on the noise characteristics and the classification method selected by the user.

[0010] The second aspect of the embodiments of this application provides a device for processing an electrocardiogram, including:

[0011] A dividing module, configured to divide the electrocardiogram to be processed into multiple heartbeat segments, each of the heartbeat segments corresponding to a heartbeat in the electrocardiogram;

[0012] A noise characteristic determination module, configured to perform data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments;

[0013] A classification module, configured to classify a plurality of the heartbeat segments based on the noise characteristics and the classification method selected by the user.

[0014] A third aspect of the embodiments of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in the first aspect above is implemented.

[0015] A fourth aspect of the embodiments of the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method described in the first aspect above is implemented.

[0016] A fifth aspect of the embodiments of the present application provides an electrocardiogram monitoring system, including a data acquisition device, a data processing device, and an interaction device, where:

[0017] The data acquisition device is configured to acquire an electrocardiogram.

[0018] The data processing device is configured to divide the electrocardiogram into a plurality of heartbeat segments, each of the heartbeat segments corresponding to a heartbeat in the electrocardiogram; perform data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments.

[0019] The interaction device is configured to obtain the processing method selected by the user for each of the heartbeat segments, and display each of the heartbeat segments based on the processing method and the noise characteristics.

[0020] Compared with the prior art, the embodiments of the present application include the following advantages:

[0021] When processing an electrocardiogram using the solution in the embodiments of the present application, the electrocardiogram to be processed can be divided into a plurality of heartbeat segments, and each heartbeat segment can correspond to a heartbeat in the electrocardiogram; then, data conversion is performed on the electrocardiogram signals in each heartbeat segment, so as to calculate the noise characteristics of each heartbeat segment; based on the noise characteristics of each heartbeat segment and the classification method selected by the user, a plurality of heartbeat segments can be classified. In the embodiments of the present application, when calculating the noise characteristics of the heartbeat segments, simple data conversion can be used for calculation, reducing the amount of calculation, and in the process of calculating the noise characteristics, no prior knowledge is required. Since the amount of calculation of the noise characteristics is small, when processing the electrocardiogram based on the noise characteristics, the time consumed is also shortened, reducing the waiting time of the user. For example, the user can classify the heartbeat segments in the electrocardiogram based on the noise characteristics, thereby shortening the classification time of the heartbeat segments. Description of the Drawings

[0022] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0023] Figure 1 It is a schematic flowchart of the steps of a method for processing an electrocardiogram provided by an embodiment of the present application;

[0024] Figure 2 It is a schematic diagram of an electrocardiogram to be detected provided by an embodiment of the present application;

[0025] Figure 3 It is a schematic diagram of a noise feature provided by an embodiment of the present application;

[0026] Figure 4 It is a schematic diagram of another noise feature provided by an embodiment of the present application;

[0027] Figure 5 It is a schematic diagram of yet another noise feature provided by an embodiment of the present application;

[0028] Figure 6 It is a schematic diagram of a noise feature sequence provided by an embodiment of the present application;

[0029] Figure 7 It is a schematic diagram of an exponential sequence provided by an embodiment of the present application;

[0030] Figure 8 It is a schematic diagram of an exponential difference sequence provided by an embodiment of the present application;

[0031] Figure 9 It is a schematic flowchart of a process for determining a threshold provided by an embodiment of the present application;

[0032] Figure 10 It is a schematic flowchart of another process for determining a threshold provided by an embodiment of the present application;

[0033] Figure 11 It is a schematic flowchart of an electrocardiogram analysis method provided by an embodiment of the present application;

[0034] Figure 12 It is a schematic diagram of a display interface provided by an embodiment of the present application;

[0035] Figure 13 It is a schematic diagram of a display interface split by direction provided by an embodiment of the present application;

[0036] Figure 14 It is a schematic diagram of an electrocardiogram analysis device provided by an embodiment of the present application;

[0037] Figure 15 It is a schematic diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0038] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0039] Dynamic electrocardiogram (Holter) is the most commonly used and simplest non-invasive examination method in cardiology. By means of skin patches, Holter can collect electrocardiogram information for 24 hours and can detect many diseases or risk factors that cannot be detected by ordinary electrocardiograms.

[0040] When processing the electrocardiogram within 24 hours, the number of electrocardiogram signals to be processed is very large. The solution of using a neural network algorithm for noise detection is difficult to meet the requirements in this scenario. On the one hand, when using a neural network algorithm to process a large number of electrocardiogram signals, the calculation amount is large and the time consumption is long. On the other hand, when using a neural network algorithm for processing, it usually needs to rely on prior knowledge and experience. However, there are differences in the electrocardiogram signals detected from different individuals, and the results obtained by relying on prior knowledge and experience do not necessarily reflect individual characteristics.

[0041] Based on this, the embodiments of the present application provide a method for processing electrocardiograms, which can detect electrocardiogram noise without relying on prior knowledge and has a small calculation amount.

[0042] The method for processing electrocardiograms provided by the embodiments of the present application can be applied to computer devices such as tablet computers, notebook computers, ultra-mobile personal computers (UMPCs), netbooks, and personal digital assistants (PDAs). The embodiments of the present application do not impose any restrictions on the specific types of computer devices.

[0043] The technical solution of the present application will be described below through specific embodiments.

[0044] Referring to Figure 1 , a schematic flowchart of the steps of a method for processing electrocardiograms provided by the embodiments of the present application is shown, which may specifically include the following steps:

[0045] S101, divide the electrocardiogram to be processed into multiple heartbeat segments, and each of the heartbeat segments corresponds to a heartbeat in the electrocardiogram.

[0046] The electrocardiogram to be processed above can be obtained by reading the detection data of wearable devices, patches, electrocardiographs, etc. The electrocardiogram can be the electrocardiogram of the user over a period of time. For example, the electrocardiogram can be the electrocardiogram of the user within 24 hours. Figure 2 is a schematic diagram of an electrocardiogram to be detected provided by an embodiment of the present application. As Figure 2 shown, the number of electrocardiogram signals in this electrocardiogram is huge.

[0047] The electrocardiogram can include a large number of heartbeats. The computer device can divide the electrocardiogram into multiple heartbeat segments, so that each heartbeat segment can correspond to a heartbeat.

[0048] In a possible implementation, the computer device can divide the electrocardiogram to be detected based on the R-wave peak. When dividing the electrocardiogram to be detected based on the R-wave peak, the computer device can identify the R-wave peaks of each heartbeat in the electrocardiogram. Exemplarily, the computer device can identify each peak in the electrocardiogram, and then determine the R-wave peak from the peaks. There are many methods for determining the R-wave peak in the prior art, which will not be elaborated here.

[0049] After determining the R-wave peak, the electrocardiogram can be segmented based on the R-wave peak. Exemplarily, taking the R-wave peak as the center, the electrocardiogram with a preset duration before and after the R-wave peak can be intercepted as a heartbeat segment, where the preset duration can be the duration of a normal heartbeat. For example, the preset duration can be 1000 ms, and the starting point Startpos of the divided segment = R - 500 ms, and the ending point Endpos = R + 500 ms. In addition, the heartbeat segment corresponding to the R-wave peak can also be determined according to the previous R-wave peak and the next R-wave peak of the R-wave peak. For example, the starting point Startpos of the heartbeat segment = LastR + 350 ms, and the ending point Endpos = NextR - 200 ms. Among them, LastR is the moment corresponding to the previous R-wave peak, and NextR is the moment corresponding to the next R-wave peak.

[0050] S102. Perform data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments.

[0051] The above noise characteristics are used to quantify the noise in the heartbeat segment. In the embodiments of the present application, different noise characteristics can be obtained by performing different processes on the electrocardiogram signals in the heartbeat segment. For the electrocardiogram in each segment, it can be regarded as a waveform composed of multiple sampling points. When calculating the noise characteristics, data conversion can be performed on the electrocardiogram signals in each heartbeat segment to obtain the noise characteristics of each heartbeat segment. For example, operations such as difference and integration can be performed based on the potential values corresponding to each sampling point to obtain the noise characteristics.

[0052] In a possible implementation, for each heartbeat segment, the computer device can perform a difference operation on the electrocardiogram signal values within the heartbeat segment to obtain the difference values corresponding to the electrocardiogram signal values. The difference operation can be two-point difference, three-point difference, four-point difference, etc., which is not limited here. After the difference operation, the computer device can perform an integral calculation on the absolute values of the difference values within the heartbeat segment to obtain the integral value corresponding to the heartbeat segment; then calculate the mean value based on this integral value as the noise feature. The mean value corresponding to each heartbeat segment can be provided as Figure 3 shown in Figure 3 where the abscissa in

[0053] can be the number of R waves, and the ordinate is the mean value, that is, the value of the noise feature.

[0054] Exemplarily, for the heartbeat segment [Startpos, Endpos], the difference Data_diff can be calculated within the range [Startpos, Endpos]; then the absolute value of the difference abs(Data_diff) can be calculated within the range [Startpos, Endpos]; then the integral of the absolute value of the difference abs(Data_diff) is obtained within the range [Startpos, Endpos] as INT(abs(Data_diff)), and then the mean value of the integral is taken nosIndex = INT(abs(Data_diff)) / (Endpos - Startpos), and nosIndex is used as the current noise feature. The larger the index, the greater the noise. Figure 4 shown in Figure 4 where the abscissa in

[0055] can be the number of R waves, and the ordinate is the sum of the differences, that is, the value of the noise feature.

[0056] In another possible implementation, for each heartbeat segment, the computer device can determine the median value of the ECG signal value within the heartbeat segment; then calculate the difference between each ECG signal value within the heartbeat segment and the median value; then perform sliding integration on the difference within the heartbeat segment to obtain a smoothed signal; then determine the target maximum and target minimum values ​​of the smoothed signal, wherein the target maximum value is a maximum value greater than a preset amplitude, and the target minimum value is a minimum value greater than a preset amplitude; the total number of target maximum values ​​and target minimum values, that is, the number of peaks and valleys greater than the preset amplitude, is used as a noise feature. The number of peaks and valleys greater than the preset amplitude that can correspond to each heartbeat segment can be given as follows: Figure 5 As shown, Figure 5 The horizontal axis in can be the number of R waves, and the vertical axis can be the number of peaks and valleys greater than a preset amplitude, that is, the value of the noise characteristic.

[0057] Exemplarily, for the heartbeat segment [Startpos, Endpos], the median medvalue can be taken within the range of [Startpos, Endpos] as the baseline value; the data to be analyzed data_ana = raw_data-medvalue; the data_ana is slidingly integrated within the range of [Startpos, Endpos], and the signal is smoothed to obtain the signal data_ana_int; the maximum and minimum values ​​of data_ana_int are calculated within the range of [Startpos, Endpos]; the maximum and minimum values ​​with amplitudes lower than a threshold (for example, 0.3mV) are eliminated; the number of maximum and minimum values ​​that meet the amplitude threshold within the range of [Startpos, Endpos] is calculated as the current noise feature, and the larger the indicator, the greater the noise.

[0058] S103: Classify the plurality of heartbeat segments based on the noise feature and a classification method selected by the user.

[0059] The larger the value of the noise feature, the greater the noise in the segment. There is a threshold value that can classify a segment as a noise segment or a non-noise segment. If the value of the noise feature is greater than or equal to the threshold value, the heartbeat segment is a noise segment; if it is less than the threshold value, the heartbeat segment is a non-noise segment. Therefore, the classification threshold can be determined based on multiple noise features of multiple heartbeat segments.

[0060] When determining the classification threshold, a sorting analysis can be performed based on the noise features to determine the threshold. The computer device can sort the noise features according to size to obtain a noise feature sequence.

[0061] Exemplarily, after calculating the noise features, nosIndex[n] can be obtained, where n is the total number of R waves; then the values in nosIndex[n] are sorted to obtain the sorted noise feature sequence sort-nosIndex[n].

[0062] Figure 6 FIG. is a schematic diagram of a noise feature sequence provided by an embodiment of the present application. As Figure 6 shown, the noise feature sequence is arranged in ascending order of noise features.

[0063] Feature amplification processing is performed on each noise feature in the noise feature sequence to obtain the feature mutation positions of the noise feature sequence; according to the feature mutation positions, classification thresholds are determined.

[0064] In a possible implementation manner, feature amplification processing can be performed through exponential calculation. The computer device can perform exponential calculation on each noise feature in the noise feature sequence to obtain an exponential sequence. Figure 7 FIG. is a schematic diagram of an exponential sequence provided by an embodiment of the present application. For example, exponential calculation can be performed on the noise feature sequence sort-nosIndex[n] to obtain the exponential sequence log(sort-nosIndex)[n].

[0065] From Figure 7 it can be seen that there are mutation points in the trend of the exponential values. In order to determine the mutation points, the exponential difference between each exponential in the exponential sequence and the previous exponential can be continuously calculated to obtain an exponential difference sequence. Figure 8 FIG. is a schematic diagram of an exponential difference sequence provided by an embodiment of the present application. The difference sequence can be: diff(log(sort-nosIndex))[n]. Where diff(log(nosIndex)) = log(nosIndex(n)) - log(nosIndex(n - 1)).

[0066] Based on the exponential difference sequence, thresholds for dividing noise segments and non-noise segments can be determined; then according to the thresholds, noise segments with noise are determined from multiple segments. Generally, the proportion of noise signals present in an electrocardiogram is determined. For example, the electrocardiogram signals within 24 hours may generally have 5% noise signals. Based on this, the thresholds can be determined from the difference sequence. The peaks in the above difference sequence can be used to characterize the mutation positions of the noise features. Therefore, in this embodiment, the thresholds can be determined based on the peaks in the difference sequence.

[0067] When determining the threshold, the number of categories to be classified can be determined; thus, based on the number of categories and the peak value of the noise feature sequence after feature amplification processing, the feature mutation position can be determined. If it is necessary to divide into two categories, only one threshold is required. If it is necessary to divide into three categories, two thresholds can be determined.

[0068] In a possible implementation manner, only one threshold can be determined, and this threshold is used to determine whether a segment is a noise segment or a non-noise segment. The computer device can determine whether there is a peak value in the first interval arranged at the back in the exponential difference sequence; if there is a peak value in the first interval, the computer device can determine the threshold based on whether there is a peak value greater than a preset value in the first interval. Exemplarily, if there is a peak value greater than the preset value in the first interval, the computer device can use the value of the noise feature of the segment corresponding to the first peak value greater than the preset value that appears in the first interval as the threshold; if there is no peak value greater than the preset value in the first interval, the computer device can use the value of the noise feature of the segment corresponding to the maximum value of the peak values in the first interval as the threshold. If there is no peak value in the first interval, the computer device can determine the threshold based on whether there is a peak value in the second interval. The second interval includes the first interval and is larger than the first interval. If there is a peak value in the second interval, the computer device can use the value of the noise feature of the segment corresponding to the maximum value of the peak values in the second interval as the threshold; if there is no peak value in the second interval, the computer device can use the value of the noise feature of the segment corresponding to the preset position in the difference sequence as the threshold. After determining the threshold, for each noise segment, if its corresponding exponential difference is greater than or equal to the threshold, it is determined that the segment is a noise segment; if its corresponding exponential difference is less than the threshold, it is determined that the segment is a non-noise segment.

[0069] Exemplarily, the first interval can be the last 5% interval of the exponential difference sequence, the second interval can be the last 10% interval of the exponential difference sequence, and the above-mentioned preset position can be the position corresponding to the last 5% of the exponential difference sequence. In the embodiments of the present application, the peak value where diff(log(sort-nosIndex))[n] is greater than 0.05 can be obtained, the array PeakValue and the number of peak values PeakNum can be obtained, and the segmentation line threshold can be determined based on PeakValue. As Figure 9 shown, it can be determined whether there is a peak value within the last 5% range of the exponential difference sequence. If there is a peak value within the last 5% range, it can be searched backward starting from the 5% position to determine whether there is a peak value with a threshold greater than 0.05. If there is a peak value with a threshold greater than 0.05, the threshold Thd can be determined as the noise coefficient corresponding to the first peak value with a threshold greater than 0.05 that appears. If there is no peak value with a threshold greater than 0.05 within the last 5% range, the threshold Thd can be determined as the noise coefficient corresponding to the peak value with the largest amplitude within the last 5% range.

[0070] In another possible implementation, multiple thresholds can be determined. The multiple thresholds are used to divide the segments into non-noise segments and noise segments of different levels. The computer device can determine whether there is a peak in the second half of the difference sequence; if there is a peak in the second half of the difference sequence, it starts searching for the peak from the middle position of the difference sequence; based on the searched peak, multiple thresholds are determined, and the multiple thresholds are used to divide the segments into non-noise segments and noise segments of different levels. Each time a peak appears, it can indicate that the value of the noise feature here has mutated once. Therefore, the computer device can sequentially use the value of the noise feature of the segment corresponding to the peak that appears as the threshold. Among them, the number of thresholds can be set by the user himself, and the device can select the values of the noise features of the segments corresponding to a specified number of peaks from the peaks in the second half as the thresholds. If there is a peak in the second half of the difference sequence, the value of the noise feature of the segment corresponding to the preset position in the difference sequence is used as the threshold.

[0071] Exemplarily, as Figure 10 shown, it can be determined whether there is a peak in the [50%, 100%] range of the exponential difference sequence. If there is a peak in the [50%, 100%] range, it can start searching backward from the position corresponding to 50%, and then the peaks that appear are sequentially determined as Th1, Th2, Th3, Th4, Th5, etc. When there is no longer a peak or the number of peaks that appear reaches a preset value, for example, when 5 peaks are searched, it can be determined to end this threshold determination. When there is no peak in the [50%, 100%] range of the exponential difference sequence, the threshold Thd can be determined as the noise coefficient corresponding to the peak with the largest amplitude in the last 5% range.

[0072] Assume there are 5 thresholds: Th1, Th2, Th3, Th4, Th5, then:

[0073] If (nosIndex < Thd1), non-noise segment;

[0074] If (nosIndex >= Thd1), level I noise segment;

[0075] If (nosIndex >= Thd2), level II noise segment;

[0076] If (nosIndex >= Thd3), level III noise segment;

[0077] If (nosIndex >= Thd4), level IV noise segment;

[0078] If (nosIndex >= Thd5), level V noise segment.

[0079] In the embodiments of the present application, each noise segment corresponding to a heartbeat can have a corresponding value of the noise feature.

[0080] When processing an electrocardiogram, a user can select a classification method, so that heartbeats can be classified based on noise characteristics and the classification method selected by the user.

[0081] The computer device can determine a classification threshold according to the number of classifications selected by the user, and then quickly classify each heartbeat in the electrocardiogram based on the method in this application and display it to the user.

[0082] Exemplarily, heartbeats can be divided into three categories. The first category can be heartbeats determined to be ventricular premature beats, atrial premature beats, and normal heartbeats; the second category is heartbeats suspected of ventricular premature beats, suspected atrial premature beats, and suspected normal heartbeats, and the third category is artifact heartbeats.

[0083] Such as Figure 11 shown, based on 2 mutation positions, two classification thresholds can be determined, so as to divide heartbeats into: XX heartbeats, suspected XX heartbeats, and artifact heartbeats.

[0084] In the embodiment of this application, when performing cutting, cutting is performed based on the R-wave peak, which is simple and convenient; the noise characteristics and threshold corresponding to each segment are determined using a computational mathematical operation method with a small amount of calculation, thereby reducing the time-consuming of noise detection. Since the time-consuming of noise detection is relatively low, the processing time of the electrocardiogram can be reduced, enabling the electrocardiogram to be processed quickly. In addition, when detecting noise in the electrocardiogram signal in the embodiment of this application, it is based on global sorting analysis of the heartbeat segments in the electrocardiogram and does not require prior knowledge, realizing personalized processing of each electrocardiogram to a certain extent.

[0085] Current electrocardiogram processing software has an inefficient and time-consuming editing process and does not conform to the doctor's usage habits. In traditional processing software, when processing an electrocardiogram, an algorithm is first used for analysis, and then the doctor edits. The doctor cannot edit during the algorithm analysis process. Based on this, a method for processing an electrocardiogram is provided in this embodiment.

[0086] Referring to Figure 12 , a schematic flowchart of steps of another method for processing an electrocardiogram provided by an embodiment of this application is shown, which may specifically include the following steps:

[0087] S1201, divide the electrocardiogram to be processed into multiple heartbeat segments, and each of the heartbeat segments corresponds to a heartbeat in the electrocardiogram.

[0088] S1202, perform data conversion on the electrocardiogram signal in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments.

[0089] S1203. Classify multiple heart beat segments based on the noise characteristics and the classification method selected by the user.

[0090] The method in this embodiment can be used for a user to edit an electrocardiogram using a computer device. When the user is editing, it is possible to determine to interact based on the displayed content. In this embodiment, when processing the electrocardiogram, the first heart beat classification in the electrocardiogram can be recognized. The heart beats in the electrocardiogram are divided into premature ventricular contractions, premature atrial contractions, normal heart beats, unclassified heart beats or artifact heart beats. Compared with the prior art, the differences in the classification of the solution in this application are as shown in Table 1.

[0091]

[0092] Table 1

[0093] When displaying based on the differences in Table 1, the display interface of the solution in this application can be as Figure 13 shown. Among them, the point graph area can be used to display each electrocardiogram segment, and the template area can display multiple templates.

[0094] The user can select a type of heart beat for processing. The above splitting methods can include splitting by direction, splitting by noise characteristics or splitting by amplitude. For example, "(V) Premature Ventricular Contraction" can be automatically split by direction, and the templates used for automatic splitting by direction can include:

[0095] (V) Premature Ventricular Contraction - Template 1: Q wave (negative);

[0096] (V) Premature Ventricular Contraction - Template 2: RS wave (positive and negative, similar amplitude);

[0097] (V) Premature Ventricular Contraction - Template 3: Rs wave (positive and negative, larger positive amplitude);

[0098] (V) Premature Ventricular Contraction - Template 4: R wave (positive);

[0099] (V) Premature Ventricular Contraction - Template 5: small qrs (low amplitude).

[0100] For example, normal heart beats can be split by amplitude; premature atrial contractions can also be split by noise characteristics.

[0101] When displaying heart beats, the heart beats can be displayed in a certain order. In this application, the sorting method of heart beats is extended, as shown in Table 2:

[0102]

[0103] Table 2

[0104] Among them, when sorting by noise, the noise characteristics corresponding to each segment can be obtained by the method in the previous embodiment.

[0105] The user can select a preferred sorting method to facilitate batch editing. For example, the user can select multiple electrocardiogram segments with close sorting in the interface for batch processing.

[0106] S1201 - S1203 in this embodiment is similar to S101 - S103 in the previous embodiment and can be referred to each other, so details are not described here.

[0107] S1204, if the user selects to sort the heartbeat segments of the target category by noise, then display each of the heartbeat segments corresponding to the target category according to the noise characteristics.

[0108] The user can select the sorting method of the heartbeat segments, and the computer device can display the heartbeat segments according to the sorting method selected by the user.

[0109] For example, the user can select to sort according to the noise characteristics, so that the computer device can display the heartbeat segments in ascending order of the noise characteristics. The user can directly perform batch processing on the heartbeats with large noise.

[0110] S1205, when multiple of the heartbeat segments are selected, perform batch processing on the selected multiple heartbeat segments according to the batch processing method selected by the user.

[0111] Multiple heartbeat segments can be displayed on the display interface. The user can select multiple heartbeats for batch processing. For example, a batch of heartbeats with large noise characteristics can be selected and marked as artifact heartbeats.

[0112] For each first heartbeat category, after splitting once according to the template, it can be split again based on the secondary splitting method selected by the user. The secondary splitting method can include automatic splitting and manual splitting. Automatic splitting can be performed based on correlation coefficient, lead time, NNN interval ratio, noise characteristics, amplitude, direction, etc., and can also be performed based on the superposition diagram. Manual splitting can be performed by the user for batch editing or by the user according to the superposition diagram.

[0113] After splitting the electrocardiogram twice, multiple groups can be obtained. The user can choose to use an algorithm for analysis at this time. The analysis of the electrocardiogram can include noise identification, and the noise identification can be realized by the method described in the previous embodiment.

[0114] In the embodiment of the present application, when sorting the ambient noise, the noise characteristics corresponding to each ECG segment can be determined using the solution in the previous embodiment, and the calculation is simple. In this embodiment, the ECG can be analyzed based on an algorithm after being split according to the user's selection, so as to provide the user with more choices and better conform to the user's editing habits.

[0115] It should be noted that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0116] Referring to Figure 14 , a schematic diagram of a processing device for an electrocardiogram provided by an embodiment of the present application is shown, which may specifically include a segment division module 1401, a noise feature determination module 1402, and a classification module 1403, where:

[0117] The division module 1401 is configured to divide the electrocardiogram to be processed into multiple heartbeat segments, and each of the heartbeat segments corresponds to a heartbeat in the electrocardiogram;

[0118] The noise feature determination module 1402 is configured to perform data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments;

[0119] The classification module 1403 is configured to classify multiple heartbeat segments based on the noise characteristics and the classification method selected by the user.

[0120] In a possible implementation manner, the above classification module 1403 includes:

[0121] The classification threshold determination sub-module is configured to determine a classification threshold based on the multiple noise characteristics of the multiple heartbeat segments;

[0122] The classification sub-module is configured to classify multiple heartbeat segments based on the classification threshold.

[0123] In a possible implementation manner, the above classification threshold determination sub-module includes:

[0124] The sorting unit is configured to sort the multiple noise characteristics to obtain a noise characteristic sequence;

[0125] The feature amplification processing unit is configured to perform feature amplification processing on each of the noise characteristics in the noise characteristic sequence to obtain the feature mutation positions of the noise characteristic sequence;

[0126] The classification threshold determination unit is configured to determine the classification threshold according to the feature mutation positions.

[0127] In a possible implementation, the above classification threshold determination unit includes:

[0128] A category quantity determination subunit, configured to determine the quantity of categories to be classified;

[0129] A feature mutation position determination subunit, configured to determine the feature mutation position based on the quantity of categories and the peak value of the noise feature sequence after feature amplification processing.

[0130] In a possible implementation, the above device further includes:

[0131] A display module, configured to, if the user selects to sort the heartbeat segments of the target category by noise, display each of the heartbeat segments corresponding to the target category according to the noise features.

[0132] In a possible implementation, the above device further includes:

[0133] A batch processing module, configured to, when multiple heartbeat segments are selected, perform batch processing on the selected multiple heartbeat segments according to the batch processing method selected by the user.

[0134] In a possible implementation, the above device further includes:

[0135] A target template determination module, configured to determine the target template selected by the user;

[0136] A target heartbeat segment display module, configured to display each target heartbeat segment corresponding to the target template.

[0137] In a possible implementation, the above device further includes:

[0138] A secondary classification threshold determination module, configured to, if the user selects to classify the target heartbeat segments, determine the secondary classification threshold of the target heartbeat segments according to the noise features;

[0139] A secondary classification module, configured to classify each of the target heartbeat segments based on the secondary classification threshold.

[0140] In a possible implementation, the above device further includes:

[0141] A sorting display module, configured to display each of the target heartbeat segments according to the sorting method selected by the user.

[0142] For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For related parts, refer to the description in the method embodiment section.

[0143] Figure 15The structural schematic diagram of a computer device provided by an embodiment of this application. As Figure 15 shown, the computer device 150 of this embodiment includes: at least one processor 1500 ( Figure 15 only one is shown in the figure), a memory 1501, and a computer program 1502 stored in the memory 1501 and executable on the at least one processor 1500. When the processor 1500 executes the computer program 1502, it implements the steps in any of the above method embodiments.

[0144] The computer device 150 may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud computer device. The computer device may include, but is not limited to, a processor 1500 and a memory 1501. Those skilled in the art can understand that Figure 15 merely examples of the computer device 150, which do not constitute a limitation on the computer device 150, and may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0145] An embodiment of this application also provides an electrocardiogram monitoring system, including a data acquisition device, a data processing device, and an interaction device, where:

[0146] The data acquisition device is used to acquire electrocardiograms;

[0147] The data processing device is used to divide the electrocardiogram into multiple cardiac cycle segments, each cardiac cycle segment corresponding to one cardiac cycle in the electrocardiogram; perform data conversion on the electrocardiogram signals in each cardiac cycle segment to obtain the noise characteristics of each cardiac cycle segment;

[0148] The interaction device is used to obtain the processing methods selected by the user for each cardiac cycle segment, and display each cardiac cycle segment based on the processing methods and the noise characteristics.

[0149] Among them, the above data acquisition device may include a wearable device, a patch, or an electrocardiograph, etc., and is used to acquire the electrocardiogram data of the user within a certain period of time, so as to obtain an electrocardiogram;

[0150] The above data processing device may be the processing device for the above electrocardiogram, and is used to perform data processing on the electrocardiogram based on the electrocardiogram processing method selected by the user obtained by the interaction device.

[0151] The above interaction device is used to interact with a user, so as to obtain a user instruction and display an electrocardiogram based on the user instruction. In a possible implementation manner, the interaction device may include a display screen, and the display screen may display each cardiac beat segment. The user may select an electrocardiogram processing method according to each displayed cardiac beat segment. For example, the user may select to perform classification, and then the data processing device may classify each cardiac beat segment according to the noise characteristics and display the classification result on the display screen. The electrocardiogram processing method may also include batch processing. The interaction device may identify the target cardiac beat segments corresponding to the batch processing instruction selected by the user, so as to display the processing result of the target cardiac beat segments. The electrocardiogram processing method may also include sorting. According to the sorting method selected by the user, the interaction device may display each cardiac beat segment in sequence based on the noise characteristics.

[0152] The specific processing method and interaction method in this embodiment may refer to the description in the above method embodiment and will not be elaborated here.

[0153] The so-called processor 1500 may be a central processing unit (CPU). The processor 1500 may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0154] In some embodiments, the memory 1501 may be an internal storage unit of the computer device 150, such as the hard disk or memory of the computer device 150. In other embodiments, the memory 1501 may also be an external storage device of the computer device 150, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 150. Further, the memory 1501 may also include both the internal storage unit and the external storage device of the computer device 150. The memory 1501 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory 1501 may also be used to temporarily store data that has been output or will be output.

[0155] An embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.

[0156] An embodiment of the present application provides a computer program product. When the computer program product runs on a computer device, the computer device is caused to execute the steps in the above-mentioned various method embodiments.

[0157] The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for processing an electrocardiogram, characterized in that, comprising: dividing the electrocardiogram to be processed into a plurality of heartbeat segments, each of the heartbeat segments corresponding to a heartbeat in the electrocardiogram; performing data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise characteristics of each of the heartbeat segments; classifying the plurality of heartbeat segments based on the noise characteristics and the classification method selected by the user.

2. The method according to claim 1, characterized in that, the classifying the plurality of heartbeat segments based on the noise characteristics and the classification method selected by the user comprises: determining a classification threshold based on the plurality of noise characteristics of the plurality of heartbeat segments; classifying the plurality of heartbeat segments based on the classification threshold.

3. The method according to claim 2, characterized in that, the determining a classification threshold based on the plurality of noise characteristics of the plurality of heartbeat segments comprises: sorting the plurality of noise characteristics to obtain a noise characteristic sequence; performing feature amplification processing on each of the noise characteristics in the noise characteristic sequence to obtain the feature mutation positions of the noise characteristic sequence; determining the classification threshold according to the feature mutation positions.

4. The method according to claim 3, characterized in that, the performing feature amplification processing on each of the noise characteristics in the noise characteristic sequence to obtain the feature mutation positions of the noise characteristic sequence comprises: determining the number of categories to be classified; determining the feature mutation positions based on the number of categories and the peak values of the noise characteristic sequence after feature amplification processing.

5. The method according to any one of claims 1-4, characterized in that, the method further comprises: if the user selects to sort the heartbeat segments of the target category by noise, then displaying each of the heartbeat segments corresponding to the target category according to the noise characteristics.

6. The method according to claim 5, characterized in that, the method further comprises: when a plurality of the heartbeat segments are selected, performing batch processing on the selected plurality of heartbeat segments according to the batch processing method selected by the user.

7. The method according to claim 5, characterized in that, the method further comprises: determining the target template selected by the user; displaying each of the target heartbeat segments corresponding to the target template.

8. The method according to claim 7, characterized in that, the method further comprises: if the user selects to classify the target heartbeat segments, then determining a secondary classification threshold for the target heartbeat segments according to the noise characteristics; classifying each of the target heartbeat segments based on the secondary classification threshold.

9. The method according to claim 7, characterized in that, the method further comprises: displaying each of the target heartbeat segments according to the sorting method selected by the user.

10. An electrocardiogram processing device, characterized in that, comprising: a dividing module, configured to divide the electrocardiogram to be processed into a plurality of heartbeat segments, each of the heartbeat segments corresponding to a heartbeat in the electrocardiogram; A noise feature determination module, configured to perform data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise features of each of the heartbeat segments; A classification module, configured to classify a plurality of the heartbeat segments based on the noise features and the classification method selected by the user.

11. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, when the processor executes the computer program, the method according to any one of claims 1-9 is implemented.

12. A computer-readable storage medium storing a computer program, wherein, when the computer program is executed by a processor, the method according to any one of claims 1-9 is implemented.

13. An electrocardiogram monitoring system, wherein, it includes a data acquisition device, a data processing device, and an interaction device, wherein: the data acquisition device is configured to acquire an electrocardiogram; the data processing device is configured to divide the electrocardiogram into a plurality of heartbeat segments, each of the heartbeat segments corresponding to a heartbeat in the electrocardiogram; perform data conversion on the electrocardiogram signals in each of the heartbeat segments to obtain the noise features of each of the heartbeat segments; the interaction device is configured to obtain the electrocardiogram processing method selected by the user for each of the heartbeat segments, and display each of the heartbeat segments based on the electrocardiogram processing method and the noise features.

Images