Heart rate data analysis system and method

Abstract

The invention relates to the technical field of data processing, and provides a heart rate data analysis system and method.The heart rate data analysis method comprises the steps that heart rate monitoring signals are collected through wearable heart rate monitoring equipment, and motion acceleration data at all moments are recorded; analyzing the discrete and differential performance of the central rate monitoring data in the monitoring time period, and quantifying the heart rate change intense index in the monitoring time period; obtaining a motion transformation index of the monitoring time period based on the discrete representation of the motion acceleration data in the monitoring time period; obtaining the interference source possibility of each independent component signal; screening a plurality of interference source component signals according to the interference source possibility; obtaining the frequency decoupling degree of each interference source component signal; obtaining a preprocessing demand degree of each interference source component signal; and performing smoothing based on the preprocessing demand degree, performing reconstruction to obtain a processed heart rate monitoring signal, and performing data analysis. The invention aims to solve the problem that heart rate data monitoring is influenced by a motion mode to generate motion artifacts.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a heart rate data analysis system and method. Background Art

[0002] Heart rate (HR) is a key physiological parameter that reflects heart health and is commonly used to monitor various aspects of an individual's cardiovascular health, exercise status, and psychological stress. With growing awareness of health management, heart rate data analysis systems have gained widespread application in healthcare, fitness, sports analysis, and health monitoring. Traditional heart rate monitoring relies on specialized equipment in medical institutions. However, with the rapid development of wearable technology, devices such as smart bracelets and smart watches are becoming increasingly popular, allowing users to obtain their own heart rate data in real time. These devices not only monitor heart rate changes in real time, but also analyze the user's exercise status and provide health advice, thereby helping users better manage their health.

[0003] Existing analysis methods for heart rate data usually use fixed templates or unified algorithms, which cannot fully consider the impact of behavioral pattern factors on heart rate fluctuation patterns, and it is difficult to provide refined and personalized risk assessments; and current heart rate data analysis systems usually rely on ECG (electrocardiogram) or PPG (photoplethysmography) signals collected by wearable devices. These signals are easily affected by various non-physiological factors such as motion artifacts, ambient light interference, contact resistance changes, sensor drift, etc. during the acquisition process, resulting in abnormal characteristics such as noise superposition, baseline drift, peak distortion or signal truncation in the original generation stage of heart rate data, which in turn affects the accuracy of subsequent heart rate data analysis. Summary of the Invention

[0004] The present invention provides a heart rate data analysis method and system to solve the problem that existing heart rate data monitoring is affected by exercise patterns and produces motion artifacts. The technical solutions adopted are as follows: The present invention proposes a heart rate data analysis method, which includes the following steps: The wearable heart rate monitoring device collects heart rate monitoring signals, obtains heart rate monitoring data for the monitoring period and several moments therein, and records motion acceleration data at each moment; Analyze the discrete and differential performance of the heart rate monitoring data during the monitoring period and quantify the heart rate variation intensity index during the monitoring period; based on the discrete performance of the motion acceleration data during the monitoring period and the time difference of its discrete relationship with the heart rate monitoring data, combine the heart rate variation intensity index to obtain the motion transformation index of the monitoring period; based on the time series distribution of the extreme points in each independent component signal of the heart rate monitoring data and the area difference between the peak areas, obtain the mixing degree of each independent component signal, and then combine the motion transformation index to obtain the possibility of the interference source of each independent component signal; Screening several interference source component signals based on the possibility of the interference source; decomposing each interference source component signal by EMD to obtain several IMF component signals; based on the correlation between the instantaneous frequency performance of each IMF component signal and the order value of the IMF component signal, combined with the quality index of the data points in each IMF component signal, obtaining the frequency decoupling degree of each interference source component signal; combining the interference source possibility of the interference source component signal, obtaining the preprocessing requirement degree of each interference source component signal; The component signals of each interference source are smoothed based on the preprocessing requirement, and the processed heart rate monitoring signal is obtained by combining the independent component signals for reconstruction.

[0005] Optionally, the heart rate variation intensity index during the monitoring period is obtained by: Arrange all heart rate monitoring data in the monitoring period in ascending order to obtain a heart rate ascending sequence, obtain the difference between any two adjacent heart rate monitoring data in the heart rate ascending sequence, and use the two adjacent heart rate monitoring data corresponding to the maximum value of the difference between all two adjacent heart rate monitoring data as segmented heart rate data; the segmented heart rate data on the left side of the two segmented heart rate data and all heart rate data on the left side in the heart rate ascending sequence constitute a low heart rate data set for the monitoring period, and the segmented heart rate data on the right side of the two segmented heart rate data and all heart rate data on the right side in the heart rate ascending sequence constitute a high heart rate data set for the monitoring period; The difference obtained by subtracting the mean of the heart rate monitoring data in the low heart rate data set from the mean of the heart rate monitoring data in the high heart rate data set is used as the heart rate dispersion coefficient of the monitoring period; the absolute value of the difference between the heart rate monitoring data at any two adjacent moments in the monitoring period is obtained, and the product of the mean of the absolute values ​​of the difference between the heart rate monitoring data at all two adjacent moments in the monitoring period and the heart rate dispersion coefficient is used as the heart rate change intensity index of the monitoring period.

[0006] Optionally, obtaining the motion transformation index during the monitoring period includes the following specific methods: Arrange all the motion acceleration data in the monitoring period in ascending order to obtain an acceleration ascending sequence, and obtain a low acceleration data set and a high acceleration data set of the monitoring period based on the difference between adjacent motion acceleration data in the acceleration ascending sequence; Obtain the absolute value of the difference between the minimum value of all motion acceleration data at the corresponding moment in the high acceleration data set and the minimum value of all heart rate monitoring data at the corresponding moment in the high heart rate data set, and record it as the high heart rate deviation time difference during the monitoring period; obtain the absolute value of the difference between the number of motion acceleration data in the high acceleration data set and the number of heart rate monitoring data in the high heart rate data set, and record it as the high heart rate quantity deviation during the monitoring period; Based on the heart rate change intensity index during the monitoring period, as well as the high heart rate deviation time difference and the high heart rate quantity deviation, the motion transformation index of the monitoring period is obtained; the motion transformation index is positively correlated with the heart rate change intensity index, and negatively correlated with the high heart rate deviation time difference and the high heart rate quantity deviation.

[0007] Optionally, the obtaining of the mixing degree of each independent component signal includes the following specific methods: The heart rate monitoring signal composed of the heart rate monitoring data during the monitoring period is subjected to ICA independent component analysis to obtain a number of independent component signals; For any independent component signal, a number of extreme value points are obtained, including a number of maximum value points and a number of minimum value points, and the time interval between any two adjacent extreme value points is obtained. The standard deviation of the time interval between all two adjacent extreme value points is used as the extreme value distribution fluctuation coefficient of the independent component signal; Based on the moments corresponding to any maximum point and its two adjacent minimum points, the peak area of ​​the maximum point is constructed, and the area of ​​the peak area of ​​the maximum point is obtained by integration; the absolute value of the difference between the areas of the peak areas of any two maximum points is obtained, and the average of the absolute values ​​of the differences in all the obtained areas and the product of the extreme value distribution fluctuation coefficient are used as the degree of mixing of the independent component signal.

[0008] Optionally, obtaining the possibility of interference sources of each independent component signal includes the following specific methods: The product of the mixing degree of any independent component signal and the motion transformation index of the monitoring period is used as the interference source possibility of the independent component signal.

[0009] Optionally, the screening of several interference source component signals according to the possibility of the interference source includes the following specific methods: Arrange all independent component signals in ascending order of their interference source possibilities to obtain a component signal sequence, obtain the difference between the interference source possibilities of adjacent independent component signals in the component signal sequence, and use the right independent component signal of the two independent component signals corresponding to the maximum value of the difference between all the interference source possibilities as the interference source component signal; The other independent component signals on the right side of the interference source component signal in the component signal sequence are taken as interference source component signals.

[0010] Optionally, obtaining the frequency decoupling degree of each interference source component signal includes the following specific methods: Based on the instantaneous frequency performance of each IMF component signal and the order value of the IMF component signal, the decomposition disorder degree of each interference source component signal is obtained; Calculate the quality index of each data point in any IMF component signal of any interference source component signal, and obtain the average of the quality indexes of all data points in the IMF component signal as the periodic stability coefficient of the IMF component signal; Based on the decomposition chaos degree of the interference source component signal and the cyclic stability mean of all IMF component signals of the interference source component signal, the frequency decoupling degree of the interference source component signal is obtained. The frequency decoupling degree is negatively correlated with the decomposition chaos degree and positively correlated with the cyclic stability mean.

[0011] Optionally, the obtaining of the decomposition disorder degree of each interference source component signal includes the following specific methods: Performing Fourier transform on any IMF component signal of any interference source component signal to obtain several instantaneous frequencies of the IMF component signal, and taking the mean of all instantaneous frequencies as the frequency representation value of the IMF component signal; Arranging the frequency representation values ​​of each IMF component signal in the interference source component signal according to the order of the IMF component signal to obtain a frequency representation sequence of the interference source component signal, and arranging the order values ​​of each IMF component signal to obtain an order value sequence of the interference source component signal; The Pearson correlation coefficient between the frequency representation sequence and the order value sequence is obtained, and normalized based on the value range of the Pearson correlation coefficient, and the normalized result is used as the decomposition chaos degree of the interference source component signal.

[0012] Optionally, the preprocessing requirement of each interference source component signal is specifically obtained by: The Euclidean norm of the inversely proportional normalized result of the frequency decoupling degree of any interference source component signal and the normalized result of the interference source possibility is used as the preprocessing requirement of the interference source component signal.

[0013] The present invention also proposes a heart rate data analysis system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0014] The beneficial effects of the present invention are as follows: the present invention obtains heart rate monitoring signals in real time through a wearable heart rate monitoring device and records motion acceleration data, quantifies the drastic changes in heart rate through the discrete distribution and fluctuation difference of the heart rate monitoring data during the monitoring period, and further obtains a motion transformation index to reflect the possibility that the heart rate monitoring data is affected by motion artifacts in combination with the discrete performance of the motion acceleration data and the time difference with the discrete distribution of the heart rate monitoring data; through independent component analysis, the distribution of extreme points and the peak area difference in the independent component component signal are quantified to reflect that the independent component component signal does not conform to the independent decomposition process of the independent component, and then combines the motion transformation index to present the possibility of the interference source; and through the local feature analysis of the quality index of the IMF component signal under the EMD decomposition of the interference source component signal, and the frequency feature extraction reflects whether the interference source component signal is clearly decomposed and periodically stable, and then obtains the frequency decoupling degree. The interference source component signal with a smaller frequency decoupling degree and a greater possibility of interference source needs to be subjected to a greater filtering requirement, and then the acquired processed heart rate monitoring signal is reconstructed to eliminate the influence of motion artifacts on the heart rate monitoring data, thereby improving the accuracy of heart rate data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 paying any creative work.

[0016] Figure 1 A flowchart of a heart rate data analysis method provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] See also Figure 1, which shows a flow chart of a heart rate data analysis method provided by one embodiment of the present invention, the method comprising the following steps: Step S001: collect heart rate monitoring signals through a wearable heart rate monitoring device, obtain heart rate monitoring data for a monitoring period and several moments therein, and record motion acceleration data at each moment.

[0019] The purpose of this embodiment is to eliminate motion artifacts caused by factors such as exercise in the heart rate monitoring signal collected by a wearable heart rate monitoring device, thereby improving the accuracy of the heart rate monitoring signal and thus improving the effect of data analysis.

[0020] Specifically, the user's heart rate data is collected through a wearable heart rate monitoring device (such as an ECG bracelet or PPG smart watch), and the original physiological electrical signals or photoplethysmography signals are collected and converted into the user's heart rate monitoring signal. The sampling time interval of the heart rate monitoring signal is set in the device, and acceleration data is recorded at the same sampling time interval. The acceleration data is obtained through the built-in acceleration sensor of the bracelet or smart watch to obtain the motion acceleration data at each moment. The original signal includes the timestamp, signal value and channel number corresponding to the sampling point, and the heart rate signal is obtained after conversion.

[0021] Furthermore, the time period length is preset. In this embodiment, the time period length is described as 10 minutes. The heart rate monitoring data at each moment corresponding to the collected heart rate monitoring signals are divided based on the time period length to obtain several monitoring time periods and the heart rate monitoring data at each moment therein. Then, subsequent correction processing of the heart rate monitoring data is performed once in each monitoring time period. In this embodiment, the most recent 10-minute time period is used as the monitoring time period for subsequent processing.

[0022] Step S002: Analyze the discrete and differential performance of the heart rate monitoring data during the monitoring period, and quantify the heart rate change intensity index during the monitoring period; based on the discrete performance of the motion acceleration data during the monitoring period, and the time difference of its discrete relationship relative to the heart rate monitoring data, combined with the heart rate change intensity index, obtain the motion transformation index of the monitoring period; based on the time series distribution of the extreme points in the independent component component signals of the heart rate monitoring data, and the area difference between the peak areas, obtain the mixing degree of each independent component component signal, and then combine the motion transformation index to obtain the possibility of the interference source of each independent component component signal.

[0023] It should be noted that by analyzing the impact of exercise patterns on the fluctuation performance of heart rate monitoring data, data support can be provided for the subsequent component identification in the heart rate monitoring data. By introducing motion sensing features such as motion acceleration data, a behavioral interpretation of the heart rate signal fluctuation pattern can be performed, which can distinguish non-physiological heart rate fluctuations caused by exercise from real heart rate changes, and significantly improve the interference targeting of subsequent decomposition processing.

[0024] Preferably, in one embodiment of the present invention, the discrete and differential performance of the heart rate monitoring data during the monitoring period is analyzed to quantify the heart rate variation intensity index during the monitoring period, including the following specific methods: Arrange all the heart rate monitoring data in the monitoring period in ascending order to obtain a heart rate ascending sequence, obtain the difference between any two adjacent heart rate monitoring data in the heart rate ascending sequence (obtained by subtracting the previous one from the latter heart rate monitoring data), and use the two adjacent heart rate monitoring data corresponding to the maximum value of the difference between all two adjacent heart rate monitoring data as the segmented heart rate data; the segmented heart rate data on the left side of the two segmented heart rate data and all the heart rate data on the left side in the heart rate ascending sequence constitute the low heart rate data set of the monitoring period, and the segmented heart rate data on the right side of the two segmented heart rate data and all the heart rate data on the right side in the heart rate ascending sequence constitute the high heart rate data set of the monitoring period.

[0025] Furthermore, the difference obtained by subtracting the mean of the heart rate monitoring data in the low heart rate data set from the mean of the heart rate monitoring data in the high heart rate data set is used as the heart rate dispersion coefficient of the monitoring period; the absolute value of the difference between the heart rate monitoring data at any two adjacent moments in the monitoring period is obtained, and the product of the mean of the absolute values ​​of the difference between the heart rate monitoring data at all two adjacent moments in the monitoring period and the heart rate dispersion coefficient is used as the heart rate change intensity index of the monitoring period.

[0026] It should be noted that the greater the difference in the mean of the heart rate monitoring data between the high heart rate data set and the low heart rate data set, the more discrete the heart rate monitoring data during the monitoring period. At the same time, the greater the difference in the mean of the heart rate monitoring data at adjacent moments, the more frequent the drastic changes in the heart rate monitoring data, and the greater the heart rate change intensity index.

[0027] It should be further explained that changes in the user's heart rate during the detection process are often caused by drastic changes in their own motion state or changes in the heart rate data itself. Changes in heart rate caused by drastic changes in their own motion state are usually normal behavior and do not indicate abnormal behavior in the user's heart rate changes.

[0028] Preferably, in one embodiment of the present invention, based on the discrete representation of the motion acceleration data in the monitoring period and the time difference of the discrete relationship relative to the heart rate monitoring data, combined with the heart rate change intensity index, the motion transformation index of the monitoring period is obtained, including the specific method of: Arrange all motion acceleration data in the monitoring period in ascending order to obtain an acceleration ascending sequence. According to the method for obtaining the low heart rate data set and the high heart rate data set of the monitoring period, based on the difference between adjacent motion acceleration data in the acceleration ascending sequence, obtain the low acceleration data set and the high acceleration data set of the monitoring period; obtain the minimum value of all motion acceleration data in the high acceleration data set at the corresponding moment, and the absolute value of the difference between the minimum value of all heart rate monitoring data in the high heart rate data set at the corresponding moment, and record it as the high heart rate deviation time difference of the monitoring period; obtain the number of motion acceleration data in the high acceleration data set, and the absolute value of the difference between the number of heart rate monitoring data in the high heart rate data set, and record it as the high heart rate quantity deviation of the monitoring period.

[0029] Furthermore, based on the heart rate change intensity index of the monitoring period, the high heart rate deviation time difference and the high heart rate quantity deviation, the motion transformation index of the monitoring period is obtained; the motion transformation index is positively correlated with the heart rate change intensity index, and negatively correlated with the high heart rate deviation time difference and the high heart rate quantity deviation.

[0030] As an example, the product of the high heart rate deviation time difference and the high heart rate quantity deviation during the monitoring period is obtained, and the ratio of the heart rate change intensity index to the product is used as the motion transformation index of the monitoring period; it is particularly noted that in order to avoid the denominator being 0, which makes the fraction meaningless, this embodiment adds a hyperparameter to the numerator and denominator respectively during the ratio calculation process, and the hyperparameter in this embodiment is described as 0.1.

[0031] What needs to be explained is that the smaller the high heart rate deviation time difference and the smaller the high heart rate quantity deviation, the more likely that the drastic fluctuation of the heart rate monitoring data is caused by the change of exercise mode. At the same time, the larger the heart rate change intensity index is as a basis, the larger the exercise change index is further obtained, which means that the drastic change in the user's heart rate may be caused by a drastic change in the user's exercise behavior, which in turn causes the heart rate data to show unexpected fluctuations, and the larger the motion artifact.

[0032] It should be further explained that the real heart rate signal exhibits a certain rhythmicity and stability, and motion artifacts usually manifest as high-frequency oscillations (such as jitters during running); disturbances do not reflect real fluctuations in physiological state, but are caused by limb movements, environmental vibrations, etc., and are non-target source signals. They are interference sources and should be separated from the heart rate monitoring data; ICA independent component analysis can decompose the heart rate monitoring signal into several original signal components, enhancing the ability to remove motion artifacts.

[0033] Preferably, in one embodiment of the present invention, based on the temporal distribution of extreme value points in each independent component signal of the heart rate monitoring data and the area difference between the peak areas, the mixing degree of each independent component signal is obtained, and then combined with the motion transformation index to obtain the possibility of the interference source of each independent component signal, the specific method includes: The heart rate monitoring signal composed of the heart rate monitoring data during the monitoring period is subjected to ICA independent component analysis to obtain a number of independent component component signals. In this embodiment, a total of 5 independent component component signals are obtained, wherein the independent component component signal is set based on the prior knowledge of ICA independent component analysis. ICA independent component analysis is a well-known technology and will not be described in detail in this embodiment. For any independent component component signal, a number of extreme points are obtained, including a number of maximum points and a number of minimum points. The time interval between any two adjacent extreme points is obtained, and the standard deviation of the time interval between all two adjacent extreme points is used as the extreme value distribution fluctuation coefficient of the independent component component signal. Among them, the existing algorithm for extreme point extraction is used to obtain the extreme points of the independent component component signal, which will not be described in detail in this embodiment.

[0034] Furthermore, based on the moments corresponding to any maximum point and its two adjacent minimum points, the peak area of ​​the maximum point is constructed, that is, the time period between the corresponding moments of the two minimum points, and the curve of the independent component component signal constitutes the peak area; it is particularly noted that if any maximum point is close to the boundary of the monitoring period and there is no minimum point on one side, then it is constructed based on the corresponding boundary to the corresponding moment part of the maximum point; the area of ​​the peak area of ​​the maximum point is obtained by integration; the absolute value of the difference between the areas of the peak areas of any two maximum points is obtained, and the mean of the absolute values ​​of the differences of all the obtained areas and the product of the extreme value distribution fluctuation coefficient are used as the mixing degree of the independent component component signal.

[0035] It should be noted that the greater the area difference of the peak area in the independent component signal during the ICA decomposition process, and the more uneven the distribution of extreme points, the less the decomposition process of the independent component signal conforms to the decomposition law of ICA, indicating that the internal chaos of the component signal is worse and the independence is worse, that is, the waveform is unstable and there is high-frequency information in the component signal.

[0036] Furthermore, the product of the mixing degree of any independent component signal and the motion transformation index of the monitoring period is used as the possibility of the interference source of the independent component signal.

[0037] It should be noted that the motion transformation index is used as a weight reference for component selection, that is, in the period when the motion transformation index is higher, the component signals containing high-frequency components or frequent changes are preferentially classified as interference sources, and the remaining component signals with stable waveforms and good rhythmicity are regarded as the main components of the heart rate; that is, in the process of independent component analysis of the heart rate monitoring data, the possibility of the component signal being an interference source is obtained.

[0038] Step S003, screening several interference source component signals according to the possibility of the interference source; decomposing each interference source component signal by EMD to obtain several IMF component signals, based on the correlation between the instantaneous frequency performance of each IMF component signal and the order value of the IMF component signal, combined with the quality difference index of the data points in each IMF component signal, obtain the frequency decoupling degree of each interference source component signal; combined with the interference source possibility of the interference source component signal, obtain the preprocessing requirement degree of each interference source component signal.

[0039] Preferably, in one embodiment of the present invention, a plurality of interference source component signals are screened according to the possibility of the interference source, including the specific method of: Specifically, all independent component component signals are arranged in ascending order of their interference source possibilities to obtain a component signal sequence, and the difference between the interference source possibilities of adjacent independent component component signals in the component signal sequence is obtained (obtained by subtracting the interference source possibility of the previous independent component component signal from the next independent component component signal), and the independent component component signal on the right side of the two independent component component signals corresponding to the maximum value of the difference between all the interference source possibilities is taken as the interference source component signal; and the other independent component component signals on the right side of the interference source component signal in the component signal sequence are taken as the interference source component signals.

[0040] It should be noted that the component signals of the possible interference sources are divided according to the maximum difference to obtain the interference source component signals; at the same time, the screening process is made more selective to avoid mistakenly deleting the main heart rate signal or retaining artifacts; after completing the independent component analysis (ICA) guided by the motion transformation index, the component signals that may belong to the interference source have been preliminarily identified; in dynamic scenarios, the interference components in some heart rate data may have a certain degree of statistical correlation or mixing with the target physiological signals, so that there is still residual coupling between some components after ICA decomposition, which manifests as weak independence, and the interference source component signals need to be further decomposed and filtered.

[0041] It should be further explained that in order to further improve the accuracy of signal purification, a local feature extraction mechanism is introduced for interference source component signals with insignificant independence, so as to achieve deeper identification and qualitative analysis of potential interference components; by combining nonlinear analysis methods (such as empirical mode decomposition EMD), multimodal decomposition and spectral structure analysis are performed on the retained signal, so as to identify high-frequency disturbances, pseudo-peaks and short-time unstable waveforms mixed in the main components, thereby achieving deep purification of the signal.

[0042] Preferably, in one embodiment of the present invention, each interference source component signal is decomposed by EMD to obtain a plurality of IMF component signals, and based on the correlation between the instantaneous frequency performance of each IMF component signal and the order value of the IMF component signal, combined with the quality index of the data point in each IMF component signal, the frequency decoupling degree of each interference source component signal is obtained, including the specific method of: Perform EMD decomposition on any interference source component signal to obtain several IMF component signals of the interference source component signal, and the several IMF component signals have order values ​​and are arranged in ascending order. EMD decomposition is a well-known technology and will not be described in detail in this embodiment.

[0043] Further, any IMF component signal of the interference source component signal is subjected to Fourier transform to obtain several instantaneous frequencies of the IMF component signal (several frequencies in the spectrum obtained by Fourier transform), and the mean of all instantaneous frequencies is used as the frequency expression value of the IMF component signal; the frequency expression values ​​of each IMF component signal in the interference source component signal are arranged according to the order of the IMF component signal to obtain a frequency expression sequence of the interference source component signal, and the order values ​​of each IMF component signal are arranged to obtain an order value sequence of the interference source component signal; the Pearson correlation coefficient of the frequency expression sequence and the order value sequence is obtained, and normalized based on the value range of the Pearson correlation coefficient, and the normalized result is used as the decomposition chaos degree of the interference source component signal; wherein the value range of the Pearson correlation coefficient is , then normalizing its range is to linearly transform the range to In the linear transformation process, the sum of the Pearson correlation coefficient plus 1 is divided by 2, and the quotient obtained is the normalized result.

[0044] Furthermore, a quality index is calculated for each data point in any IMF component signal of any interference source component signal (the quality index calculation is an existing method and will not be repeated in this embodiment), and the average quality index of all data points in the IMF component signal is obtained as the periodic stability coefficient of the IMF component signal.

[0045] Furthermore, based on the decomposition chaos degree of the interference source component signal and the cyclic stability mean of all IMF component signals of the interference source component signal, the frequency decoupling degree of the interference source component signal is obtained, and the frequency decoupling degree is negatively correlated with the decomposition chaos degree and positively correlated with the cyclic stability mean.

[0046] As an example, the ratio of the mean of the cyclic stability of all IMF component signals of any interference source component signal to the decomposition chaos degree is used as the frequency decoupling degree of the interference source component signal; it is particularly noted that in order to avoid the denominator being 0 resulting in meaningless fractions, this embodiment adds a hyperparameter to the numerator and denominator respectively during the ratio calculation process, and the hyperparameter in this embodiment is described as 0.001.

[0047] It should be noted that the closer the correlation coefficient between the frequency expression sequence and the order value sequence is to -1, that is, negative correlation, the smaller the decomposition chaos, the clearer the EMD decomposition of the interference source component signal, and the higher the periodic stability of the IMF component signal, the more uniform and stable the periodic change, thereby obtaining a greater frequency decoupling degree, reflecting the clear decomposition process of the interference source component signal.

[0048] It should be further explained that for the collected heart rate monitoring data, independent component analysis is used to identify the different components in the heart rate monitoring data. The ability of independent component analysis is "global statistical independence decomposition", which mainly relies on the statistical independence between signal sources for unmixing. However, the real heart rate plus motion interference signal may be a nonlinear superposition, and ICA cannot completely decouple it, and cannot retain the timing and frequency stratification information of the components, which is not suitable for instantaneous feature analysis. Therefore, the preliminary identification of independent component analysis is used, and then the empirical mode decomposition is used to perform deeper component identification, so as to realize the accurate identification and stratification processing of the main fluctuations of heart rate, local interference artifacts and signal mutations, thereby improving the accuracy and stability of heart rate data processing.

[0049] Preferably, in one embodiment of the present invention, the preprocessing requirement of each interference source component signal is obtained in combination with the interference source possibility of the interference source component signal, including the specific method of: The Euclidean norm of the inversely proportional normalized result of the frequency decoupling degree of any interference source component signal and the normalized result of the interference source possibility is used as the preprocessing requirement of the interference source component signal.

[0050] It should be noted that this embodiment adopts Model to present inverse proportional relationship and normalization processing, is the input of the model, It represents an exponential function with a natural constant as the base. At the same time, the sigmoid function is used for separate normalization processing. The implementer can set the inverse proportional function and normalization function according to the actual situation.

[0051] It should be noted that the Euclidean norm reflects the degree of interaction between the two factors, avoids one-sided judgments, and increases the robustness of the preprocessing strategy. In terms of signal processing, combining the frequency decoupling degree and the possibility of interference sources helps to accurately assess the degree of interference in the heart rate monitoring signal, thereby accurately preprocessing the heart rate signal.

[0052] Step S004: smoothing each interference source component signal based on the preprocessing requirement, combining the independent component component signal to reconstruct the processed heart rate monitoring signal, and performing data analysis on the processed heart rate monitoring signal.

[0053] Specifically, this embodiment uses a Gaussian filtering algorithm to smooth the interference source component signal, takes the preprocessing requirement of any interference source component signal as the filtering strength value of the filter, and then smoothes to obtain several processed interference source component signals, combines other independent component signals except the interference source component signal, and reconstructs the obtained signal as the processed heart rate monitoring signal, that is, eliminates motion artifacts in the heart rate monitoring signal.

[0054] Furthermore, based on the processed heart rate monitoring signal, that is, the "purified" heart rate monitoring signal, feature extraction is used to identify phenomena such as arrhythmia, paroxysmal tachycardia and stress response, so as to realize the analysis of heart rate data and eliminate the problem of inaccurate heart rate monitoring data caused by motion artifacts in wearable heart rate monitoring devices.

[0055] At this point, this embodiment is completed.

[0056] Another embodiment of the present invention provides a heart rate data analysis system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, steps S001 to S004 of the above method are implemented.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A heart rate data analysis method, characterized in that: The method comprises the following steps: The wearable heart rate monitoring device collects heart rate monitoring signals, obtains heart rate monitoring data for the monitoring period and several moments therein, and records motion acceleration data at each moment; Analyze the discrete and differential performance of the heart rate monitoring data during the monitoring period and quantify the heart rate variation intensity index during the monitoring period; based on the discrete performance of the motion acceleration data during the monitoring period and the time difference of its discrete relationship with the heart rate monitoring data, combine the heart rate variation intensity index to obtain the motion transformation index of the monitoring period; based on the time series distribution of the extreme points in each independent component signal of the heart rate monitoring data and the area difference between the peak areas, obtain the mixing degree of each independent component signal, and then combine the motion transformation index to obtain the possibility of the interference source of each independent component signal; Screening several interference source component signals based on the possibility of the interference source; decomposing each interference source component signal by EMD to obtain several IMF component signals; based on the correlation between the instantaneous frequency performance of each IMF component signal and the order value of the IMF component signal, combined with the quality index of the data points in each IMF component signal, obtaining the frequency decoupling degree of each interference source component signal; combining the interference source possibility of the interference source component signal, obtaining the preprocessing requirement degree of each interference source component signal; The component signals of each interference source are smoothed based on the preprocessing requirement, and the processed heart rate monitoring signal is obtained by combining the independent component signals for reconstruction.

2. A heart rate data analysis method according to claim 1, characterized in that: The heart rate variation intensity index during the monitoring period is obtained in the following specific method: Arrange all heart rate monitoring data in the monitoring period in ascending order to obtain a heart rate ascending sequence, obtain the difference between any two adjacent heart rate monitoring data in the heart rate ascending sequence, and use the two adjacent heart rate monitoring data corresponding to the maximum value of the difference between all two adjacent heart rate monitoring data as segmented heart rate data; the segmented heart rate data on the left side of the two segmented heart rate data and all heart rate data on the left side in the heart rate ascending sequence constitute a low heart rate data set for the monitoring period, and the segmented heart rate data on the right side of the two segmented heart rate data and all heart rate data on the right side in the heart rate ascending sequence constitute a high heart rate data set for the monitoring period; The difference obtained by subtracting the mean of the heart rate monitoring data of the low heart rate data set from the mean of the heart rate monitoring data of the high heart rate data set is used as the heart rate dispersion coefficient of the monitoring period; The absolute value of the difference between the heart rate monitoring data of any two adjacent moments in the monitoring period is obtained, and the product of the average of the absolute value of the difference between the heart rate monitoring data of all two adjacent moments in the monitoring period and the heart rate dispersion coefficient is used as the heart rate change intensity index of the monitoring period.

3. A heart rate data analysis method according to claim 2, characterized in that: The specific method of obtaining the motion transformation index during the monitoring period includes: Arrange all the motion acceleration data in the monitoring period in ascending order to obtain an acceleration ascending sequence, and obtain a low acceleration data set and a high acceleration data set of the monitoring period based on the difference between adjacent motion acceleration data in the acceleration ascending sequence; Obtaining the absolute value of the difference between the minimum value of all motion acceleration data at the corresponding moment in the high acceleration data set and the minimum value of all heart rate monitoring data at the corresponding moment in the high heart rate data set, and recording it as the high heart rate deviation time difference during the monitoring period; Obtaining an absolute value of the difference between the number of motion acceleration data in the high acceleration data set and the number of heart rate monitoring data in the high heart rate data set, and recording the difference as a high heart rate number deviation during the monitoring period; Based on the heart rate change intensity index during the monitoring period, as well as the high heart rate deviation time difference and the high heart rate quantity deviation, the motion transformation index of the monitoring period is obtained; the motion transformation index is positively correlated with the heart rate change intensity index, and negatively correlated with the high heart rate deviation time difference and the high heart rate quantity deviation.

4. A heart rate data analysis method according to claim 1, characterized in that: The specific method of obtaining the mixing degree of each independent component signal includes: The heart rate monitoring signal composed of the heart rate monitoring data during the monitoring period is subjected to ICA independent component analysis to obtain a number of independent component signals; For any independent component signal, a number of extreme value points are obtained, including a number of maximum value points and a number of minimum value points, and the time interval between any two adjacent extreme value points is obtained. The standard deviation of the time interval between all two adjacent extreme value points is used as the extreme value distribution fluctuation coefficient of the independent component signal; Based on the time corresponding to any maximum point and its two adjacent minimum points, the peak area of ​​the maximum point is constructed, and the area of ​​the peak area of ​​the maximum point is obtained by integration; The absolute value of the difference between the areas of the peak regions of any two maximum points is obtained, and the average of all the obtained absolute values ​​of the area differences multiplied by the extreme value distribution fluctuation coefficient is used as the mixing degree of the independent component signal.

5. The heart rate data analysis method according to claim 1, characterized in that: The specific method of obtaining the possibility of interference sources of each independent component signal includes: The product of the mixing degree of any independent component signal and the motion transformation index of the monitoring period is used as the interference source possibility of the independent component signal.

6. A heart rate data analysis method according to claim 1, characterized in that: The specific method of screening a plurality of interference source component signals according to the possibility of the interference source includes: Arrange all independent component signals in ascending order of their interference source possibilities to obtain a component signal sequence, obtain the difference between the interference source possibilities of adjacent independent component signals in the component signal sequence, and use the right independent component signal of the two independent component signals corresponding to the maximum value of the difference between all the interference source possibilities as the interference source component signal; The other independent component signals on the right side of the interference source component signal in the component signal sequence are taken as interference source component signals.

7. A heart rate data analysis method according to claim 1, characterized in that: The specific method of obtaining the frequency decoupling degree of each interference source component signal includes: Based on the instantaneous frequency performance of each IMF component signal and the order value of the IMF component signal, the decomposition disorder degree of each interference source component signal is obtained; Calculate the quality index of each data point in any IMF component signal of any interference source component signal, and obtain the average of the quality indexes of all data points in the IMF component signal as the periodic stability coefficient of the IMF component signal; Based on the decomposition chaos degree of the interference source component signal and the cyclic stability mean of all IMF component signals of the interference source component signal, the frequency decoupling degree of the interference source component signal is obtained. The frequency decoupling degree is negatively correlated with the decomposition chaos degree and positively correlated with the cyclic stability mean.

8. A heart rate data analysis method according to claim 7, characterized in that: The specific method for obtaining the decomposition disorder degree of each interference source component signal is as follows: Performing Fourier transform on any IMF component signal of any interference source component signal to obtain several instantaneous frequencies of the IMF component signal, and taking the mean of all instantaneous frequencies as the frequency representation value of the IMF component signal; Arranging the frequency representation values ​​of each IMF component signal in the interference source component signal according to the order of the IMF component signal to obtain a frequency representation sequence of the interference source component signal, and arranging the order values ​​of each IMF component signal to obtain an order value sequence of the interference source component signal; The Pearson correlation coefficient between the frequency representation sequence and the order value sequence is obtained, and normalized based on the value range of the Pearson correlation coefficient, and the normalized result is used as the decomposition chaos degree of the interference source component signal.

9. The heart rate data analysis method according to claim 1, characterized in that: The specific method for obtaining the preprocessing requirement of each interference source component signal is as follows: The Euclidean norm of the inversely proportional normalized result of the frequency decoupling degree of any interference source component signal and the normalized result of the interference source possibility is used as the preprocessing requirement of the interference source component signal.

10. A heart rate data analysis system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the heart rate data analysis method according to any one of claims 1 to 9 are implemented.

Images