Three-dimensional electrocardiogram scattergram analysis method, device and medium based on heart rate variability

Abstract

The present application relates to a three-dimensional electrocardiogram (ECG) scatter plot analysis method, device, and medium based on heart rate variability, and relates to the field of biomedical signal processing technology. The present application can map the user's heart rate data into three-dimensional points in a three-dimensional coordinate system after preprocessing, where the coordinate axes of the three-dimensional coordinate system are time domain index, frequency domain index, and nonlinear index, respectively, to display heart rate changes from three dimensions. Compared with traditional two-dimensional scatter plots, the present application can present heart rate change characteristics more comprehensively and three-dimensionally, capture more potential heart disease-related information, and construct a three-dimensional ECG scatter plot containing more heart rate change information to deeply explore the characteristics of heart rate data, quantify the heart rate scatter index, and scientifically and accurately assess the risk level of heart diseases such as arrhythmia and atrial fibrillation based on the scatter index, thereby improving the accuracy and reliability of heart disease risk assessment.

Description

Technical Field

[0001] The present application relates to the field of biomedical signal processing technology, and in particular to a three-dimensional electrocardiogram scatter plot analysis method, device, and medium based on heart rate variability. Background Art

[0002] Heart rate variability (HRV), a key indicator of cardiac autonomic function, has attracted considerable attention in the diagnosis of cardiac diseases. By quantifying subtle variations in heartbeat-to-beat cycles, HRV reveals the dynamic balance between the sympathetic and parasympathetic nervous systems. Disturbances in this balance are often early signs of cardiovascular disease. While traditional two-dimensional heart rate scatter plot analysis methods can intuitively visualize the relationship between adjacent cardiac cycles, they suffer from limitations in information representation and insufficient sensitivity to abnormal data. Technically, a two-dimensional scatter plot maps two adjacent cardiac cycles (NN intervals) onto the horizontal and vertical axes, respectively, and uses the distribution of the scatter plots to identify patterns in heart rate fluctuations. While this method effectively identifies typical ECG patterns such as sinus rhythm and premature atrial beats, it struggles to capture the multidimensional characteristics of complex arrhythmias. For example, in the early stages of paroxysmal atrial fibrillation, a two-dimensional scatter plot may only reveal slight dispersion in the distribution of the points, failing to distinguish physiological fluctuations from pathological abnormalities. Clinical studies have shown that two-dimensional analysis has a high miss rate of up to 28% for latent conduction block and a sensitivity of less than 60% for early warning of long QT syndrome. The root of its limitations lies in the fact that two-dimensional graphs can only reflect the linear relationship between two cardiac cycles, making it difficult to fully capture complex heart rate patterns. When a patient presents with premature ventricular bigeminy combined with a wandering sinus rhythm, the two-dimensional scatter plot can cause feature confusion due to data overlap. Furthermore, traditional methods rely on manual delineation of scatter plot distribution areas and lack the ability to quantify small fluctuations. Two-dimensional analysis methods may not accurately identify millisecond-level HRV abnormalities caused by myocardial ischemia or cryptic rhythm disturbances caused by diabetic autonomic neuropathy, limiting the accuracy of heart disease risk assessment. The convergence of deep learning and medical imaging technologies is placing higher demands on the spatial and temporal resolution of HRV analysis in the medical field. In Holter monitoring, a single 24-hour recording can generate over 100,000 cardiac cycle data sets. Traditional two-dimensional methods are unable to accurately analyze these large volumes of data. Therefore, a more advanced analysis method is urgently needed that integrates multi-dimensional ECG information to construct a visualization model with temporal and spatial characteristics, thereby improving the accuracy and reliability of heart disease risk assessment. Summary of the Invention

[0003] Based on this, a three-dimensional electrocardiogram scatter plot analysis method, equipment and medium based on heart rate variability are provided to solve the current technical problem of being unable to quantify the heart rate scatter index and assess the risk level of heart diseases such as arrhythmia and atrial fibrillation, resulting in low accuracy and poor reliability in heart disease risk assessment.

[0004] In one aspect, a method for analyzing a three-dimensional electrocardiogram scatter plot based on heart rate variability is provided, the method comprising:

[0005] Collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data;

[0006] The coordinate axes of the three-dimensional coordinate system are defined as a time domain index, a frequency domain index, and a nonlinear index, respectively. The time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system, and a scatter point set is formed by the three-dimensional points.

[0007] Drawing a three-dimensional electrocardiogram scatter plot according to the scatter point set;

[0008] Analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points;

[0009] Calculating a scatter index based on the outliers and the clustering points;

[0010] The user's heart health risk level is determined based on the scatter index.

[0011] In one embodiment, collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data includes:

[0012] Collect the user's heart rate data as RR interval series ,in Represents the total number of data points, and the sampling frequency is ;

[0013] The RR interval sequence is processed using median filtering, and the calculation formula is: , Indicates the RR interval as the center, select the total data points, and the median is taken as the result after filtering; where k is the kth value of the RR interval sequence, and m is the number of values ​​of the RR interval sequence taken as half of the sampling frequency;

[0014] The RR interval sequence after median filtering is mapped to interval.

[0015] In one embodiment, the coordinate axes defining the three-dimensional coordinate system are a time domain index, a frequency domain index, and a nonlinear index, respectively. The time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system. Constructing a scatter point set using the three-dimensional points includes:

[0016] A three-dimensional coordinate system is defined with the time domain index as the X-axis, the frequency domain index as the Y-axis, and the nonlinear index as the Z-axis;

[0017] Get the time domain indicators of preprocessed heart rate data , frequency domain indicators and nonlinear indicators , mapping the heart rate data into three-dimensional points in the three-dimensional coordinate system ;

[0018] The three-dimensional points form a scattered point set .

[0019] In one embodiment, the time domain index of the pre-processed heart rate data is obtained , frequency domain indicators and nonlinear indicators include:

[0020] Get the time domain indicators of preprocessed heart rate data is the standard deviation of the NN interval;

[0021] Get the frequency domain indicators of preprocessed heart rate data is the ratio of low frequency power (LF) to high frequency power (HF);

[0022] Obtain nonlinear indicators of preprocessed heart rate data is the approximate entropy.

[0023] In one embodiment, analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points includes:

[0024] Based on the normal heart rate data of healthy people, a three-dimensional confidence ellipse is established;

[0025] The Mahalanobis distance is used to measure the The degree of deviation from the mean of the scattered points in the three-dimensional confidence ellipse , ,in is the scatter mean of healthy people, representing the center position of normal heart rate data, is the covariance matrix, T is the transpose operator;

[0026] Set as the critical value of the chi-square distribution with 3 degrees of freedom and 95% confidence level as a determination threshold, comparing the degree of deviation with the determination threshold;

[0027] When the degree of deviation is greater than the determination threshold, determining the three-dimensional point corresponding to the degree of deviation as an abnormal point;

[0028] The abnormal points are divided into outliers and clusters. The outliers are , the gathering point is , where the parameters , ; median(D) represents the median of the Euclidean distances between all points in the data set, and D is the degree of deviation MinPts is the threshold for determining core points, which represents the data set The minimum number of points that must be included in the neighborhood; In the neighborhood radius Set to 10% of the typical distance.

[0029] In one embodiment, calculating the scatter index based on the outliers and the clusters includes:

[0030] The outlier index OI is set to measure the proportion of the outliers in the total data points: , N is the total data points;

[0031] The aggregation index CI is set to measure the degree of aggregation of the aggregation points: ;

[0032] The scatter index CSI is obtained by fusing the outlier index and the clustering index: ,in and is the weight, and satisfy , weight and By minimizing the value of To confirm, As a clinical diagnostic label, Indicates normal, Indicates an exception.

[0033] In one embodiment, determining the user's heart health risk level based on the scatter index includes:

[0034] Setting different scatter index thresholds, calculating the true positive rate and false positive rate under the corresponding scatter index thresholds, drawing an ROC curve, and selecting the scatter index thresholds that distinguish different risk levels based on the ROC curve to distinguish the heart health risk level;

[0035] The scatter index threshold is set to include a first threshold T1 and a second threshold T2, and the user's heart health risk level is ; CSI is the scatter index, Low is a low risk level, Medium is a medium risk level, and High is a high risk level.

[0036] In one embodiment, the method further comprises:

[0037] Use sliding window calculation to obtain the scatter index sequence value at each time point and construct the scatter index sequence , t represents the time point;

[0038] The scatter index sequence is smoothed by exponentially weighted moving average (EWMA) to obtain a smoothed scatter index sequence. , ;in is the smoothed scatter index value at time point t (target output), is the original scatter index value at time point t (current observation value), is the historical smoothed value at time point t-1 (the state at the previous moment), is the smoothing factor (weight coefficient), which is used to control the weight distribution between the current value and the historical value;

[0039] Determine the scatter index sequence after smoothing Whether it crosses the risk threshold, if so, An early warning is triggered when the risk threshold is crossed three times in a row.

[0040] On the other hand, a computer device is provided, 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 steps of a three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability are implemented.

[0041] On the other hand, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the three-dimensional electrocardiogram scattergram analysis method based on heart rate variability are implemented.

[0042] The above-mentioned three-dimensional ECG scatter plot analysis method, equipment and medium based on heart rate variability, by mapping the user's heart rate data into three-dimensional points in a three-dimensional coordinate system after preprocessing, the coordinate axes of the three-dimensional coordinate system are time domain indicators, frequency domain indicators and nonlinear indicators respectively, and displaying heart rate changes from three dimensions. Compared with traditional two-dimensional scatter plots, it can present heart rate change characteristics more comprehensively and three-dimensionally, capture more potential heart disease-related information, and by constructing a three-dimensional ECG scatter plot containing more heart rate change information, deeply explore the characteristics of heart rate data, quantify the heart rate scatter index, and scientifically and accurately evaluate the risk level of heart diseases such as arrhythmia and atrial fibrillation based on the scatter index, thereby improving the accuracy and reliability of heart disease risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 This is a flow chart of a three-dimensional electrocardiogram scattergram analysis method based on heart rate variability in one embodiment of the present application;

[0045] Figure 2 A schematic diagram of a process for constructing a three-dimensional electrocardiogram scattergram in one embodiment of the present application;

[0046] Figure 3 This is a schematic diagram of a process for calculating a scatter index in one embodiment of the present application;

[0047] Figure 4 This is a flowchart of a dynamic risk update process in one embodiment of the present application;

[0048] Figure 5 This is a diagram of the internal structure of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0050] In one embodiment, Figure 1 As shown, a three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability is provided, comprising the following steps:

[0051] Step S1, collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data;

[0052] Step S2, defining the coordinate axes of a three-dimensional coordinate system as a time domain index, a frequency domain index, and a nonlinear index, obtaining the time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series, and mapping them into three-dimensional points in the three-dimensional coordinate system, and forming a scatter point set through the three-dimensional points;

[0053] Step S3, drawing a three-dimensional electrocardiogram scatter plot based on the scatter point set;

[0054] Step S4, analyzing the three-dimensional points in the three-dimensional ECG scattergram to identify outliers and clusters among the abnormal points;

[0055] Step S5, calculating a scatter index based on the outliers and the cluster points;

[0056] Step S6: determining the user's heart health risk level based on the scatter index.

[0057] Specifically, by mapping the user's heart rate data into three-dimensional points in a three-dimensional coordinate system after preprocessing, the coordinate axes of the three-dimensional coordinate system are time domain indicators, frequency domain indicators and nonlinear indicators respectively, and the heart rate changes are displayed from three dimensions. Compared with the traditional two-dimensional scatter plot, it can present the heart rate change characteristics more comprehensively and three-dimensionally, and capture more potential heart disease-related information. By constructing a three-dimensional electrocardiogram scatter plot containing more heart rate change information, the heart rate data characteristics are deeply explored, the heart rate scatter index is quantified, and the risk level of heart diseases such as arrhythmia and atrial fibrillation is scientifically and accurately evaluated based on the scatter index, thereby improving the accuracy and reliability of heart disease risk assessment.

[0058] In this embodiment, collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data includes:

[0059] Collect the user's heart rate data as RR interval series ,in Represents the total number of data points, and the sampling frequency is ;

[0060] The RR interval sequence is processed using median filtering, and the calculation formula is: , Indicates the RR interval as the center, select the total data points, and the median is taken as the result after filtering; where k is the kth value of the RR interval sequence, and m is the number of values ​​of the RR interval sequence taken as half of the sampling frequency;

[0061] The RR interval sequence after median filtering is mapped to interval.

[0062] Among them, in the actual ECG signal acquisition process, it is inevitable that noise such as baseline drift and motion artifacts will be mixed in. These noises will interfere with the analysis of the real heart rate data, so median filtering is used for processing.

[0063] In order to facilitate the subsequent unified analysis and calculation, the RR interval is mapped to Interval, calculated as: ; This can eliminate the impact of data dimensions and make data from different sources comparable.

[0064] After completing data preprocessing, a three-dimensional coordinate system is defined to characterize the heart rate change characteristics from multiple dimensions.

[0065] like Figure 2 As shown, in this embodiment, the coordinate axes defining the three-dimensional coordinate system are the time domain index, the frequency domain index, and the nonlinear index, respectively. The time domain index, the frequency domain index, and the nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system. The scatter point set formed by the three-dimensional points includes:

[0066] A three-dimensional coordinate system is defined with the time domain index as the X-axis, the frequency domain index as the Y-axis, and the nonlinear index as the Z-axis;

[0067] Get the time domain indicators of preprocessed heart rate data , frequency domain indicators and nonlinear indicators , mapping the heart rate data into three-dimensional points in the three-dimensional coordinate system ;

[0068] The three-dimensional points form a scattered point set .

[0069] In this embodiment, the time domain index of the heart rate data after preprocessing is obtained , frequency domain indicators and nonlinear indicators include:

[0070] Get the time domain indicators of preprocessed heart rate data is the standard deviation of the NN interval;

[0071] Get the frequency domain indicators of preprocessed heart rate data is the ratio of low frequency power (LF) to high frequency power (HF);

[0072] Obtain nonlinear indicators of preprocessed heart rate data is the approximate entropy.

[0073] Specifically, the X-axis (time domain index) is represented by the standard deviation of NN interval (SDNN).

[0074] SDNN reflects the overall discreteness of the RR interval over a period of time, can intuitively reflect the variability of heart rate, and is an important time domain indicator for evaluating the cardiac autonomic nervous system regulation function.

[0075] Y-axis (frequency domain index): is the ratio of low-frequency power (LF) to high-frequency power (HF).

[0076] ;in is the power spectral density estimated by the Welch method. LF and HF are associated with sympathetic and parasympathetic nerve activity, respectively. By calculating their ratio, the balance between the sympathetic and parasympathetic nerves in the cardiac autonomic nervous system was analyzed.

[0077] Z-axis (non-linear index): expressed using approximate entropy (ApEn).

[0078] ;in is the pattern matching probability, , Approximate entropy is used to quantify the complexity and regularity of time series. In heart rate analysis, it reflects the nonlinear dynamic characteristics of cardiac activity and helps to discover potential cardiac dysfunction.

[0079] Finally, each preprocessed RR interval , mapped to a three-dimensional point according to the three-dimensional coordinate system defined above , through the three-dimensional heart rate points together to form a scattered point set , thereby completing the construction of a three-dimensional ECG scatter plot, which comprehensively displays the characteristics of heart rate changes from multiple dimensions.

[0080] In this embodiment, analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points includes:

[0081] Based on the normal heart rate data of healthy people, a three-dimensional confidence ellipse is established;

[0082] The Mahalanobis distance is used to measure the The degree of deviation from the mean of the scattered points in the three-dimensional confidence ellipse , ,in is the scatter mean of healthy people, representing the center position of normal heart rate data, is the covariance matrix, T is the transpose operator;

[0083] Set as the critical value of the chi-square distribution with 3 degrees of freedom and 95% confidence level as a determination threshold, comparing the degree of deviation with the determination threshold;

[0084] When the degree of deviation is greater than the determination threshold, determining the three-dimensional point corresponding to the degree of deviation as an abnormal point;

[0085] The abnormal points are divided into outliers and clusters. The outliers are , the gathering point is , where the parameters , Median (D) represents the median of the Euclidean distances between all points in the data set, and D is the degree of deviation. MinPts is the threshold for determining core points, which represents the data set The minimum number of points that must be included in the neighborhood; In the neighborhood radius Set to 10% of the typical distance.

[0086] The covariance matrix is ​​used to describe the degree of dispersion and correlation of data in various dimensions. The judgment threshold is used to define the boundaries of the normal range. When the critical value is exceeded, it means that the data point deviates from the normal distribution and can be determined as an outlier.

[0087] Outliers refer to data points that deviate significantly from the normal range, reflecting sudden abnormalities in cardiac activity.

[0088] Clusters are identified using the DBSCAN clustering algorithm. The DBSCAN algorithm automatically identifies high-density clustered areas based on the density distribution of data points. These clusters are often associated with specific heart diseases and have important diagnostic value.

[0089] In this embodiment, calculating the scatter index according to the outliers and the cluster points includes:

[0090] The outlier index OI is set to measure the proportion of the outliers in the total data points: , N is the total data points;

[0091] The aggregation index CI is set to measure the degree of aggregation of the aggregation points: ;

[0092] The scatter index CSI is obtained by fusing the outlier index and the clustering index: ,in and is the weight, and satisfy , weight and By minimizing the value of To confirm, As a clinical diagnostic label, Indicates normal, Indicates an exception.

[0093] A larger Outlier Index (OI) indicates a greater number of sudden abnormalities in cardiac activity. The Clustering Index (CI) comprehensively considers the size and density of clusters and assesses the degree of clustering and potential risk of abnormal data related to heart disease.

[0094] Weight and The value of is determined in such a way that the CSI can more accurately reflect the true risk status of the heart.

[0095] like Figure 3 As shown, in this embodiment, determining the user's heart health risk level according to the scatter index includes:

[0096] Setting different scatter index thresholds, calculating the true positive rate and false positive rate under the corresponding scatter index thresholds, drawing an ROC curve, and selecting the scatter index thresholds that distinguish different risk levels based on the ROC curve to distinguish the heart health risk level;

[0097] The scatter index threshold is set to include a first threshold T1 and a second threshold T2, and the user's heart health risk level is ; CSI is the scatter index, Low is a low risk level, Medium is a medium risk level, and High is a high risk level.

[0098] After calculating the scatter index, it is converted into an intuitive cardiac risk level to provide a reference for clinical diagnosis.

[0099] in and Optimization and determination are performed through the ROC curve. The ROC curve comprehensively considers the true positive rate and false positive rate. By analyzing a large amount of clinical data, it finds the threshold that best distinguishes different risk levels, making the risk level classification more scientific and reasonable, and providing a clear basis for doctors to judge the patient's heart health status.

[0100] like Figure 4 As shown, in this embodiment, the method further includes:

[0101] Use sliding window calculation to obtain the scatter index sequence value at each time point and construct the scatter index sequence , t represents the time point;

[0102] The scatter index sequence is smoothed by exponentially weighted moving average (EWMA) to obtain a smoothed scatter index sequence. , ;in is the smoothed scatter index value at time point t (target output), is the original scatter index value at time point t (current observation value), is the historical smoothed value at time point t-1 (the state at the previous moment), is the smoothing factor (weight coefficient), which is used to control the weight distribution between the current value and the historical value;

[0103] Determine the scatter index sequence after smoothing Whether it crosses the risk threshold, if so, An early warning is triggered when the risk threshold is crossed three times in a row.

[0104] like Figure 4 As shown, the sliding window is used to calculate the scatter index sequence value at the i-th time point, that is, t=i. When the scatter index sequence after smoothing is When the risk threshold is not crossed, the sliding window calculation is used to obtain the scatter index sequence value at the i+1th time point for the next round of judgment.

[0105] The above processing method highlights the impact of recent data while taking historical data into consideration, making the CSI sequence smoother and more stable. Crossing the risk threshold three times in a row triggers an early warning, promptly alerting doctors that the patient's heart risk condition has changed so that appropriate diagnostic and treatment measures can be taken.

[0106] In the above-mentioned three-dimensional ECG scatter plot analysis method based on heart rate variability, the user's heart rate data is mapped into three-dimensional points in a three-dimensional coordinate system after preprocessing. The coordinate axes of the three-dimensional coordinate system are time domain indicators, frequency domain indicators and nonlinear indicators respectively, and the heart rate changes are displayed from three dimensions. Compared with traditional two-dimensional scatter plots, it can present the heart rate change characteristics more comprehensively and three-dimensionally, and capture more potential heart disease-related information. By constructing a three-dimensional ECG scatter plot containing more heart rate change information, the heart rate data characteristics are deeply explored, the heart rate scatter index is quantified, and the risk level of heart diseases such as arrhythmia and atrial fibrillation is scientifically and accurately evaluated based on the scatter index, thereby improving the accuracy and reliability of heart disease risk assessment.

[0107] Comprehensive information: By constructing a three-dimensional ECG scatter plot, heart rate changes are displayed from three dimensions: RR interval, ΔRR, and RRV. Compared with traditional two-dimensional scatter plots, it can present the characteristics of heart rate changes more comprehensively and three-dimensionally, and capture more potential heart disease-related information.

[0108] Accurate assessment: Principal component analysis and clustering algorithms are used to determine the normal three-dimensional distribution area, and the heart rate scatter index is calculated in combination with the weight coefficient matrix optimized by machine learning. This improves the accuracy of abnormal data recognition and risk assessment, and reduces the possibility of misdiagnosis and missed diagnosis.

[0109] High clinical value: It provides clinicians with a more scientific and intuitive heart disease risk assessment tool, which helps to detect potential heart disease risks early, formulate personalized diagnosis and treatment plans in a timely manner, and improve patients' treatment effects and quality of life.

[0110] Technological innovation: This method combines three-dimensional graphics analysis with machine learning technology, which is innovative in the field of heart rate variability analysis. It provides new ideas and methods for the development of heart disease diagnosis technology and has broad application prospects and promotion value.

[0111] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store three-dimensional electrocardiogram scatter plot analysis data based on heart rate variability. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability is implemented.

[0112] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0113] In one embodiment, a computer device is provided, 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 following steps are performed:

[0114] Collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data;

[0115] The coordinate axes of the three-dimensional coordinate system are defined as a time domain index, a frequency domain index, and a nonlinear index, respectively. The time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system, and a scatter point set is formed by the three-dimensional points.

[0116] Drawing a three-dimensional electrocardiogram scatter plot according to the scatter point set;

[0117] Analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points;

[0118] Calculating a scatter index based on the outliers and the clustering points;

[0119] The user's heart health risk level is determined based on the scatter index.

[0120] For specific limitations on the steps implemented when the processor executes the computer program, please refer to the above limitations on the method for three-dimensional electrocardiogram analysis based on heart rate variability, which will not be repeated here.

[0121] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0122] Collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data;

[0123] The coordinate axes of the three-dimensional coordinate system are defined as a time domain index, a frequency domain index, and a nonlinear index, respectively. The time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system, and a scatter point set is formed by the three-dimensional points.

[0124] Drawing a three-dimensional electrocardiogram scatter plot according to the scatter point set;

[0125] Analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points;

[0126] Calculating a scatter index based on the outliers and the clustering points;

[0127] The user's heart health risk level is determined based on the scatter index.

[0128] For specific limitations on the steps implemented when the computer program is executed by the processor, please refer to the above limitations on the method for three-dimensional electrocardiogram analysis based on heart rate variability, which will not be repeated here.

[0129] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0130] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0131] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and such modifications and improvements are intended to fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability, characterized in that: include: Collecting the user's heart rate data and preprocessing it to obtain the preprocessed heart rate data; The coordinate axes of the three-dimensional coordinate system are defined as a time domain index, a frequency domain index, and a nonlinear index, respectively. The time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system, and a scatter point set is formed by the three-dimensional points. Drawing a three-dimensional electrocardiogram scatter plot according to the scatter point set; Analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points; Calculating a scatter index based on the outliers and the clustering points; determining a user's heart health risk level based on the scatter index; The step of analyzing the three-dimensional points in the three-dimensional electrocardiogram scattergram to identify outliers and clusters among the abnormal points includes: Based on the normal heart rate data of healthy people, a three-dimensional confidence ellipse is established; The Mahalanobis distance is used to measure the The degree of deviation from the mean of the scattered points in the three-dimensional confidence ellipse , ,in is the scatter mean of healthy people, representing the center position of normal heart rate data, is the covariance matrix, T is the transpose operator; Set as the critical value of the chi-square distribution with 3 degrees of freedom and 95% confidence level as a determination threshold, comparing the degree of deviation with the determination threshold; When the degree of deviation is greater than the determination threshold, determining the three-dimensional point corresponding to the degree of deviation as an abnormal point; The abnormal points are divided into outliers and clusters. The outliers are , the gathering point is , where the parameters , ; median(D) represents the median of the Euclidean distances between all points in the data set, and D is the degree of deviation MinPts is the threshold for determining core points, which represents the data set The minimum number of points that must be included in the neighborhood.

2. The three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability according to claim 1, characterized in that: The collecting and preprocessing of the user's heart rate data to obtain the preprocessed heart rate data includes: Collect the user's heart rate data as RR interval series ,in Represents the total number of data points, and the sampling frequency is ; The RR interval sequence is processed using median filtering, and the calculation formula is: , Indicates the RR interval as the center, select the total data points, and the median is taken as the result after filtering; where k is the kth value of the RR interval sequence, and m is the number of values ​​of the RR interval sequence taken as half of the sampling frequency; The RR interval sequence after median filtering is mapped to interval.

3. The three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability according to claim 1, characterized in that: The coordinate axes of the three-dimensional coordinate system are respectively a time domain index, a frequency domain index, and a nonlinear index. The time domain index, frequency domain index, and nonlinear index of the preprocessed heart rate data in each time series are obtained and mapped into three-dimensional points in the three-dimensional coordinate system. The scatter point set formed by the three-dimensional points includes: A three-dimensional coordinate system is defined with the time domain index as the X-axis, the frequency domain index as the Y-axis, and the nonlinear index as the Z-axis; Get the time domain indicators of preprocessed heart rate data , frequency domain indicators and nonlinear indicators , mapping the heart rate data into three-dimensional points in the three-dimensional coordinate system ; The three-dimensional points form a scattered point set .

4. The three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability according to claim 3, characterized in that: The time domain index of the pre-processed heart rate data is obtained , frequency domain indicators and nonlinear indicators include: Get the time domain indicators of preprocessed heart rate data is the standard deviation of the NN interval; Get the frequency domain indicators of preprocessed heart rate data is the ratio of low-frequency power to high-frequency power; Obtain nonlinear indicators of preprocessed heart rate data is the approximate entropy.

5. The three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability according to claim 1, characterized in that: Calculating the scatter index according to the outliers and the cluster points includes: The outlier index OI is set to measure the proportion of the outliers in the total data points: , N is the total data points; The aggregation index CI is set to measure the degree of aggregation of the aggregation points: ; The scatter index CSI is obtained by fusing the outlier index and the clustering index: ,in and is the weight, and satisfy , weight and By minimizing the value of To confirm, As a clinical diagnostic label, Indicates normal, Indicates an exception.

6. The three-dimensional electrocardiogram scattergram analysis method based on heart rate variability according to claim 1, characterized in that: Determining the user's heart health risk level according to the scatter index includes: Setting different scatter index thresholds, calculating the true positive rate and false positive rate under the corresponding scatter index thresholds, drawing an ROC curve, and selecting the scatter index thresholds that distinguish different risk levels based on the ROC curve to distinguish the heart health risk level; The scatter index threshold is set to include a first threshold T1 and a second threshold T2, and the user's heart health risk level is ; CSI is the scatter index, Low is a low risk level, Medium is a medium risk level, and High is a high risk level.

7. The three-dimensional electrocardiogram scatter plot analysis method based on heart rate variability according to claim 1, characterized in that: The method further comprises: Use sliding window calculation to obtain the scatter index sequence value at each time point and construct the scatter index sequence , t represents the time point; The scatter index sequence is smoothed by exponentially weighted moving average to obtain a smoothed scatter index sequence. , ;in is the smoothed scatter index value at time point t, is the original scatter index value at time point t, is the historical smoothed value at time point t-1, is a smoothing factor used to control the weight distribution between current value and historical value; Determine the scatter index sequence after smoothing Whether it crosses the risk threshold, if so, An early warning is triggered when the risk threshold is crossed three times in a row.

8. 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 steps of the method according to any one of claims 1 to 7 are implemented.

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

Images