Arrhythmia early warning system based on individual heart rate variability analysis

Abstract

The invention relates to the technical field of heart rate monitoring, in particular to an arrhythmia early warning system based on individual heart rate variability analysis, which comprises a feature expression acquisition module, a division module, a heart rate fluctuation rule degree acquisition module, a non-disease possibility acquisition module and a correction module, dividing a heart rate fluctuation time period in the heart rate time sequence into an increasing trend section and a decreasing trend section, determining a main change trend of the heart rate fluctuation time period, and obtaining the heart rate fluctuation rule degree of the heart rate fluctuation time period according to the association between the increasing trend section and the decreasing trend section in combination with the characteristics of the main trend section. The non-disease possibility of the heart rate fluctuation time period is obtained based on the characteristic expression and the heart rate fluctuation rule degree of the heart rate fluctuation time period, and reverse correction is performed on the heart rate variability of the heart rate fluctuation time period according to the non-disease possibility, so that the credibility of the heart rate variability can be improved, and the accuracy of the heart rate variability is ensured; therefore, the accuracy of early warning of arrhythmia of the patient is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of heart rate monitoring, and in particular to an early warning system for arrhythmia based on individual heart rate variability analysis. Background Art

[0002] Heart rate variability, as an important indicator for early detection of arrhythmias, can effectively warn of potential arrhythmias. Early arrhythmias can manifest in various forms of heart rate changes, such as tachycardia, bradycardia, arrhythmia, conduction block, and sinus arrest. Existing technologies use individual heart rate variability to provide early warning of arrhythmias, but heart rate data can fluctuate due to strenuous exercise, emotional fluctuations, and changes in the external environment, making heart rate variability unreliable and reducing the accuracy of early warnings of arrhythmias. Summary of the Invention

[0003] In order to solve the technical problem in existing arrhythmia early warning systems that other influencing factors may interfere with heart rate variability and thus affect the accuracy of heart rate variability, the present invention aims to provide an arrhythmia early warning system based on individual heart rate variability analysis. The technical solutions adopted are as follows: The present invention provides an early warning system for arrhythmia based on individual heart rate variability analysis, comprising: A characteristic performance acquisition module, configured to determine characteristic performance of a heart rate fluctuation period in a heart rate time series sequence based on the instability of the heart rate in the heart rate time series sequence; A division module is used to divide the heart rate fluctuation period into an increasing trend segment and a decreasing trend segment according to the heart rate change trend, and determine the main change trend of the heart rate fluctuation period; a heart rate fluctuation regularity acquisition module, configured to obtain the degree of heart rate fluctuation regularity during the heart rate fluctuation period based on the correlation between the increasing trend segment and the decreasing trend segment and in combination with the characteristics of the main trend segment; the main trend segment being the trend segment corresponding to the main change trend; a non-symptom possibility obtaining module, configured to obtain the non-symptom possibility of the heart rate fluctuation period based on the characteristic manifestations and the regularity of the heart rate fluctuation; The correction module is used to reversely correct the heart rate variability during the heart rate fluctuation period according to the non-symptom possibility.

[0004] In an exemplary embodiment, the heart rate fluctuation regularity acquisition module is specifically configured to: According to the fluctuation of the length of each increasing trend segment and the fluctuation of the length of each decreasing trend segment in the heart rate fluctuation period, the distribution regularity of the heart rate fluctuation period is obtained; A characteristic of heart rate variation during the heart rate fluctuation period is obtained based on the distribution regularity, the lengths of the target increasing trend segment and the target decreasing trend segment, and the first fitting error and the second fitting error; the target increasing trend segment is the longest increasing trend segment among the increasing trend segments during the heart rate fluctuation period, and the target decreasing trend segment is the longest decreasing trend segment among the decreasing trend segments during the heart rate fluctuation period; the first fitting error is the fitting error between the target increasing trend segment and its fitting straight line, and the second fitting error is the fitting error between the target decreasing trend segment and its fitting straight line; The degree of regularity of the heart rate fluctuation is obtained according to the regularity characteristics of the heart rate variation, the number of main trend segments, and the time intervals between adjacent main trend segments.

[0005] In an exemplary embodiment, the process of obtaining the distribution regularity includes: obtaining the distribution regularity based on a first variance and a second variance; the first variance is the variance of the length of each increasing trend segment in the heart rate fluctuation period, and the second variance is the variance of the length of each decreasing trend segment in the heart rate fluctuation period; the distribution regularity is inversely proportional to the first variance and the second variance.

[0006] In an exemplary embodiment, the process of acquiring the heart rate variation regularity characteristics includes: Obtaining a variation error characteristic of the target increasing trend segment according to the length of the target increasing trend segment and the first fitting error; Obtaining a variation error characteristic of the target reduction trend segment according to the length of the target reduction trend segment and a second fitting error; fusing the change error features of the target increasing trend segment and the target decreasing trend segment to obtain a comprehensive change error feature; The distribution regularity and the comprehensive characteristics of the change error are integrated to obtain the heart rate change regularity characteristics.

[0007] In an exemplary embodiment, the process of obtaining the regularity of heart rate fluctuations includes: The degree of regularity of heart rate fluctuation is obtained based on the characteristics of the heart rate change pattern, the proportion of the number of main trend segments, and the average time interval; the degree of regularity of heart rate fluctuation is directly proportional to the characteristics of the heart rate change pattern and the proportion of the number, and inversely proportional to the average time interval; the proportion of the number is obtained by the number of main trend segments and the total number of trend segments in the heart rate fluctuation period; the average time interval is the average value of the time intervals between all two adjacent main trend segments.

[0008] In an exemplary embodiment, the feature performance acquisition module is specifically configured to: Dividing the heart rate time series into a heart rate fluctuation period and a heart rate stable period according to the difference between two adjacent heart rates; Determine the difference in heart rate instability between periods of heart rate fluctuation and adjacent periods of heart rate stability; The difference between the maximum heart rate and the minimum heart rate in the heart rate fluctuation period and the time interval therebetween are determined, and the characteristic performance is obtained in combination with the heart rate instability difference.

[0009] In an exemplary embodiment, the process of obtaining the heart rate instability difference includes: Calculating the average of the absolute values of the differences between all two adjacent heart rates in the target period to obtain the heart rate instability of the target period; the target period is any one of the heart rate fluctuation period and the heart rate stable period; The absolute value of the difference between the heart rate instability of the heart rate fluctuation period and the adjacent heart rate stable period is calculated to obtain the heart rate instability difference.

[0010] In an exemplary embodiment, the process of dividing the increasing trend segment and the decreasing trend segment includes: forming the increasing trend segment with continuous heart rates having an increasing trend, and forming the decreasing trend segment with continuous heart rates having a decreasing trend; The process of obtaining the main change trend includes: obtaining the number of increasing trend segments and the number of decreasing trend segments in the heart rate fluctuation period, and taking the change trend corresponding to the trend segment with the largest number as the main change trend.

[0011] In an exemplary embodiment, the arrhythmia early warning system further includes an arrhythmia risk level acquisition module, configured to: A weighted summation is performed based on the length weights of multiple heart rate fluctuation periods and the target heart rate variability to obtain an arrhythmia risk index; the length weight is obtained from the length of the heart rate fluctuation period; the target heart rate variability is the heart rate variability after reverse correction; The time feature and the arrhythmia risk index are integrated to obtain the arrhythmia risk level; the time feature is obtained by averaging the time intervals between all two adjacent heart rate fluctuation periods.

[0012] In an exemplary embodiment, the arrhythmia early warning system further includes a warning module for: Comparing the arrhythmia risk level with a preset first risk level threshold and a second risk level threshold; wherein the first risk level threshold is greater than the second risk level threshold; If the arrhythmia risk level is greater than the first risk level threshold, determining that the arrhythmia is at high risk; If the arrhythmia risk level is less than or equal to the first risk level threshold and greater than or equal to the second risk level threshold, determining that the arrhythmia is of medium risk; If the arrhythmia risk level is less than the second risk level threshold, the arrhythmia is determined to be of low risk.

[0013] The present invention has the following beneficial effects: the characteristics of the heart rate fluctuation period will have a great impact on the heart rate variability, so the heart rate fluctuation period is analyzed to obtain characteristic performance, and then the heart rate fluctuation period is divided based on the change trend, so as to analyze the correlation between the trend segments, and obtain the degree of heart rate fluctuation regularity in the heart rate fluctuation period, and then according to the characteristic performance and the degree of heart rate fluctuation regularity, obtain the non-symptomatic possibility of the heart rate fluctuation period, and the non-symptomatic possibility characterizes that the cause of the heart rate fluctuation is the possibility of non-symptomatic, which is the impact of other influencing factors on the heart rate, so that the heart rate variability in the heart rate fluctuation period is corrected according to the non-symptomatic possibility, and the credibility of the corrected heart rate variability is greatly improved, ensuring the accuracy of heart rate variability, thereby improving the accuracy of early warning of patients' arrhythmias. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a schematic structural diagram of an arrhythmia early warning system based on individual heart rate variability analysis provided by one embodiment of the present invention; Figure 2 This is a flowchart of steps corresponding to various modules of an arrhythmia early warning system based on individual heart rate variability analysis provided by one embodiment of the present invention; Figure 3 This is a specific implementation flow chart of a feature expression acquisition module provided by an embodiment of the present invention; Figure 4 is a flow chart of obtaining heart rate instability differences provided by one embodiment of the present invention; Figure 5 This is a specific implementation flow chart of a heart rate fluctuation regularity acquisition module provided by an embodiment of the present invention; Figure 6 This is a flow chart for obtaining the regular characteristics of heart rate changes provided by one embodiment of the present invention; Figure 7 This is a specific implementation flow chart of the arrhythmia risk level acquisition module provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] To further illustrate the technical means and effects employed by the present invention to achieve its intended objectives, the following detailed description of the specific embodiments, structures, features, and effects of the present invention is provided in conjunction with the accompanying drawings and preferred embodiments. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this invention pertains. The data and information collected in this application were obtained with full consent and authorization.

[0017] This embodiment provides an arrhythmia early warning system based on individual heart rate variability analysis (hereinafter referred to as the arrhythmia early warning system). Figure 1 As shown, it includes: a feature expression acquisition module, a classification module, a heart rate fluctuation regularity acquisition module, a non-symptom possibility acquisition module, and a correction module. Each module can be a software module, which is essentially a corresponding method step; or it can be a hardware module, in which the executed method steps are configured in the hardware module so that the hardware module realizes the corresponding function. Accordingly, the arrhythmia early warning system can be a software system, configured in a related processor, computer host, and related medical monitoring platform; or it can be a hardware system, such as a server, computer host, etc. This embodiment does not limit the specific configuration of each module and the arrhythmia early warning system.

[0018] The purpose of the arrhythmia early warning system provided in this embodiment is to analyze whether heart rate fluctuations within a patient's heart rate time series are caused by changes in the external environment or the patient's physiological state. Based on the likelihood that these fluctuations are non-symptomatic, the system then corrects heart rate variability during these periods. A heart rate time series is formed by continuously measuring the time intervals between adjacent R-wave peaks (i.e., the RR interval) and arranging them in chronological order. This sequence serves as the core data foundation for analyzing heart rate variability, cardiac rhythm stability, and autonomic nervous system function. The time interval between two adjacent R waves is the RR interval (which directly reflects the heartbeat interval). Heart rate variability refers to the phenomenon in which the intervals between heartbeats vary from beat to beat. The application scenario of the arrhythmia early warning system is that heart rate variability is not only affected by cardiac health but can also fluctuate due to changes in the external environment or physiological state. For example, strenuous exercise, emotional fluctuations, hypoxia, and temperature changes can all cause temporary heart rate fluctuations, which are generally non-symptomatic. These non-symptomatic factors can cause significant fluctuations in heart rate variability in the short term, reducing the accuracy of arrhythmia warnings.

[0019] In an exemplary embodiment, a heart rate timing sequence within a patient's historical time period is obtained, wherein the length of the historical time period is set according to actual needs. In order to ensure the accuracy of arrhythmia warning, the historical time period can be set longer. In this embodiment, the length of the most recent week is used as the historical time period. When conditions permit, the patient can wear a professional heart rate detection device, or wear a smart watch or smart bracelet with a heart rate detection function. The heart rate detection principle of the smart watch or smart bracelet is a prior art and will not be repeated here.

[0020] It should be understood that each heart rate in the heart rate time series is an instantaneous heart rate. To facilitate subsequent processing, each heart rate in the obtained heart rate time series is numerically normalized and dimensionless. In an exemplary embodiment, a maximum-minimum normalization method can be used. Specifically, the maximum and minimum values of each heart rate in the heart rate time series are obtained, and then the maximum-minimum normalization method is used to normalize each heart rate. All heart rates involved in the subsequent steps are normalized heart rates.

[0021] like Figure 2 As shown, the method steps corresponding to each module are as follows: A characteristic performance acquisition module, configured to determine characteristic performance of a heart rate fluctuation period in a heart rate time series sequence based on the instability of the heart rate in the heart rate time series sequence; A division module is used to divide the heart rate fluctuation period into an increasing trend segment and a decreasing trend segment according to the heart rate change trend, and determine the main change trend of the heart rate fluctuation period; A heart rate fluctuation regularity acquisition module is used to obtain the heart rate fluctuation regularity during the heart rate fluctuation period based on the correlation between the increasing trend segment and the decreasing trend segment and the characteristics of the main trend segment; A non-symptom possibility acquisition module is used to obtain the non-symptom possibility of the heart rate fluctuation period based on the characteristic performance and the regularity of the heart rate fluctuation; The correction module is used to reversely correct the heart rate variability during the heart rate fluctuation period according to the possibility of non-symptom.

[0022] The specific implementation process of each module is described below with reference to the accompanying drawings.

[0023] The feature performance acquisition module is used to determine the feature performance of the heart rate fluctuation period in the heart rate time series based on the instability of the heart rate in the heart rate time series.

[0024] Heart rate fluctuations are often related to the regulatory capacity of the autonomic nervous system. Under normal circumstances, heart rate fluctuates within a certain range; this fluctuation is physiological and generally regular. However, sudden increases in heart rate fluctuations may indicate a problem with autonomic nervous system regulation, particularly as it is closely associated with the occurrence of arrhythmias, and therefore warrants special attention. Therefore, based on the instability of heart rate within a heart rate time series, the characteristic manifestations of heart rate fluctuation periods can be determined.

[0025] In an exemplary embodiment, Figure 3 As shown, a specific implementation process of the feature expression acquisition module is given below: Step 1-1: Divide the heart rate time series into a heart rate fluctuation period and a heart rate stable period according to the difference between two adjacent heart rates.

[0026] Since the heart rate time series sequence includes multiple heart rates in the time series, the difference between every two adjacent heart rates is obtained, specifically, the absolute value of the difference between every two adjacent heart rates is calculated. The larger the absolute value of the difference, the more unstable the two adjacent heart rates are. Then, the absolute value of the difference between the two adjacent heart rates is used as the instability of the previous heart rate of the two adjacent heart rates. Taking the i-th heart rate and the i+1-th heart rate as an example, the absolute value of the difference between the i-th heart rate and the i+1-th heart rate is used as the instability of the i-th heart rate. It should be understood that if there is no instability in the last heart rate in the time series, the instability will no longer be obtained and it will not participate in subsequent data processing.

[0027] An instability threshold is preset, which is used to determine whether the instability of each heart rate is high. The value range of the instability threshold is 0-1, and its specific value is set according to actual judgment needs. In this embodiment, 0.3 is used as an example.

[0028] The instability of each heart rate is compared with the instability threshold, and the instability greater than or equal to the instability threshold and the instability less than the instability threshold are obtained.

[0029] The heart rate corresponding to the instability greater than or equal to the instability threshold is taken as the first heart rate, the moments corresponding to each first heart rate are taken as the heart rate fluctuation moments, and the period formed by continuous heart rate fluctuation moments is taken as the heart rate fluctuation period, thereby obtaining several heart rate fluctuation periods.

[0030] The heart rate corresponding to an instability less than the instability threshold is used as the second heart rate, the moments corresponding to each second heart rate are used as heart rate stable moments, and the period consisting of consecutive heart rate stable moments is used as a heart rate stable period, thereby obtaining a number of heart rate stable periods. It should be understood that a heart rate stable period is adjacent to a heart rate fluctuation period.

[0031] Step 1-2: Determine the difference in heart rate instability between a heart rate fluctuation period and an adjacent heart rate stable period.

[0032] Heart rate fluctuation is one of the important manifestations of irregular heartbeat. Stable heart rate may indicate that the patient's physiological state is relatively stable, while fluctuating heart rate may reflect that the patient's physiological state has fluctuated or is abnormal. The greater the difference between the heart rate fluctuation period and the adjacent stable heart rate period, the greater the change in the patient's condition.

[0033] For any heart rate fluctuation period, it has an adjacent heart rate stable period. If the heart rate fluctuation period is at the two ends of the heart rate time series, the heart rate fluctuation period may have only one adjacent heart rate stable period, namely: the left adjacent heart rate stable period or the right adjacent heart rate stable period. If the heart rate fluctuation period is at other positions in the heart rate time series, the heart rate fluctuation period has two adjacent heart rate stable periods, namely: the left adjacent heart rate stable period and the right adjacent heart rate stable period.

[0034] According to the heart rate instability of the heart rate fluctuation period and the adjacent heart rate stable period, the heart rate instability difference between the heart rate fluctuation period and the adjacent heart rate stable period is determined. In an exemplary embodiment, Figure 4 As shown, a specific process of obtaining the heart rate instability difference is given as follows: Step 1-2-1: Calculate the average of the absolute values of the differences between all two adjacent heart rates in the target period to obtain the heart rate instability of the target period.

[0035] For ease of explanation, any one of the heart rate fluctuation periods and the heart rate stable period is set as the target period. The absolute value of the difference between each two adjacent heart rates in the target period (i.e., the instability of each heart rate in the target period) is obtained, and then the average value is calculated. This average value is the heart rate instability of the target period. This results in the heart rate instability of each heart rate fluctuation period and each heart rate stable period.

[0036] Step 1-2-2: Calculate the absolute value of the difference between the heart rate instability of the heart rate fluctuation period and the adjacent heart rate stable period to obtain the heart rate instability difference.

[0037] The heart rate instability during the heart rate fluctuation period and the heart rate instability during the heart rate stable period adjacent to the heart rate fluctuation period are determined, and the absolute value of the difference between the two is calculated. The result obtained is the heart rate instability difference corresponding to the heart rate fluctuation period.

[0038] It should be understood that if there are two adjacent heart rate stable periods on the left and right sides of the heart rate fluctuation period, the heart rate instability difference between the two adjacent heart rate stable periods on the left and right sides should be calculated respectively, and then the average value of these two heart rate instability differences should be obtained as the heart rate instability difference corresponding to the heart rate fluctuation period.

[0039] The greater the difference in heart rate instability, the greater the difference in heart rate instability between the heart rate fluctuation period and the adjacent heart rate stable period, and the more obvious the feature expression, that is, the higher the degree of feature expression.

[0040] Step 1-3: Determine the difference between the maximum heart rate and the minimum heart rate during the heart rate fluctuation period and the time interval between them, and combine the difference in heart rate instability to obtain characteristic performance.

[0041] Get the maximum heart rate and minimum heart rate during the heart rate fluctuation period, and then calculate the difference between the maximum heart rate and the minimum heart rate, which is the range of the heart rate during the heart rate fluctuation period. The larger the range, the greater the fluctuation amplitude of the heart rate during the heart rate fluctuation period, and the more obvious the feature performance, that is, the higher the degree of feature performance.

[0042] The time interval between the maximum heart rate and the minimum heart rate in the heart rate fluctuation period is obtained. The shorter the time interval, the faster the heart rate changes in the heart rate fluctuation period, the more drastic the change, and the more obvious the feature performance, that is, the higher the degree of feature performance.

[0043] According to the difference between the maximum heart rate and the minimum heart rate in the heart rate fluctuation period and the time interval between them, combined with the difference in heart rate instability during the heart rate fluctuation period, the characteristic performance of the heart rate fluctuation period is obtained. Through the above analysis, it can be seen that the characteristic performance of the heart rate fluctuation period is proportional to the difference in heart rate instability during the heart rate fluctuation period, proportional to the difference between the maximum heart rate and the minimum heart rate, and inversely proportional to the time interval between the maximum heart rate and the minimum heart rate. In combination with this logical relationship, a specific quantification method is given as follows: ; in, represents the characteristic performance of the bth heart rate fluctuation period, represents the difference in heart rate instability during the bth heart rate fluctuation period, Indicates the extreme difference of the heart rate in the bth heart rate fluctuation period, represents the time interval between the maximum heart rate and the minimum heart rate in the bth heart rate fluctuation period, represents the exponential function with the natural constant e as the base, Used for Perform negative correlation normalization.

[0044] Using the above process, we can obtain the characteristic characteristics of each heart rate fluctuation period. Drastic heart rate fluctuations are often a precursor or manifestation of arrhythmia. If the heart rate fluctuation period differs significantly from the stable heart rate period, it means that the patient's heart rate regularity has changed significantly, which may be an early sign of arrhythmia.

[0045] The division module is used to divide the heart rate fluctuation period into an increasing trend segment and a decreasing trend segment according to the heart rate change trend, and determine the main change trend of the heart rate fluctuation period.

[0046] During periods of heart rate fluctuation not caused by illness, heart rate changes often exhibit a certain regularity. Unlike arrhythmias caused by illness, heart rate fluctuations follow a specific pattern rather than being erratic. For example, during exercise, as the intensity gradually increases, the heart rate gradually rises, and this upward trend is relatively stable. During periods of emotional excitement, the heart rate rises rapidly, but as the emotions gradually calm, it gradually decreases, and this fluctuation is somewhat predictable. Therefore, heart rate fluctuation periods can be divided into increasing and decreasing trend segments based on the heart rate change trend. An increasing trend segment indicates a gradual increase in heart rate over time, while a decreasing trend segment indicates a gradual decrease in heart rate over time.

[0047] Continuous heart rates with an increasing trend constitute an increasing trend segment. Specifically, taking the i-th heart rate as an example, the difference between the i+1-th heart rate and the i-th heart rate is calculated. If the difference is positive, it indicates that the i-th heart rate is in an increasing trend and the i-th heart rate is experiencing positive fluctuations. Therefore, the heart rates corresponding to the continuous positive fluctuations in time sequence constitute an increasing trend segment, thereby obtaining a plurality of increasing trend segments. It should be understood that an isolated heart rate in an increasing trend does not constitute a separate increasing trend segment.

[0048] Continuous heart rates with a decreasing trend constitute a decreasing trend segment. Specifically, taking the i-th heart rate as an example, the difference between the i+1-th heart rate and the i-th heart rate is calculated. If the difference is negative, it indicates that the i-th heart rate is in a decreasing trend, and the i-th heart rate is experiencing negative fluctuations. Therefore, the heart rates corresponding to the consecutive negative fluctuations in time sequence constitute a decreasing trend segment, thereby obtaining a plurality of decreasing trend segments. It should be understood that an isolated heart rate in a decreasing trend does not constitute a separate decreasing trend segment.

[0049] The main change trend of the heart rate fluctuation period is determined based on the number of increasing trend segments and decreasing trend segments contained in the heart rate fluctuation period. The main change trend represents the most important change trend exhibited by the heart rate fluctuation period. In an exemplary embodiment, the number of increasing trend segments and the number of decreasing trend segments in the heart rate fluctuation period are obtained, the maximum number is determined from these two numbers, and the change trend corresponding to the trend segment with the largest number is defined as the main change trend. Accordingly, the trend segment corresponding to the main change trend is defined as the main trend segment. The main trend segment can determine the main change trend of the heart rate during the heart rate fluctuation period. If the number of main trend segments accounts for a large proportion, it means that the heart rate fluctuation during the heart rate fluctuation period has a strong regularity, that is, the heart rate fluctuates along a change trend most of the time; if the number accounts for a small proportion, it means that the trend segments are more dispersed and the regularity of the heart rate fluctuation is weak. For example: if the number of increasing trend segments is greater than the number of decreasing trend segments, the main change trend is an increasing trend, and the main trend segment is an increasing trend segment.

[0050] The heart rate fluctuation regularity acquisition module is used to obtain the heart rate fluctuation regularity degree during the heart rate fluctuation period based on the correlation between the increasing trend segment and the decreasing trend segment and the characteristics of the main trend segment.

[0051] In an exemplary embodiment, Figure 5 As shown, the heart rate fluctuation regularity acquisition module is specifically used to: Step 3-1: Obtain the distribution regularity of the heart rate fluctuation period according to the fluctuation of the length of each increasing trend segment and the fluctuation of the length of each decreasing trend segment in the heart rate fluctuation period.

[0052] The length of each increasing trend segment in the heart rate fluctuation period is obtained, and then the fluctuation of the length of the increasing trend segment is obtained. In an exemplary embodiment, the fluctuation is characterized by variance. Then, the variance of the length of each increasing trend segment in the heart rate fluctuation period is obtained and defined as the first variance.

[0053] The length of each decreasing trend segment in the heart rate fluctuation period is obtained, and then the fluctuation of the length of the decreasing trend segment is obtained. In an exemplary embodiment, the fluctuation is characterized by variance. Then, the variance of the length of each decreasing trend segment in the heart rate fluctuation period is obtained and defined as the second variance.

[0054] The length of each trend segment represents the duration of the trend segment, that is, the duration of the changing trend represented by the trend segment. Then, according to the duration and distribution of each increasing trend segment and each decreasing trend segment, the distribution law of the heart rate fluctuation period can be determined. Then, according to the first variance and the second variance, the distribution regularity of the heart rate fluctuation period is obtained. The distribution regularity represents the distribution regularity of each trend segment in the heart rate fluctuation period. Since the variance represents the fluctuation of the length, the larger the variance, the more irregular the distribution of the trend segment. Therefore, the distribution regularity of the heart rate fluctuation period is inversely proportional to the first variance and the second variance. In an exemplary embodiment, a specific quantitative method of the distribution regularity of the heart rate fluctuation period is given as follows: ; in, Indicates the distribution regularity of the bth heart rate fluctuation period, represents the first variance, represents the second variance.

[0055] The larger the value is, the more stable the fluctuation of heart rate in the bth heart rate fluctuation period is, the greater the regularity is, and the more likely it is that the heart rate fluctuation is caused by non-symptomatic factors.

[0056] Step 3-2: According to the distribution regularity, the lengths of the target increasing trend segment and the target decreasing trend segment, as well as the first fitting error and the second fitting error, the heart rate variation regularity characteristics during the heart rate fluctuation period are obtained.

[0057] Obtain the longest length of each increasing trend segment in the heart rate fluctuation period, and define the increasing trend segment corresponding to the longest length as the target increasing trend segment. Similarly, obtain the longest length of each decreasing trend segment in the heart rate fluctuation period, and define the decreasing trend segment corresponding to the longest length as the target decreasing trend segment.

[0058] A two-dimensional coordinate system is constructed with time as the horizontal coordinate and heart rate as the vertical coordinate, and each heart rate in the target increasing trend segment and each heart rate in the target decreasing trend segment are mapped into the two-dimensional coordinate system.

[0059] Perform linear fitting on each heart rate in the target increasing trend segment, for example, using the least squares method, to obtain a first fitting line. Then, calculate the absolute value of the difference between each heart rate in the target increasing trend segment and the fitted heart rate at the same abscissa point on the first fitting line as the fitting error for that abscissa point, thereby obtaining the fitting error for each abscissa point in the target increasing trend segment. Calculate the average of the fitting errors for each abscissa point in the target increasing trend segment as the fitting error between the target increasing trend segment and the first fitting line, which is defined as the first fitting error.

[0060] Similarly, a linear fit is performed on each heart rate in the target decreasing trend segment, for example, using the least squares method, to obtain a second fitting line. The absolute value of the difference between each heart rate in the target decreasing trend segment and the fitted heart rate at the same abscissa point on the second fitting line is then calculated as the fitting error for that abscissa point, thereby obtaining the fitting error for each abscissa point in the target decreasing trend segment. The average of the fitting errors for each abscissa point in the target decreasing trend segment is then calculated as the fitting error between the target decreasing trend segment and the second fitting line, defined as the second fitting error.

[0061] A larger first fitting error indicates a greater difference between the increasing target trend segment and the first fitted line, a greater difference between the increasing target trend segment and the linear increasing trend, and a more irregular heart rate variation, that is, a less obvious heart rate variation pattern. A larger second fitting error indicates a greater difference between the decreasing target trend segment and the second fitted line, a greater difference between the decreasing target trend segment and the linear decreasing trend, and a more irregular heart rate variation, that is, a less obvious heart rate variation pattern.

[0062] According to the distribution regularity of the heart rate fluctuation period, the length of the target increasing trend segment and the target decreasing trend segment in the heart rate fluctuation period, as well as the first fitting error and the second fitting error, the regularity characteristics of the heart rate change in the heart rate fluctuation period are obtained. The stronger the distribution regularity of the heart rate fluctuation period, the more obvious the regularity characteristics of the heart rate change in the heart rate fluctuation period. The longer the length of the target increasing trend segment and the target decreasing trend segment, the more time points they include, that is, the more heart rates they include, and the more reliable the fitting error obtained. Therefore, the length of the target increasing trend segment and the target decreasing trend segment essentially serves as the credibility of the corresponding fitting error. In an exemplary embodiment, as Figure 6 As shown, a specific process of obtaining the characteristics of heart rate variation is given below: Step 3-2-1: According to the length of the target increasing trend segment and the first fitting error, obtain the change error characteristics of the target increasing trend segment.

[0063] Normalize the length of the target increasing trend segment and use the normalized length of the target increasing trend segment as the coefficient of the first fitting error. That is, multiply the normalized length of the target increasing trend segment by the first fitting error, and the result is the change error characteristic of the target increasing trend segment. The specific normalization method is: , where x is the object to be normalized.

[0064] Step 3-2-2: According to the length of the target reduction trend segment and the second fitting error, the variation error characteristics of the target reduction trend segment are obtained.

[0065] Normalize the length of the target decreasing trend segment and use the normalized length of the target decreasing trend segment as the coefficient of the second fitting error. That is, multiply the normalized length of the target decreasing trend segment by the second fitting error, and the result is the change error characteristic of the target decreasing trend segment. The normalization method is as follows: .

[0066] Step 3-2-3: Fuse the change error characteristics of the target increasing trend segment and the target decreasing trend segment to obtain the comprehensive change error characteristics.

[0067] The average value of the change error characteristics of the target increasing trend segment and the change error characteristics of the target decreasing trend segment is calculated, and the result is the comprehensive change error characteristics.

[0068] Step 3-2-4: Fuse the comprehensive features of distribution regularity and variation error to obtain the regularity features of heart rate variation.

[0069] The larger the comprehensive characteristic of the variation error, the more irregular the heart rate variation, that is, the less obvious the regularity of the heart rate variation. The stronger the distribution regularity of the heart rate fluctuation period, the more obvious the regularity of the heart rate variation during the heart rate fluctuation period. Therefore, the regularity of the heart rate variation is proportional to the distribution regularity and inversely proportional to the comprehensive characteristic of the variation error. In an exemplary embodiment, a specific quantification method of the regularity of the heart rate variation is given as follows: ; in, Indicates the regular characteristics of heart rate changes in the bth heart rate fluctuation period, Represents the comprehensive characteristics of the change error in the bth heart rate fluctuation period.

[0070] Step 3-3: According to the regular characteristics of heart rate changes, the number of main trend segments, and the time intervals between adjacent main trend segments, the regularity of heart rate fluctuations is obtained.

[0071] The regular characteristics of heart rate changes during the heart rate fluctuation period, the number of main trend segments in the heart rate fluctuation period, and the time intervals between adjacent main trend segments all affect the regularity of heart rate fluctuations during the heart rate fluctuation period. Therefore, the regularity of heart rate fluctuations during the heart rate fluctuation period is obtained based on these parameters.

[0072] In one exemplary embodiment, the proportion of the main trend segments is determined based on the number of main trend segments in the heart rate fluctuation period and the total number of trend segments in the heart rate fluctuation period (i.e., the sum of the number of increasing trend segments and the number of decreasing trend segments). Specifically, the proportion of the main trend segments is calculated as the ratio of the number of main trend segments to the total number of trend segments in the heart rate fluctuation period. A larger proportion of the main trend segments indicates a more pronounced main fluctuation trend and a higher degree of regularity in the heart rate fluctuation.

[0073] Determine the time interval between each two adjacent main trend segments during the heart rate fluctuation period. Then calculate the average time interval between all two adjacent main trend segments during the heart rate fluctuation period, defining it as the average time interval. A shorter average time interval indicates that the heart rate fluctuates continuously within the main trend within a short period of time during the heart rate fluctuation period. This indicates a more pronounced main fluctuation trend and a higher degree of regularity in the heart rate fluctuations.

[0074] Based on the heart rate fluctuation regularity characteristics of the heart rate fluctuation period, the number and proportion of the main trend segments in the heart rate fluctuation period, and the average time interval corresponding to the heart rate fluctuation period, the degree of heart rate fluctuation regularity during the heart rate fluctuation period is obtained. The degree of heart rate fluctuation regularity is proportional to the heart rate fluctuation regularity characteristics and the number and proportion, and inversely proportional to the average time interval. In an exemplary embodiment, a specific quantification method is given as follows: ; in, Indicates the regularity of heart rate fluctuation in the bth heart rate fluctuation period, Indicates the number of main trend segments in the bth heart rate fluctuation period, Indicates the total number of trend segments in the bth heart rate fluctuation period, Indicates the proportion of the main trend segments in the bth heart rate fluctuation period, represents the average time interval corresponding to the bth heart rate fluctuation period.

[0075] The non-symptom possibility acquisition module is used to obtain the non-symptom possibility of the heart rate fluctuation period based on the characteristic performance and the regularity of the heart rate fluctuation.

[0076] According to the characteristic manifestations of the heart rate fluctuation period and the regularity of the heart rate fluctuation, the non-symptom possibility of the heart rate fluctuation period is obtained. Through the above analysis, the more regular the heart rate fluctuation, that is, the higher the regularity of the heart rate fluctuation, the higher the possibility that the heart rate fluctuation period is non-symptom. Therefore, the non-symptom possibility of the heart rate fluctuation period is proportional to the regularity of the heart rate fluctuation. The characteristic manifestations of the heart rate fluctuation period reflect the severity of the heart rate fluctuation during the heart rate fluctuation period, which is usually a precursor or manifestation of arrhythmia. Then, the higher the characteristic manifestations of the heart rate fluctuation period, the higher the possibility of arrhythmia in the heart rate fluctuation period, that is, the higher the possibility of being a symptom, and the lower the possibility of being non-symptom. Therefore, the non-symptom possibility of the heart rate fluctuation period is inversely proportional to the characteristic manifestations. In an exemplary embodiment, a specific quantification process of the non-symptom possibility of the heart rate fluctuation period is given as follows: ; in, The non-symptom possibility of the bth heart rate fluctuation period is obtained by adopting the above method.

[0077] The correction module is used to reversely correct the heart rate variability during the heart rate fluctuation period according to the possibility of non-symptom.

[0078] Obtain the initial heart rate variability during the heart rate fluctuation period. Heart rate variability can be calculated using methods such as time domain analysis, frequency domain analysis, and nonlinear analysis. The most common method is time domain analysis, in which standard deviation is one of the most common indicators of heart rate variability. Specifically, obtain each RR interval during the heart rate fluctuation period, then calculate the standard deviation of all RR intervals during the heart rate fluctuation period to represent the heart rate variability during the heart rate fluctuation period. Under normal circumstances, the RR intervals are relatively uniform, with small fluctuations and low heart rate variability. However, during arrhythmias, the RR intervals become irregular, exhibiting periodic alternations of short and long intervals. A larger value of heart rate variability indicates better heart health, greater adaptability of the heart to environmental changes, and a lower risk of arrhythmias.

[0079] Based on the likelihood of non-symptoms during the heart rate fluctuation period, the heart rate variability during the heart rate fluctuation period is reversely corrected. That is, the higher the likelihood of non-symptoms, the more the heart rate variability needs to be reduced to avoid making the heart rate variability unreliable. In an exemplary embodiment, a correction method is provided as follows: ; in, represents the corrected heart rate variability of the bth heart rate fluctuation period, which is defined as the target heart rate variability; represents the initial heart rate variability of the bth heart rate fluctuation period, that is, the heart rate variability before correction.

[0080] The above process corrects the heart rate variability of each heart rate fluctuation period, thereby obtaining a more accurate heart rate variability of each heart rate fluctuation period. Subsequently, arrhythmia warning can be performed based on the corrected heart rate variability of each heart rate fluctuation period.

[0081] In an exemplary embodiment, the arrhythmia early warning system further includes an arrhythmia risk level acquisition module, such as Figure 7 As shown, the arrhythmia risk degree acquisition module is used to implement the following process: Step 6: Perform weighted summation based on the length weights of multiple heart rate fluctuation periods and the target heart rate variability to obtain an arrhythmia risk index; Heart rate variability isn't a momentary indicator of health; it's closely linked to a patient's long-term health. Target heart rate variability for a single period of heart rate fluctuation doesn't fully reflect a person's heart health. Non-symptom-related factors can cause short-term disturbances in heart rate variability, but these disturbances are likely temporary and don't represent long-term health issues. Therefore, target heart rate variability is analyzed across multiple periods of heart rate fluctuation within a heart rate time series.

[0082] The length of each heart rate fluctuation period is obtained, and the length weight of each heart rate fluctuation period is obtained based on the length. The longer the length of the heart rate fluctuation period, the more data it contains, and the greater the importance of its target heart rate variability. Therefore, the longer the length of the heart rate fluctuation period, the greater its length weight. In an exemplary embodiment, the sum of the lengths of all heart rate fluctuation periods is obtained, and then the ratio of the length of each heart rate fluctuation period to the sum of the lengths is calculated as the length weight of each heart rate fluctuation period. In this way, the sum of the length weights of all heart rate fluctuation periods is made to be 1.

[0083] According to the length weight of each heart rate fluctuation period, the target heart rate variability of each heart rate fluctuation period is weighted and summed. The result is the arrhythmia risk index, which is as follows: ; Among them, Q represents the arrhythmia risk index, represents the length weight of the bth heart rate fluctuation period, and B represents the number of heart rate fluctuation periods. The higher the arrhythmia risk index, the higher the arrhythmia risk level.

[0084] Step 7: Fuse the time characteristics and arrhythmia risk indicators to obtain the arrhythmia risk level.

[0085] Get the time interval between each two adjacent heart rate fluctuation periods, and then calculate the average time interval between all two adjacent heart rate fluctuation periods. The smaller the average value of the time interval, the more likely it is that the rapid and irregular beats are caused by arrhythmia, and the higher the risk of arrhythmia. The average value of the time interval is negatively normalized, and the result is the time feature. Then, the larger the time feature, the higher the risk of arrhythmia. Therefore, the time feature and the arrhythmia risk index are integrated to obtain the risk level of arrhythmia, as follows: ; Among them, F represents the risk level of arrhythmia, It represents the average value of the time interval between all two adjacent heart rate fluctuation periods. Represents time characteristics.

[0086] After obtaining the risk level of arrhythmia, early warning of arrhythmia can be carried out. The higher the risk level of arrhythmia, the higher the risk of arrhythmia.

[0087] In an exemplary embodiment, the arrhythmia early warning system further includes an early warning module, and the early warning module implements the following process: A first risk level threshold and a second risk level threshold are preset, the first risk level threshold is greater than the second risk level threshold, the numerical ranges of the first risk level threshold and the second risk level threshold are both 0-1, and the specific numerical values of these two thresholds are set according to actual judgment needs. As an example, the first risk level threshold is 0.7 and the second risk level threshold is 0.4.

[0088] The arrhythmia risk level is compared with a first risk level threshold and a second risk level threshold. If the arrhythmia risk level is greater than the first risk level threshold, the arrhythmia is determined to be high risk; if the arrhythmia risk level is less than or equal to the first risk level threshold and greater than or equal to the second risk level threshold, the arrhythmia is determined to be medium risk; if the arrhythmia risk level is less than the second risk level threshold, the arrhythmia is determined to be low risk.

[0089] When the arrhythmia is judged to be high-risk, a real-time warning signal can be immediately sent to the patient's emergency contacts (such as family members or doctors); when the arrhythmia is judged to be medium-risk, relevant reminders to seek medical treatment as soon as possible can be conveyed to the patient (such as sending reminder instructions to a smart watch or smart bracelet); when the arrhythmia is judged to be low-risk, no additional reminders may be given to the patient.

[0090] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. An early warning system for arrhythmia based on individual heart rate variability analysis, characterized by: include: A characteristic performance acquisition module is used to determine the characteristic performance of the heart rate fluctuation period of the heart rate time series sequence based on the instability of the heart rate in the heart rate time series sequence; A division module is used to divide the heart rate fluctuation period into an increasing trend segment and a decreasing trend segment according to the heart rate change trend, and determine the main change trend of the heart rate fluctuation period; a heart rate fluctuation regularity acquisition module, configured to obtain the degree of heart rate fluctuation regularity during the heart rate fluctuation period based on the correlation between the increasing trend segment and the decreasing trend segment and in combination with the characteristics of the main trend segment; the main trend segment being the trend segment corresponding to the main change trend; a non-symptom possibility obtaining module, configured to obtain the non-symptom possibility of the heart rate fluctuation period based on the characteristic manifestations and the regularity of the heart rate fluctuation; The correction module is used to reversely correct the heart rate variability during the heart rate fluctuation period according to the non-symptom possibility.

2. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 1, characterized in that: The heart rate fluctuation regularity acquisition module is specifically used for: According to the fluctuation of the length of each increasing trend segment and the fluctuation of the length of each decreasing trend segment in the heart rate fluctuation period, the distribution regularity of the heart rate fluctuation period is obtained; Obtaining a heart rate variation regularity characteristic during the heart rate fluctuation period based on the distribution regularity, the lengths of the target increasing trend segment and the target decreasing trend segment, and the first fitting error and the second fitting error; The target increasing trend segment is the longest increasing trend segment among the increasing trend segments in the heart rate fluctuation period, and the target decreasing trend segment is the longest decreasing trend segment among the decreasing trend segments in the heart rate fluctuation period; the first fitting error is the fitting error between the target increasing trend segment and its fitting straight line, and the second fitting error is the fitting error between the target decreasing trend segment and its fitting straight line; The degree of regularity of the heart rate fluctuation is obtained according to the regularity characteristics of the heart rate variation, the number of main trend segments, and the time intervals between adjacent main trend segments.

3. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 2, characterized in that: The process of obtaining the distribution regularity includes: obtaining the distribution regularity based on a first variance and a second variance; the first variance is the variance of the length of each increasing trend segment in the heart rate fluctuation period, and the second variance is the variance of the length of each decreasing trend segment in the heart rate fluctuation period; the distribution regularity is inversely proportional to the first variance and the second variance.

4. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 2, characterized in that: The process of acquiring the regular characteristics of heart rate changes includes: Obtaining a variation error characteristic of the target increasing trend segment according to the length of the target increasing trend segment and the first fitting error; Obtaining a variation error characteristic of the target reduction trend segment according to the length of the target reduction trend segment and a second fitting error; fusing the change error features of the target increasing trend segment and the target decreasing trend segment to obtain a comprehensive change error feature; The distribution regularity and the comprehensive characteristics of the change error are integrated to obtain the heart rate change regularity characteristics.

5. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 2, characterized in that: The process of obtaining the regularity of the heart rate fluctuation includes: The degree of regularity of heart rate fluctuation is obtained based on the characteristics of the heart rate change pattern, the proportion of the number of main trend segments, and the average time interval; the degree of regularity of heart rate fluctuation is directly proportional to the characteristics of the heart rate change pattern and the proportion of the number, and inversely proportional to the average time interval; the proportion of the number is obtained by the number of main trend segments and the total number of trend segments in the heart rate fluctuation period; the average time interval is the average value of the time intervals between all two adjacent main trend segments.

6. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 1, characterized in that: The feature expression acquisition module is specifically used to: Dividing the heart rate time series into a heart rate fluctuation period and a heart rate stable period according to the difference between two adjacent heart rates; Determine the difference in heart rate instability between periods of heart rate fluctuation and adjacent periods of heart rate stability; The difference between the maximum heart rate and the minimum heart rate in the heart rate fluctuation period and the time interval therebetween are determined, and the characteristic performance is obtained in combination with the heart rate instability difference.

7. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 6, characterized in that: The process of obtaining the heart rate instability difference includes: Calculating the average of the absolute values of the differences between all two adjacent heart rates in the target period to obtain the heart rate instability of the target period; the target period is any one of the heart rate fluctuation period and the heart rate stable period; The absolute value of the difference between the heart rate instability of the heart rate fluctuation period and the adjacent heart rate stable period is calculated to obtain the heart rate instability difference.

8. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 1, characterized in that: The process of dividing the increasing trend segment and the decreasing trend segment includes: forming the increasing trend segment with the continuous heart rate with the increasing trend, and forming the decreasing trend segment with the continuous heart rate with the decreasing trend; The process of obtaining the main change trend includes: obtaining the number of increasing trend segments and the number of decreasing trend segments in the heart rate fluctuation period, and taking the change trend corresponding to the trend segment with the largest number as the main change trend.

9. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 1, characterized in that: The arrhythmia early warning system further includes an arrhythmia risk level acquisition module, which is used to: Obtaining an arrhythmia risk index based on weighted summation of length weights of multiple heart rate fluctuation periods and target heart rate variability; wherein the length weight is obtained from the length of the heart rate fluctuation period; The target heart rate variability is the heart rate variability after reverse correction; The time feature and the arrhythmia risk index are integrated to obtain the arrhythmia risk level; the time feature is obtained by averaging the time intervals between all two adjacent heart rate fluctuation periods.

10. The arrhythmia early warning system based on individual heart rate variability analysis as claimed in claim 9, characterized in that: The arrhythmia early warning system further includes a warning module for: Comparing the arrhythmia risk level with a preset first risk level threshold and a second risk level threshold; wherein the first risk level threshold is greater than the second risk level threshold; If the arrhythmia risk level is greater than the first risk level threshold, determining that the arrhythmia is at high risk; If the arrhythmia risk level is less than or equal to the first risk level threshold and greater than or equal to the second risk level threshold, determining that the arrhythmia is of medium risk; If the arrhythmia risk level is less than the second risk level threshold, the arrhythmia is determined to be of low risk.

Images