Typical heartbeat recognition method, device, computer equipment and storage medium

Abstract

The application relates to a typical heartbeat recognition method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring heartbeat data of a target object; clustering each heartbeat data according to at least two heartbeat description features corresponding to the heartbeat data to obtain heartbeat clustering results corresponding to each heartbeat description feature; and screening out target typical heartbeat data satisfying a preset typicality condition in the heartbeat clustering results. The method can more comprehensively screen out typical heartbeat data which can represent the overall heartbeat data of the target object, and efficiently recognize the typical heartbeat of the target object in the heartbeat data of the target object.

Description

Technical Field

[0001] This application relates to the field of electrocardiogram signal technology, and in particular to a typical heartbeat recognition method, device, computer equipment, storage medium, and computer program product. Background Technology

[0002] Electrocardiogram (ECG) signals comprehensively reflect the electrical activity of the heart and play a crucial role in the diagnosis of heart diseases and the assessment of cardiac function. Heartbeats are the fundamental component of an ECG signal. A normal ECG signal consists of several heartbeats, each heartbeat (a complete cardiac cycle) including a sequence of P waves, QRS complexes, T waves, and U waves. The accuracy of heartbeat classification is of great significance for the diagnosis of heart diseases and the assessment of cardiac function.

[0003] In related technologies, automated heartbeat recognition algorithms can effectively identify heartbeat types. However, due to significant individual differences, existing heartbeat recognition algorithms can only guarantee overall performance across large datasets. For specific patients, the overall algorithm performance often differs greatly from individual application results. This necessitates acquiring individual electrocardiogram (ECG) data to establish personalized heartbeat classification models for specific patients.

[0004] In related technologies, when acquiring individual electrocardiogram (ECG) data, doctors often only annotate the first 5 minutes of the data to build a personalized heartbeat classification model. However, this method is rigid and lacks specificity. The amount of annotated ECG data from the first 5 minutes is limited, making it impossible to efficiently and comprehensively identify a person's typical heartbeats. Therefore, related technologies suffer from low efficiency in identifying typical heartbeats. Summary of the Invention

[0005] Therefore, it is necessary to provide a typical heartbeat recognition method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve the recognition efficiency of typical heartbeats in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a typical heartbeat recognition method. The method includes:

[0007] Obtain the heartbeat data of the target object;

[0008] Based on at least two heartbeat description features corresponding to the heartbeat data, cluster each heartbeat data to obtain heartbeat clustering results corresponding to each heartbeat description feature;

[0009] From the heartbeat clustering results, target typical heartbeat data that meet the preset typicality conditions are selected.

[0010] In one embodiment, the step of filtering out target typical heartbeat data that meet preset typicality conditions from the heartbeat clustering results includes:

[0011] In the heartbeat clustering results corresponding to each of the heartbeat descriptive features, heartbeat data that meet the preset candidate conditions are selected to obtain a candidate heartbeat dataset corresponding to each of the heartbeat descriptive features.

[0012] The candidate heartbeat datasets corresponding to each of the heartbeat description features are deduplicated to obtain the target typical heartbeat data.

[0013] In one embodiment, the step of deduplicating the candidate heartbeat datasets corresponding to each of the heartbeat description features to obtain the target typical heartbeat data includes:

[0014] The candidate heartbeat datasets corresponding to each of the heartbeat description features are deduplicated to obtain the typical heartbeat datasets after deduplication.

[0015] Determine the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset; the typical value is determined based on the correlation between the corresponding typical heartbeat data and other typical heartbeat data; the other typical heartbeat data are the typical heartbeat data in the typical heartbeat dataset other than the corresponding typical heartbeat data.

[0016] Typical heartbeat data with corresponding typical values ​​greater than a preset typical value threshold are selected from the typical heartbeat dataset and used as the target typical heartbeat data.

[0017] In one embodiment, the method further includes:

[0018] Typical heartbeat data whose typical values ​​in the typical heartbeat dataset are less than the preset typical value threshold are used as backup typical heartbeat data.

[0019] In response to a cardiac beat data replenishment request, data that meets the cardiac beat data replenishment request is selected from the backup typical cardiac beat data and added to the target typical cardiac beat data.

[0020] In one embodiment, determining the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset includes:

[0021] For any typical heartbeat data in the typical heartbeat dataset, determine the correlation between the typical heartbeat data and each other typical heartbeat data, and obtain the correlations corresponding to the typical heartbeat data.

[0022] The typical value corresponding to any typical heartbeat data is obtained by determining the reciprocal of the sum of all correlations corresponding to any typical heartbeat data.

[0023] In one embodiment, the algorithm used to cluster the heartbeat data is a density clustering algorithm; each heartbeat clustering result includes several data clusters and several noise points; the data clusters include several core points and several boundary points; the step of filtering heartbeat data that meet preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat descriptive feature to obtain a candidate heartbeat dataset corresponding to each heartbeat descriptive feature includes:

[0024] Noise points that meet the first preset candidate number are selected from several noise points in each heartbeat clustering result, and core points that meet the second preset candidate number and boundary points that meet the third preset candidate number are selected from each data cluster corresponding to each heartbeat clustering result.

[0025] Based on the noise points, core points, and boundary points selected for each heartbeat clustering result, a candidate heartbeat dataset corresponding to each heartbeat descriptive feature is determined.

[0026] In one embodiment, the method further includes:

[0027] Determine a one-dimensional array to describe the waveform morphology of the heartbeat data, and obtain the waveform morphology features corresponding to the heartbeat data;

[0028] Determine the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data to obtain the heartbeat interval characteristics corresponding to the heartbeat data;

[0029] The at least two heartbeat descriptive features are determined based on the waveform morphology features and the heartbeat interval features.

[0030] Secondly, this application also provides a typical heartbeat recognition device. The device includes:

[0031] The acquisition module is used to acquire the heartbeat data of the target object;

[0032] The clustering module is used to cluster each of the heartbeat data according to at least two heartbeat description features corresponding to the heartbeat data, and obtain heartbeat clustering results corresponding to each heartbeat description feature;

[0033] The filtering module is used to filter out target typical heartbeat data that meet preset typicality conditions from the heartbeat clustering results.

[0034] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0035] Obtain the heartbeat data of the target object;

[0036] Based on at least two heartbeat description features corresponding to the heartbeat data, cluster each heartbeat data to obtain heartbeat clustering results corresponding to each heartbeat description feature;

[0037] From the heartbeat clustering results, target typical heartbeat data that meet the preset typicality conditions are selected.

[0038] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0039] Obtain the heartbeat data of the target object;

[0040] Based on at least two heartbeat description features corresponding to the heartbeat data, cluster each heartbeat data to obtain heartbeat clustering results corresponding to each heartbeat description feature;

[0041] From the heartbeat clustering results, target typical heartbeat data that meet the preset typicality conditions are selected.

[0042] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0043] Obtain the heartbeat data of the target object;

[0044] Based on at least two heartbeat description features corresponding to the heartbeat data, cluster each heartbeat data to obtain heartbeat clustering results corresponding to each heartbeat description feature;

[0045] From the heartbeat clustering results, target typical heartbeat data that meet the preset typicality conditions are selected.

[0046] The aforementioned typical heartbeat identification method, apparatus, computer equipment, storage medium, and computer program product acquire heartbeat data of the target object; based on at least two heartbeat descriptive features corresponding to the heartbeat data, they cluster each heartbeat data from at least two aspects to obtain heartbeat clustering results corresponding to each heartbeat descriptive feature. Thus, the heartbeat descriptive features referenced when clustering heartbeat data are more comprehensive, enabling the rapid and accurate selection of target typical heartbeat data that meets preset typicality conditions and represents different heartbeat data types corresponding to different heartbeat clustering results from a large amount of heartbeat data clustering results. This makes the selected target typical heartbeat data more representative of the overall heartbeat data of the target object, more comprehensively including typical heartbeat data in the overall heartbeat data of the target object, and achieving efficient identification of typical heartbeats of the target object from its heartbeat data. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating a typical heartbeat recognition method in one embodiment;

[0048] Figure 2 This is a schematic diagram of heartbeat data to be clustered in one embodiment;

[0049] Figure 3 This is a schematic diagram illustrating the principle of a density clustering algorithm in one embodiment;

[0050] Figure 4 This is a flowchart illustrating a typical heartbeat recognition method in another embodiment;

[0051] Figure 5 This is a flowchart illustrating a typical heartbeat presentation method in one embodiment;

[0052] Figure 6 This is a scatter plot of RR intervals for typical heartbeat data in one embodiment;

[0053] Figure 7 This is a structural block diagram of a typical heartbeat recognition device in one embodiment;

[0054] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0057] In one embodiment, such as Figure 1 As shown, a typical heartbeat recognition method is provided. This embodiment illustrates the application of this method to a computer device. It is understood that the computer device can be a terminal or a server. The server can be a standalone server or a server cluster composed of multiple servers. Furthermore, the method can also be applied to systems including terminals and servers, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0058] Step S110: Obtain the heartbeat data of the target object.

[0059] In practice, computer equipment can acquire electrocardiogram (ECG) signal data of the target object within a preset time period. By preprocessing the ECG signal data, the heart rate data of the target object to be clustered can be obtained.

[0060] Specifically, computer equipment can first standardize the data format of electrocardiogram (ECG) signal data. For example, computer equipment can use a cubic spline sampling algorithm to convert ECG signal data with different sampling frequencies into ECG signal data with the same sampling frequency (e.g., 250 Hz).

[0061] Then, the computer equipment can filter the ECG signal data. For example, the computer equipment can use a Butterworth bandpass filter (0.05Hz to 100Hz) to filter the ECG signal data, obtaining filtered ECG signal data.

[0062] Then, the computer device can use the PT algorithm (Pan-Tompkins detection algorithm) to locate the R wave position in the ECG signal data, so as to annotate the heartbeat data in the ECG signal data and obtain the heartbeat annotation information corresponding to the ECG signal data.

[0063] Then, the computer can perform data cleaning on the heartbeat annotation information corresponding to the ECG signal data, removing abnormal heartbeat annotation information to obtain cleaned ECG signal data. For example, if there are multiple heartbeat annotations for the same heartbeat data in the ECG signal data, the duplicate heartbeat annotations need to be removed; some annotation positions contain explanatory information, which needs to be deleted.

[0064] Then, the computer equipment can segment the cleaned ECG signal data to obtain the heartbeat data of the target group to be clustered. Specifically, the computer equipment uses each R-wave position in the ECG signal data as a reference point and extracts ECG signal data forward and backward with a window size of a preset time length to obtain the heartbeat data of the target group corresponding to each R-wave position.

[0065] In practical applications, computer equipment can segment the ECG signal data into intervals of 0.25 seconds before and 0.45 seconds after each R-wave position, resulting in multiple heartbeat data sets to be clustered. The preset time length can also be other values; no specific restrictions are imposed here.

[0066] For the ease of understanding of those skilled in the art, Figure 2 This paper presents a schematic diagram of segmenting electrocardiogram (ECG) signal data to obtain heartbeat data to be clustered.

[0067] Step S120: Based on at least two cardiac beat description features corresponding to the cardiac beat data, cluster each cardiac beat data to obtain the cardiac beat clustering result corresponding to each cardiac beat description feature.

[0068] In practical applications, heartbeat description features can also be named heartbeat description data.

[0069] The heartbeat clustering result corresponding to each heartbeat description feature includes multiple heartbeat data clusters.

[0070] In practice, computer equipment can cluster each heartbeat data from at least two aspects according to a preset clustering algorithm based on at least two heartbeat descriptive features corresponding to the heartbeat data, so as to obtain the heartbeat clustering results corresponding to each heartbeat descriptive feature.

[0071] Step S130: In the heartbeat clustering results, select the target typical heartbeat data that meets the preset typicality conditions.

[0072] In practice, the computer device can filter out target typical heartbeat data that meets the preset typicality conditions from the heartbeat clustering results corresponding to the heartbeat description features, so that the target typical heartbeat data can more comprehensively represent the overall heartbeat data of the target object.

[0073] In this way, the computer device can control the display device to present the target typical heartbeat data of the target object, so that the user can label the heartbeat type of the target typical heartbeat data and obtain the labeled target typical heartbeat data. The labeled target typical heartbeat data can be used to build a heartbeat classification model for the target object.

[0074] In the aforementioned typical heartbeat recognition method, heartbeat data of the target object is acquired; based on at least two heartbeat descriptive features corresponding to the heartbeat data, each heartbeat data is clustered from at least two aspects to obtain heartbeat clustering results corresponding to each heartbeat descriptive feature. In this way, the heartbeat descriptive features referenced when clustering heartbeat data are more comprehensive, thereby enabling the rapid and accurate selection of target typical heartbeat data that meets preset typicality conditions and can represent different heartbeat data types corresponding to different heartbeat clustering results from the heartbeat clustering results of a large amount of heartbeat data. This makes the selected target typical heartbeat data more representative of the overall heartbeat data of the target object, and more comprehensively includes the typical heartbeat data in the overall heartbeat data of the target object, thus achieving efficient identification of the typical heartbeats of the target object from the heartbeat data of the target object.

[0075] In one embodiment, selecting target typical heartbeat data that meets preset typicality conditions from the heartbeat clustering results includes: selecting heartbeat data that meets preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat descriptive feature to obtain a candidate heartbeat dataset corresponding to each heartbeat descriptive feature; performing deduplication processing on the candidate heartbeat datasets corresponding to each heartbeat descriptive feature to obtain a typical heartbeat dataset; and selecting typical heartbeat data that meets the preset typicality conditions from the typical heartbeat dataset to obtain target typical heartbeat data.

[0076] In the specific implementation, during the process of selecting target typical heartbeat data that meets the preset typicality conditions from the heartbeat clustering results, the heartbeat clustering results corresponding to each heartbeat descriptive feature include multiple heartbeat data clusters. The computer device can select a certain number of heartbeat data from different heartbeat data clusters according to preset candidate conditions in the heartbeat clustering results corresponding to each heartbeat descriptive feature, so as to obtain the candidate heartbeat dataset corresponding to each heartbeat descriptive feature. This makes the heartbeat data in the candidate heartbeat dataset corresponding to each heartbeat descriptive feature representative of the heartbeat data classified into different heartbeat data clusters under that heartbeat descriptive feature. Therefore, the heartbeat data in the candidate heartbeat dataset corresponding to each heartbeat descriptive feature can be used as typical heartbeat data of the target object.

[0077] Since the heartbeat data is clustered according to different heartbeat description features, the candidate heartbeat datasets selected according to different heartbeat description features will contain the same heartbeat data. Therefore, it is necessary to perform duplicate screening on each candidate heartbeat dataset, that is, to deduplicate the candidate heartbeat datasets corresponding to each heartbeat description feature, and to remove redundant and duplicate heartbeat data to obtain a deduplicated candidate heartbeat dataset. This deduplicated candidate heartbeat dataset will not contain the same heartbeat data. Based on the deduplicated candidate heartbeat dataset, the target typical heartbeat data that meets the preset typicality conditions will be determined.

[0078] The technical solution of this embodiment filters out heartbeat data that meets preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat descriptive feature, thus obtaining a candidate heartbeat dataset corresponding to each heartbeat descriptive feature. This makes the heartbeat data in the candidate heartbeat dataset representative of heartbeat data that are classified into different heartbeat data clusters under each heartbeat descriptive feature. Since the heartbeat data is clustered separately according to different heartbeat descriptive features, the same heartbeat data will appear in the candidate heartbeat datasets selected according to different heartbeat descriptive features. By performing deduplication processing on the candidate heartbeat datasets corresponding to each heartbeat descriptive feature, redundant and duplicate typical heartbeat data can be eliminated, and the target typical heartbeat data that can represent the overall heartbeat data of the target object and is less numerous can be accurately selected, making the finally identified typical heartbeats more representative of the overall heartbeat data of the target object.

[0079] In one embodiment, deduplication of the candidate heartbeat datasets corresponding to each heartbeat descriptive feature is performed to obtain target typical heartbeat data. This includes: deduplicating the candidate heartbeat datasets corresponding to each heartbeat descriptive feature to obtain a deduplicated typical heartbeat dataset; determining the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset; the typical value is determined based on the correlation between the corresponding typical heartbeat data and other typical heartbeat data; other typical heartbeat data are the typical heartbeat data in the typical heartbeat dataset other than the corresponding typical heartbeat data; and selecting typical heartbeat data in the typical heartbeat dataset whose corresponding typical value is greater than a preset typical value threshold as target typical heartbeat data.

[0080] In specific implementation, during the process of deduplicating the candidate heartbeat datasets corresponding to each heartbeat descriptive feature to obtain the target typical heartbeat data, the computer device can perform deduplication on the candidate heartbeat datasets corresponding to each heartbeat descriptive feature. Since the heartbeat data in the candidate heartbeat datasets can serve as the typical heartbeat data of the target object, the deduplicated candidate heartbeat datasets can serve as the deduplicated typical heartbeat datasets of the target object. Furthermore, the deduplicated typical heartbeat datasets do not contain duplicate typical heartbeat data. Specifically, the computer device can construct a typical heartbeat database based on the heartbeat data in the candidate heartbeat datasets corresponding to each heartbeat descriptive feature, and then perform deduplication on the typical heartbeat data in the typical heartbeat database to obtain the deduplicated typical heartbeat dataset.

[0081] To improve the representativeness of the final identified typical heartbeat data presented to the user, ensuring that each identified typical heartbeat data point reflects the characteristics of a specific heartbeat type, the typical heartbeat data in the deduplicated typical heartbeat dataset is representative of the overall heartbeat data of the target object. However, it may contain heartbeat data points with high similarity within the dataset. Typical heartbeat data points in the deduplicated typical heartbeat dataset may exhibit a high degree of repetition of the same type, meaning there are typical heartbeat data points of the same type and highly similar morphology. Typical heartbeat data points with a high degree of repetition of the same type are not directly presented. Therefore, for any typical heartbeat data point in the deduplicated typical heartbeat dataset, the computer device can determine the typical value corresponding to that typical heartbeat data point based on its correlation with other typical heartbeat data points (i.e., typical heartbeat data points in the aforementioned typical heartbeat dataset other than that specific typical heartbeat data point). In this way, the computer device can determine the typical value corresponding to each typical heartbeat data point in the aforementioned deduplicated typical heartbeat dataset. The typical value can then be used to determine the degree of repetition of the corresponding typical heartbeat data point with other typical heartbeat data points in the typical heartbeat dataset.

[0082] Therefore, the computer device can filter out typical heartbeat data with corresponding typical values ​​greater than the preset typical value threshold from the above deduplicated typical heartbeat dataset, and use them as the target typical heartbeat data of the final identified target object. This results in fewer typical heartbeats presented in the end, but more representative ones, which can reduce the number of annotations required by users and improve annotation efficiency.

[0083] The technical solution of this embodiment obtains a deduplicated typical heartbeat dataset by deduplicating the candidate heartbeat datasets corresponding to each heartbeat description feature; it then determines the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset; the typical value is determined based on the correlation between the corresponding typical heartbeat data and other typical heartbeat data; other typical heartbeat data are the typical heartbeat data in the typical heartbeat dataset other than the corresponding typical heartbeat data; thus, the typical value can be used to determine the degree of similarity between the corresponding typical heartbeat data and other typical heartbeat data, and by selecting typical heartbeat data with a typical value greater than a preset typical value threshold from the typical heartbeat dataset as the finally identified target typical heartbeat data, it not only improves the identification efficiency of target typical heartbeat data that can accurately represent the overall heartbeat data of the target object, but also reduces the number of finally identified target typical heartbeat data, and makes it more representative of a large amount of heartbeat data of different heartbeat types of the target object.

[0084] In one embodiment, the method further includes: using typical heartbeat data in the typical heartbeat data set whose corresponding typical values ​​are less than a preset typical value threshold as backup typical heartbeat data; and in response to a heartbeat data supplementation request, selecting data from the backup typical heartbeat data that meets the heartbeat data supplementation request and supplementing it into the target typical heartbeat data.

[0085] In practice, the computer device can also use typical heartbeat data whose typical values ​​are less than a preset typical value threshold from the deduplicated typical heartbeat data as backup typical heartbeat data. When the user needs to supplement the presented heartbeat data, the computer device can receive a heartbeat data supplementation request and, in response, select data from the backup typical heartbeat data that meets the heartbeat data supplementation request to supplement the target typical heartbeat data for presentation to the user.

[0086] Specifically, when the target typical cardiac beat data currently presented to the user cannot meet the user's annotation needs—for example, when the user cannot determine the cardiac beat type of the target typical cardiac beat data—the user needs to supplement the presented target typical cardiac beat data. Therefore, the computer device needs to select data from the aforementioned backup typical cardiac beat data that meets the cardiac beat data supplementation request and add it to the target typical cardiac beat data.

[0087] In practical applications, the cardiac data supplementation request can carry a supplementation quantity. During the process of selecting data to supplement the target typical cardiac data that meets the supplementation request, the computer device can select spare typical cardiac data that meets the aforementioned supplementation quantity according to the typical values ​​corresponding to each spare typical cardiac data set. The typical value corresponding to the selected spare typical cardiac data set is higher than the typical value corresponding to the unselected spare typical cardiac data set.

[0088] Furthermore, in some embodiments, before the computer device receives the request for supplementary cardiac data, during the presentation of target typical cardiac data, the computer device may sort the target typical cardiac data from high to low according to the typical value corresponding to each target typical cardiac data, and present the sorted target typical cardiac data to the user.

[0089] In practical applications, before the computer device receives the request for supplementary cardiac data, during the process of presenting the target typical cardiac data, the computer device can also respond to a cardiac presentation request, which carries a presentation quantity. The computer device can select target typical cardiac data that meets the presentation quantity to present to the user. Alternatively, the computer device can also select target typical cardiac data that meets the presentation quantity from the sorted target typical cardiac data to present to the user.

[0090] Before the computer device receives the request for supplementary cardiac data, during the process of presenting the target typical cardiac data, if the number of target typical cardiac data selected by the computer device is less than the number to be presented, the computer device can select backup typical cardiac data to add to the target typical cardiac data so that the number of target typical cardiac data presented in the end meets the number to be presented.

[0091] The technical solution of this embodiment uses typical heartbeat data whose corresponding typical values ​​in the typical heartbeat data set are less than a preset typical value threshold as backup typical heartbeat data; and in response to a heartbeat data supplementation request, selects data from the backup typical heartbeat data that meets the heartbeat data supplementation request to supplement the target typical heartbeat data; in this way, typical heartbeat data that meets the user's heartbeat presentation requirements can be selected for display more flexibly.

[0092] In one embodiment, determining the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset includes: for any typical heartbeat data in the typical heartbeat dataset, determining the correlation between any typical heartbeat data and each typical heartbeat data other than any typical heartbeat data, to obtain each correlation corresponding to any typical heartbeat data; determining the reciprocal of the sum of each correlation corresponding to any typical heartbeat data, to obtain the typical value corresponding to any typical heartbeat data.

[0093] In specific implementation, during the process of determining the typical values ​​corresponding to each typical heartbeat data in the deduplicated typical heartbeat dataset, the computer device can determine the correlation between any typical heartbeat data in the typical heartbeat dataset and each typical heartbeat data in the aforementioned typical heartbeat dataset other than that typical heartbeat data, so as to obtain the correlations corresponding to that typical heartbeat data.

[0094] Then, the computer device can determine the reciprocal of the sum of all correlations corresponding to any typical heartbeat data, and use this reciprocal as the typical value corresponding to any typical heartbeat data.

[0095] In practical applications, the process of calculating typical values ​​is shown in the following formula:

[0096]

[0097]

[0098] Where, ρ X,Y This represents the correlation between two typical heartbeat data points, where cov(X,Y) represents the covariance between the two typical heartbeat data points, and σ represents the correlation between them. X and σ y represents the standard deviation of typical heartbeat data, and t represents the typical value.

[0099] In this way, the typical value corresponding to each typical heartbeat data in the deduplicated typical heartbeat dataset can be determined.

[0100] The technical solution of this embodiment determines the correlation between any typical heartbeat data in the typical heartbeat dataset and all other typical heartbeat data, thereby obtaining the correlations corresponding to the typical heartbeat data and determining the similarity between the typical heartbeat data and other typical heartbeat data. By determining the sum of the correlations corresponding to the typical heartbeat data, the degree of repetition of the typical heartbeat data in the typical heartbeat dataset can be determined. Based on the reciprocal of the sum, the typical value corresponding to the typical heartbeat data is determined, such that the typical value corresponding to the typical heartbeat data is inversely proportional to the degree of repetition of the same type. Therefore, based on the typical values ​​corresponding to each typical heartbeat data in the typical heartbeat dataset, more representative target typical heartbeat data can be selected.

[0101] In one embodiment, the algorithm used to cluster the heartbeat data is a density clustering algorithm; each heartbeat clustering result includes several data clusters and several noise points; each data cluster includes several core points and several boundary points; in the heartbeat clustering results corresponding to the heartbeat descriptive features, heartbeat data that meet preset candidate conditions are selected to obtain a candidate heartbeat dataset corresponding to each heartbeat descriptive feature, including: selecting noise points that meet a first preset candidate number from the several noise points in each heartbeat clustering result, selecting core points that meet a second preset candidate number from each data cluster corresponding to each heartbeat clustering result, and selecting boundary points that meet a third preset candidate number; based on the noise points, core points, and boundary points selected for each heartbeat clustering result, the candidate heartbeat dataset corresponding to each heartbeat descriptive feature is determined.

[0102] In practical applications, the data cluster can also be named the heartbeat data cluster.

[0103] In the specific implementation, the algorithm used by the computer device to cluster the heartbeat data of the target object is the density clustering algorithm, so that the heartbeat clustering result corresponding to each heartbeat descriptive feature includes several heartbeat data clusters and several noise points, and each heartbeat data cluster includes several heartbeat data as core points and several heartbeat data as boundary points.

[0104] For the ease of understanding of those skilled in the art, Figure 3 A schematic diagram illustrating the principle of a density-based clustering algorithm is provided. For example... Figure 3 As shown, N represents noise points; B and C represent boundary points; and the remaining point A represents the core point.

[0105] Thus, in the process of the computer device filtering out heartbeat data that meets the preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat descriptive feature, and obtaining the candidate heartbeat dataset corresponding to each heartbeat descriptive feature, the computer device can filter out noise points that meet the first preset candidate number from several noise points in each heartbeat clustering result, and filter out core points that meet the second preset candidate number and boundary points that meet the third preset candidate number from each data cluster corresponding to each heartbeat clustering result.

[0106] In practical applications, to select typical heartbeat data from the heartbeat clustering results corresponding to each heartbeat descriptive feature, the computer device can randomly select x (a first preset number) noise point samples from all noise points in the heartbeat clustering results corresponding to any heartbeat descriptive feature. These noise points are not necessarily noise; they may be outliers and serve as positive heartbeat data samples. For the core points and boundary points in each heartbeat data cluster corresponding to any heartbeat descriptive feature, num_b (a second preset number) and num_c (a third preset number) samples are randomly selected respectively. If the number of core points *p* and boundary points *q* in a given heartbeat data cluster is less than *num_b* and *num_c*, then the actual number *p* and *q* will be used. In this case, *y* samples are selected from each heartbeat data cluster corresponding to any given heartbeat descriptive feature, that is, *yi* samples are selected from the *i*th heartbeat data cluster corresponding to any given heartbeat descriptive feature, and *yi* ≤ *num_b* + *num_c*. Therefore, for any given heartbeat descriptive feature, *N* samples will be selected, where *N* = *x* + *y1* + *y2* + *yi* + ... + *yn*, and *n* represents the total number of heartbeat data clusters corresponding to any given heartbeat descriptive feature. The selected *N* samples will serve as the candidate heartbeat dataset corresponding to any given heartbeat descriptive feature. Here, *i* is greater than or equal to 1 and less than or equal to *n*.

[0107] In this way, the noise points, core points, and boundary points selected from the heartbeat clustering results corresponding to any heartbeat descriptive feature are used as candidate heartbeat datasets for that heartbeat descriptive feature. Similarly, using the above method, a computer device can determine the candidate heartbeat datasets corresponding to each heartbeat descriptive feature based on the noise points, core points, and boundary points selected from the heartbeat clustering results corresponding to each heartbeat descriptive feature.

[0108] The technical solution of this embodiment uses a density clustering algorithm to cluster the heartbeat data. Each heartbeat clustering result includes several data clusters and several noise points. Each data cluster includes several core points and several boundary points. Thus, in the heartbeat clustering result corresponding to each heartbeat descriptive feature, each heartbeat data point is divided into noise points, core points, or boundary points. By selecting noise points that meet a first preset candidate number from the noise points in each heartbeat clustering result, and selecting core points that meet a second preset candidate number and boundary points that meet a third preset candidate number from each data cluster corresponding to each heartbeat clustering result, typical heartbeat data can be selected from the heartbeat clustering result corresponding to each heartbeat descriptive feature. Then, based on the noise points, core points, and boundary points selected for each heartbeat clustering result, candidate heartbeat datasets corresponding to each heartbeat descriptive feature can be determined, making the heartbeat data in the candidate heartbeat datasets corresponding to each heartbeat descriptive feature more representative of heartbeat data classified into different heartbeat data clusters under that heartbeat descriptive feature.

[0109] In one embodiment, the method further includes: determining a one-dimensional array for describing the waveform morphology of the heartbeat data to obtain waveform morphology features corresponding to the heartbeat data; determining the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data to obtain heartbeat interval features corresponding to the heartbeat data; and determining at least two heartbeat description features based on the waveform morphology features and the heartbeat interval features.

[0110] In practical applications, waveform morphology features can be named morphological data, and heartbeat interval features can be named distance data.

[0111] In specific implementation, before the computer device performs clustering on each heartbeat data according to at least two heartbeat descriptive features corresponding to the heartbeat data to obtain the heartbeat clustering results corresponding to each heartbeat descriptive feature, the computer device can also determine a one-dimensional array used to describe the waveform morphology of the heartbeat data to obtain the waveform morphology features corresponding to the heartbeat data; and determine the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data to obtain the heartbeat interval features corresponding to the heartbeat data.

[0112] Specifically, each heartbeat data point can be visualized using a set of one-dimensional discrete data (i.e., a one-dimensional array). By determining the one-dimensional array corresponding to each heartbeat data point and standardizing it, a standardized one-dimensional array is obtained, which serves as the waveform morphology feature corresponding to the heartbeat data. For the R-wave interval information of each heartbeat data point, i.e., the distance between the current heartbeat data point and the preceding and following heartbeat data points (including the distance RRI_pre between the current heartbeat data point and the R-wave of the previous heartbeat data point, and the distance RRI_aft between the current heartbeat data point and the R-wave of the following heartbeat data point), the two distances corresponding to each heartbeat data point are standardized to obtain a standardized distance, which serves as the heartbeat interval feature corresponding to the heartbeat data point.

[0113] The standardization method used can be Z-score standardization, which can be described by the following formula:

[0114]

[0115] Where x represents the waveform morphology or interval characteristics of the heartbeat data, μ is the mean, and σ is the variance.

[0116] In practical applications, other standardized algorithms can also be used, and no specific restrictions are imposed here.

[0117] In this way, computer equipment can determine at least two descriptive features of heartbeats based on the waveform morphology and interval characteristics of the heartbeat data.

[0118] In addition to the waveform morphology features and beat interval features mentioned above, the heartbeat descriptive features can also include other heartbeat descriptive features, such as heartbeat correlation features determined by the correlation between heartbeat data. No specific restrictions are imposed here.

[0119] The technical solution of this embodiment obtains the waveform morphology features corresponding to the heartbeat data by determining a one-dimensional array used to describe the waveform morphology of the heartbeat data; it also obtains the heartbeat interval features corresponding to the heartbeat data by determining the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data; and it determines at least two heartbeat descriptive features based on the waveform morphology features and the heartbeat interval features. Thus, by clustering the heartbeat data of the target object using multiple heartbeat descriptive features, the heartbeat descriptive features referenced when clustering the heartbeat data are more comprehensive, thereby enabling more accurate selection of the overall heartbeat data of the target object and more representative typical target heartbeat data.

[0120] In another embodiment, such as Figure 4 As shown, a typical heartbeat recognition method is provided. Taking the application of this method to a computer device as an example, the method includes the following steps:

[0121] Step S402: Obtain the heartbeat data of the target object.

[0122] Step S404: Determine a one-dimensional array to describe the waveform morphology of the heartbeat data, and obtain the waveform morphology features corresponding to the heartbeat data.

[0123] Step S406: Determine the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data to obtain the heartbeat interval characteristics corresponding to the heartbeat data.

[0124] Step S408: Determine at least two heartbeat descriptive features based on waveform morphology characteristics and heartbeat interval characteristics.

[0125] Step S410: Based on at least two cardiac beat description features corresponding to the cardiac beat data, cluster each cardiac beat data to obtain the cardiac beat clustering result corresponding to each cardiac beat description feature.

[0126] Step S412: Select noise points that meet the first preset candidate number from a number of noise points in each heartbeat clustering result, and select core points that meet the second preset candidate number and boundary points that meet the third preset candidate number from each data cluster corresponding to each heartbeat clustering result.

[0127] Step S414: Based on the noise points, core points, and boundary points selected for each heartbeat clustering result, determine the candidate heartbeat dataset corresponding to the descriptive features of each heartbeat.

[0128] Step S416: Deduplication is performed on the candidate heartbeat datasets corresponding to each heartbeat description feature to obtain the deduplicated typical heartbeat dataset.

[0129] Step S418: Determine the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset.

[0130] Step S420: Select typical heartbeat data with corresponding typical values ​​greater than the preset typical value threshold from the typical heartbeat dataset, and use them as target typical heartbeat data.

[0131] Step S422: Select typical heartbeat data in the typical heartbeat dataset whose corresponding typical value is less than the preset typical value threshold as backup typical heartbeat data.

[0132] Step S424: In response to the cardiac data supplementation request, select data from the spare typical cardiac data that meets the cardiac data supplementation request and supplement it into the target typical cardiac data.

[0133] It should be noted that the specific limitations of the above steps can be found in the specific limitations of a typical heartbeat recognition method described above.

[0134] For the ease of understanding of those skilled in the art, Figure 5 A typical method for presenting heartbeats is provided, and the method is illustrated using a computer device as an example. It includes the following steps:

[0135] Step 1: Clustering of Heartbeat Data.

[0136] Using clustering algorithms, the heartbeat data of the target object is divided into multiple categories of heartbeat data clusters based on waveform morphology features and heartbeat interval features, respectively. Clustering based on waveform morphology features can be named morphological clustering, and clustering based on heartbeat interval features can be named interval clustering. For morphological clustering and interval clustering, corresponding neighborhood radii (eps) and density thresholds (MinPts) are set, respectively. Both morphological and interval clustering will generate multiple data clusters, classifying the heartbeat data into core points, boundary points, and noise points.

[0137] Step 2: Select heart rate data.

[0138] For morphological clustering, the method described in the above embodiments for selecting heartbeat data that meet the preset candidate conditions can be used to select P samples, where P = x + a1 + a2 + ai + ... + ab, b represents the total number of heartbeat data clusters corresponding to the morphological cluster, ai represents the number of samples selected in the i-th heartbeat data cluster corresponding to the morphological cluster, and ai ≤ num_b + num_c, thus obtaining the candidate heartbeat dataset corresponding to the waveform morphological features. Similarly, for interval clustering, M samples can be selected, where M = x + o1 + o2 + oi + ... + oc, c represents the total number of heartbeat data clusters corresponding to the interval cluster, oi represents the number of samples selected in the i-th heartbeat data cluster corresponding to the interval cluster, and oi ≤ num_b + num_c, thus obtaining the candidate heartbeat dataset corresponding to the heartbeat interval features.

[0139] A typical heartbeat database is constructed based on the heartbeat data in the candidate heartbeat dataset corresponding to waveform morphology features and the heartbeat data in the candidate heartbeat dataset corresponding to heartbeat interval features.

[0140] Step 3: Repeat screening.

[0141] Since waveform clustering and feature clustering are performed separately, duplicate heartbeat data may appear in the candidate heartbeat datasets selected by the two clustering methods. Therefore, deduplication processing needs to be performed on the typical heartbeat database to eliminate duplicate heartbeat data, resulting in a deduplicated typical heartbeat dataset.

[0142] Step 4: Typicality screening.

[0143] The computer device can determine the typical value t corresponding to each typical heartbeat data in the deduplicated typical heartbeat dataset according to the method for determining typical values ​​in the above embodiments.

[0144] Step 5: Presentation of Heartbeat Data.

[0145] The typical heartbeat data in the deduplicated typical heartbeat dataset are sorted according to the typical value t. The K typical heartbeat data with the highest typical value (the preset number of presentations) are selected for user annotation and used as the presented heartbeat data. The other typical heartbeat data are used as backup typical heartbeat data. If the K typical heartbeat data cannot meet the annotation requirements, the backup typical heartbeat data can be supplemented into the presented heartbeat data according to the typical value for user annotation.

[0146] For the ease of understanding of those skilled in the art, Figure 6 This provides a scatter plot of RRI (RR interval, the distance between two R waves) for typical heartbeat data after labeling. For example... Figure 6As shown, typical heartbeat data are labeled as Class N (normal heartbeat data) and Class V (ventricular ectopic beats). It should be noted that the typical heartbeat identification method of this application can not only be applied to Class N and Class V, but also to other heartbeat types such as Class S (supraventricular abnormal heartbeat data) and Class F (fusion heartbeats) as needed.

[0147] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0148] Based on the same inventive concept, this application also provides a typical heartbeat recognition device for implementing the typical heartbeat recognition method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more typical heartbeat recognition device embodiments provided below can be found in the limitations of a typical heartbeat recognition method described above, and will not be repeated here.

[0149] In one embodiment, such as Figure 7 As shown, a typical heartbeat recognition device is provided, including: an acquisition module 710, a clustering module 720, and a filtering module 730, wherein:

[0150] The acquisition module 710 is used to acquire the heartbeat data of the target object.

[0151] The clustering module 720 is used to cluster each of the heartbeat data according to at least two heartbeat description features corresponding to the heartbeat data, and obtain heartbeat clustering results corresponding to each heartbeat description feature.

[0152] The filtering module 730 is used to filter out target typical heartbeat data that meet preset typicality conditions from the heartbeat clustering results.

[0153] In one embodiment, the filtering module 730 is specifically used to filter out heartbeat data that meet preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat description feature, to obtain a candidate heartbeat dataset corresponding to each heartbeat description feature; and to perform deduplication processing on the candidate heartbeat datasets corresponding to each heartbeat description feature to obtain the target typical heartbeat data.

[0154] In one embodiment, the filtering module 730 is specifically used to perform deduplication processing on the candidate heartbeat datasets corresponding to each of the heartbeat description features to obtain a deduplicated typical heartbeat dataset; determine the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset; the typical value is determined based on the correlation between the corresponding typical heartbeat data and other typical heartbeat data; the other typical heartbeat data are the typical heartbeat data in the typical heartbeat dataset other than the corresponding typical heartbeat data; and filter out the typical heartbeat data in the typical heartbeat dataset whose corresponding typical value is greater than a preset typical value threshold as the target typical heartbeat data.

[0155] In one embodiment, the device further includes a supplementation module, configured to use typical heartbeat data in the typical heartbeat data set whose corresponding typical values ​​are less than the preset typical value threshold as backup typical heartbeat data; and in response to a heartbeat data supplementation request, to select data from the backup typical heartbeat data that meets the heartbeat data supplementation request and supplement it into the target typical heartbeat data.

[0156] In one embodiment, the filtering module 730 is specifically used to determine the correlation between any typical heartbeat data in the typical heartbeat dataset and each other typical heartbeat data, to obtain each correlation corresponding to the typical heartbeat data; and to determine the reciprocal of the sum of each correlation corresponding to the typical heartbeat data to obtain the typical value corresponding to the typical heartbeat data.

[0157] In one embodiment, the algorithm used to cluster the heartbeat data is a density clustering algorithm; each heartbeat clustering result includes several data clusters and several noise points; the data clusters include several core points and several boundary points; the filtering module 730 is specifically used to filter out noise points that meet a first preset candidate number from the several noise points in each heartbeat clustering result, and to filter out core points that meet a second preset candidate number and boundary points that meet a third preset candidate number from each data cluster corresponding to each heartbeat clustering result; based on the noise points, core points and boundary points filtered for each heartbeat clustering result, a candidate heartbeat dataset corresponding to each heartbeat descriptive feature is determined.

[0158] In one embodiment, the clustering module 720 is further configured to determine a one-dimensional array for describing the waveform morphology of the heartbeat data, thereby obtaining waveform morphology features corresponding to the heartbeat data; determine the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data, thereby obtaining heartbeat interval features corresponding to the heartbeat data; and determine the at least two heartbeat description features based on the waveform morphology features and the heartbeat interval features.

[0159] The modules in the aforementioned typical heartbeat recognition device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0160] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a typical heartbeat recognition method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0161] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0162] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0163] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0164] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0165] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0166] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0167] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0168] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A typical heartbeat recognition method, characterized in that, The method includes: Obtain the heartbeat data of the target object; Based on at least two heartbeat description features corresponding to the heartbeat data, cluster each heartbeat data to obtain heartbeat clustering results corresponding to each heartbeat description feature; In the heartbeat clustering results, target typical heartbeat data that meet preset typicality conditions are selected, including: selecting heartbeat data that meet preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat descriptive feature to obtain candidate heartbeat datasets corresponding to each heartbeat descriptive feature; performing deduplication processing on the candidate heartbeat datasets corresponding to each heartbeat descriptive feature to obtain deduplicated typical heartbeat datasets; determining the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset; the typical value is determined based on the correlation between the corresponding typical heartbeat data and other typical heartbeat data; the other typical heartbeat data... The typical heartbeat data refers to the typical heartbeat data other than the corresponding typical heartbeat data in the typical heartbeat dataset; the typical heartbeat data with typical values ​​greater than a preset typical value threshold in the typical heartbeat dataset are selected as the target typical heartbeat data; including: for any typical heartbeat data in the typical heartbeat dataset, determining the correlation between the any typical heartbeat data and each typical heartbeat data other than the any typical heartbeat data, obtaining each correlation corresponding to the any typical heartbeat data; determining the reciprocal of the sum of each correlation corresponding to the any typical heartbeat data, obtaining the typical value corresponding to the any typical heartbeat data.

2. The method according to claim 1, characterized in that, The method further includes: Typical heartbeat data whose typical values ​​in the typical heartbeat dataset are less than the preset typical value threshold are used as backup typical heartbeat data. In response to a cardiac beat data replenishment request, data that meets the cardiac beat data replenishment request is selected from the backup typical cardiac beat data and added to the target typical cardiac beat data.

3. The method according to claim 1, characterized in that, The algorithm used to cluster the heartbeat data is a density clustering algorithm; each heartbeat clustering result includes several data clusters and several noise points; the data clusters include several core points and several boundary points; the heartbeat data that meets the preset candidate conditions is selected from the heartbeat clustering results corresponding to the heartbeat descriptive features to obtain a candidate heartbeat dataset corresponding to each heartbeat descriptive feature, including: Among the noise points in each heartbeat clustering result, select noise points that meet the first preset number, and select core points that meet the second preset number and boundary points that meet the third preset number from each data cluster corresponding to each heartbeat clustering result. Based on the noise points, core points, and boundary points selected for each heartbeat clustering result, a candidate heartbeat dataset corresponding to each heartbeat descriptive feature is determined.

4. The method according to claim 1, characterized in that, The method further includes: Determine a one-dimensional array to describe the waveform morphology of the heartbeat data, and obtain the waveform morphology features corresponding to the heartbeat data; Determine the time distance between the R wave of each heartbeat data and the R waves of two adjacent heartbeat data to obtain the heartbeat interval characteristics corresponding to the heartbeat data; The at least two heartbeat descriptive features are determined based on the waveform morphology features and the heartbeat interval features.

5. A typical heartbeat recognition device, characterized in that, The device includes: The acquisition module is used to acquire the heartbeat data of the target object; The clustering module is used to cluster each of the heartbeat data according to at least two heartbeat description features corresponding to the heartbeat data, and obtain heartbeat clustering results corresponding to each heartbeat description feature; The filtering module is used to filter out target typical heartbeat data that meet preset typicality conditions from the heartbeat clustering results. The filtering module is specifically used to filter out heartbeat data that meet preset candidate conditions from the heartbeat clustering results corresponding to each heartbeat descriptive feature, to obtain a candidate heartbeat dataset corresponding to each heartbeat descriptive feature; to perform deduplication processing on the candidate heartbeat datasets corresponding to each heartbeat descriptive feature, to obtain a deduplicated typical heartbeat dataset; to determine the typical value corresponding to each typical heartbeat data in the typical heartbeat dataset; the typical value is determined based on the correlation between the corresponding typical heartbeat data and other typical heartbeat data; the other typical heartbeat data are the typical heartbeat data in the typical heartbeat dataset other than the corresponding typical heartbeat data; and to filter out typical heartbeat data in the typical heartbeat dataset whose corresponding typical value is greater than a preset typical value threshold, as the target typical heartbeat data. The filtering module is specifically used to determine the correlation between any typical heartbeat data in the typical heartbeat dataset and each other typical heartbeat data, to obtain each correlation corresponding to the typical heartbeat data; and to determine the reciprocal of the sum of each correlation corresponding to the typical heartbeat data to obtain the typical value corresponding to the typical heartbeat data.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images