A heart rate detection method based on wearable sensors

Abstract

This application relates to the field of data processing, and specifically provides a heart rate detection method based on wearable sensors, including: calculating a first possibility that all data points in the respiratory data set initially belong to noise points; calculating a second possibility that all data points in the respiratory data set belong to the Cheyne-Stokes respiration model; wherein, the Cheyne-Stokes respiration model characterizes the respiratory characteristics of heart failure patients during normal breathing; determining a final possibility that a data point belongs to a noise point based on the first possibility and the second possibility; determining whether a data point is noise data based on the final possibility; if the data point is a noise data point, removing the noise data point from the respiratory data set, and using the remaining data points to determine the heart rate change of the heart failure patient. Specifically, this method can eliminate abnormal breathing and perform more accurate real-time monitoring of the heart rate.

Description

Technical Field

[0001] This application relates to the field of data processing, and particularly to a heart rate detection method based on wearable sensors. Background Art

[0002] For patients with heart failure, real-time heart rate monitoring is of great significance because it can provide key health status information, help adjust treatment plans in a timely manner, and prevent the deterioration of the condition. Heart rate monitoring based on wearable sensors provides a continuous and non-invasive health management method for heart failure patients.

[0003] Generally, wearable smartwatches often use PPG (Pulse Wave Continuous Blood Pressure Monitoring) technology to measure the change in blood volume using optical sensors, thereby real-time monitoring the heart rate of heart failure patients.

[0004] However, the accuracy of real-time heart rate monitoring of heart failure patients by PPG technology is affected by factors such as the amplitude of movement and improper wearing, resulting in inaccurate monitoring results. Summary of the Invention

[0005] To solve the above problems, this application provides a heart rate detection method based on wearable sensors, which can eliminate abnormal respiratory data (noise data) and perform more accurate real-time monitoring of the heart rate.

[0006] To solve the above problems, this application provides a heart rate detection method based on wearable sensors, including:

[0007] Calculating the first possibility that all data points in the respiratory data set initially belong to noise points;

[0008] Calculating the second possibility that all data points in the respiratory data set belong to the Cheyne-Stokes respiration model; wherein, the Cheyne-Stokes respiration model represents the respiratory characteristics of heart failure patients during normal breathing;

[0009] Determining the final possibility that a data point belongs to a noise point based on the first possibility and the second possibility;

[0010] Determining whether the data point is noise data based on the final possibility;

[0011] If the data point is a noise data point, then removing the noise data point from the respiratory data set and using the remaining data points to determine the heart rate change of the heart failure patient.

[0012] Among them, calculating the first possibility that all data points in the respiratory data set initially belong to noise points includes:

[0013] Calculating a first possibility that a current data point in the respiratory data set initially belongs to a noise point based on the difference in respiratory depth between the current data point and its surrounding data points in the respiratory data set, so as to determine the first possibility that all data points in the respiratory data set initially belong to noise points;

[0014] Among them, the surrounding data points of the current data point include two data points on the left side and two data points on the right side of the current data point.

[0015] Among them, calculating the first possibility that a current data point in the respiratory data set initially belongs to a noise point includes:

[0016] Calculating the first possibility that a current data point in the respiratory data set initially belongs to a noise point by using the following formula:

[0017] ;

[0018] Among them, represents the first possibility that data point O initially belongs to a noise point, represents the respiratory depth of data point O, represents the respiratory depths of two data points on the left side of data point O, represents the respiratory depths of two data points on the right side of data point O, represents the maximum value function.

[0019] Among them, calculating the second possibility that all data points in the respiratory data set belong to the Chebyshev respiratory model includes:

[0020] Determining a third possibility that any two data points in the respiratory data set are adjacent upper and lower vertices of the cosine function corresponding to the Chebyshev respiratory model;

[0021] Based on the respiratory depths of the two data points with the largest third possibility and the time of the two data points with the largest third possibility, determining the cosine function corresponding to the Chebyshev respiratory model;

[0022] Based on the cosine function corresponding to the Chebyshev respiratory model, determining the second possibility that all data points in the respiratory data set belong to the Chebyshev respiratory model.

[0023] Among them, determining the third possibility that any two data points in the respiratory data set are adjacent upper and lower vertices of the cosine function corresponding to the Chebyshev respiratory model includes:

[0024] Determine the preliminary possibility that any two data points are adjacent upper and lower vertices of the cosine function corresponding to the Cheyne-Stokes respiration model based on the curvature radius direction of any two data points in the respiration data set, the number of data points between the any two data points, the first possibility that the data points between the any two data points are initially noise points, the number of data points with the same numerical values as the any two data points respectively, and the first weight correction coefficient;

[0025] Use the second weight correction coefficient to correct the preliminary possibility, so as to obtain the third possibility that the any two data points are adjacent upper and lower vertices of the cosine function corresponding to the Cheyne-Stokes respiration model;

[0026] Wherein, the first weight correction coefficient is determined based on the curvature radius direction of the i-th data point between the any two data points; the second weight correction coefficient is determined based on the derivative value of the linear function of the time interval constructed by the any two data points.

[0027] Wherein, the method includes:

[0028] If the curvature radius direction of the i-th data point between the any two data points is within then the first weight correction coefficient is 1, otherwise the first weight correction coefficient is 0;

[0029] If the derivative value of the linear function of the time interval constructed by the any two data points is less than 0, then the second weight correction coefficient is 1, if the derivative value of the linear function of the time interval constructed by the any two data points is greater than or equal to 0, then the second weight correction coefficient is 0.

[0030] Wherein, determining the preliminary possibility that the any two data points are adjacent upper and lower vertices of the cosine function corresponding to the Cheyne-Stokes respiration model includes:

[0031] Use the following formula to determine the preliminary possibility:

[0032] ;

[0033] ;

[0034] ;

[0035] Wherein, represents the preliminary possibility that any two data points OA and OB are adjacent upper and lower vertices of the cosine function corresponding to the Cheyne-Stokes respiration model, represents the exponential function with the natural constant e as the base, max represents taking the maximum value, represents the curvature radius direction of data point OA, Indicates the curvature radius direction of data point OB, Indicates the number of data points with the same value as data point OA, Indicates the number of data points with the same value as data point OB, Indicates the first possibility that the th data point between data points OA and OB is preliminarily a noise point, m represents the number of data points between data points OA and OB, Indicates the first weight correction coefficient of the first possibility that the th data point between data points OA and OB is preliminarily a noise point, Indicates the relationship rationality under the angle rationality that any two data points OA and OB are adjacent upper and lower vertices of the corresponding cosine function, Indicates the angle rationality that any two data points OA and OB are adjacent upper and lower vertices of the corresponding cosine function.

[0036] Among them, determining the cosine function corresponding to the Chebyshev breathing model based on the breathing depths of the two data points with the largest third possibility and the time of the two data points with the largest third possibility includes:

[0037] Determining the amplitude of the cosine function by the breathing depth difference of the two data points with the largest third possibility, and determining the period of the cosine function by the time difference of the two data points with the largest third possibility;

[0038] Constructing a reference formula using the amplitude of the cosine function, the period of the cosine function, the breathing depths of the two data points with the largest third possibility, and the time of the two data points with the largest third possibility;

[0039] Obtaining the phase parameter and offset parameter of the cosine function corresponding to the Chebyshev breathing model based on the reference formula;

[0040] Constructing the cosine function corresponding to the Chebyshev breathing model based on the phase parameter and the offset parameter.

[0041] Among them, determining the second possibility that all data points in the breathing data set belong to the Chebyshev breathing model based on the cosine function corresponding to the Chebyshev breathing model includes:

[0042] Determining the fourth possibility that the current data point in the breathing data set is preliminarily a Chebyshev breathing model based on the cosine function corresponding to the Chebyshev breathing model, so as to determine the fourth possibility that all data points are preliminarily a Chebyshev breathing model;

[0043] Determining the correction coefficient of the fourth possibility based on the moment corresponding to the current data point and the moments corresponding to the left and right adjacent data points of the current data point;

[0044] Correct the second possibility based on the correction coefficient of the fourth possibility, so as to obtain the second possibility that the current data point belongs to the Chebyshev respiration model, and thus determine the second possibility that all data points belong to the Chebyshev respiration model.

[0045] Construct a cosine function corresponding to the Chebyshev respiration model based on the phase parameter and the offset parameter, including:

[0046] The cosine function corresponding to the Chebyshev respiration model is:

[0047] ;

[0048] where represents the absolute value of the difference between the respiration depth of data point A and the respiration depth of data point B, represents the time of data point A and the time of data point B the absolute value of the time difference, T represents the time of any data point, M represents the respiration depth of the data point calculated by substituting the time of any data point into the cosine function corresponding to the Chebyshev respiration model, represents the phase parameter, represents the offset parameter.

[0049] Different from the prior art, a heart rate detection method based on a wearable sensor provided by the present application includes calculating the first possibility that all data points in the respiration data set initially belong to noise points; calculating the second possibility that all data points in the respiration data set belong to the Chebyshev respiration model; wherein, the Chebyshev respiration model characterizes the respiration characteristics of heart failure patients during normal breathing; determining the final possibility that a data point belongs to a noise point based on the first possibility and the second possibility; determining whether the data point is noise data based on the final possibility; if the data point is a noise data point, removing the noise data point from the respiration data set, and using the remaining data points to determine the heart rate change of the heart failure patient. Specifically, this method can eliminate abnormal respiration data (noise data) in the respiration data set and perform more accurate real-time monitoring of the heart rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings, where:

[0051] Figure 1Schematic diagram of the first embodiment of a heart rate detection method based on a wearable sensor according to the present invention;

[0052] Figure 2 A scatter plot of data monitored for a preset time;

[0053] Figure 3 is Figure 1 Schematic diagram of the process of an embodiment of step S12 in;

[0054] Figure 4 A scatter plot of the time intervals of all data points between adjacent upper and lower vertices of the cosine function provided by this application;

[0055] Figure 5 is Figure 3 Schematic diagram of the process of an embodiment of step S22 in. Detailed implementation manners

[0056] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0057] When generally using PPG technology to perform real-time heart rate monitoring on heart failure patients, affected by factors such as the amplitude of movement and improper wearing, the monitoring results will be inaccurate. Specifically, when generally using a smart watch to perform real-time heart rate monitoring on heart failure patients, the accelerometer in the smart watch is specifically used to monitor the minute movements of the wrist for monitoring the depth of breathing. However, due to factors such as the amplitude of movement and improper wearing, some noise data will appear, and these noise data will cause the monitoring results to be inaccurate. The present application further determines the normal breathing data of heart failure patients through the distinguishing features between the noise data and the normal breathing data of heart failure patients, and finally monitors the heart rate through respiratory sinus arrhythmia (RSA) to determine the heart rate change of heart failure patients.

[0058] Specifically, please refer to Figure 1 , which is a schematic diagram of the process of the first embodiment of a heart rate detection method based on a wearable sensor according to the present invention. The method includes:

[0059] Step S11: Calculate the first possibility that all data points in the breathing data set initially belong to noise points.

[0060] First, the breathing data monitored by the accelerometer in the smart watch is divided into groups at a preset time, such as five minutes, ten minutes, etc. The depth of the data monitored within each preset time (i.e., the depth of breathing) is plotted as Figure 2The scatter plot shown. In the scatter plot, the time intervals between two adjacent data points may be different because the time intervals of pulse beats are different.

[0061] Each set of collected respiratory data is defined as a respiratory data set. In a respiratory data set, there is normal respiratory data of the heart failure patients collected, and there is also abnormal noise data caused by excessive movement and improper wearing. The abnormal noise data will affect the accuracy of heart rate monitoring of heart failure patients. Therefore, the noise data in the respiratory data set needs to be removed.

[0062] In one embodiment, step S11 specifically includes: calculating the first possibility that the current data point in the respiratory data set initially belongs to a noise point based on the difference in respiratory depth between the current data point and the surrounding data points of the current data point in the respiratory data set, so as to determine the first possibility that all data points in the respiratory data set initially belong to noise points. Among them, the surrounding data points of the current data point include two data points on the left side and two data points on the right side of the current data point.

[0063] Specifically, since the characteristic of noise is the undulation of respiratory depth with the surrounding data points, the first possibility that a data point initially belongs to a noise point can be judged according to the undulation degree between the data point and the surrounding data points. Taking the data point O as an example, it can be calculated according to the difference in respiratory depth between the data point O and the four surrounding data points (two on the left side and two on the right side) of the data point O.

[0064] In the embodiment of the present invention, a specific analysis is carried out on two data points on the left side and two data points on the right side of the data point O. Of course, in some other embodiments of the present invention, it is also possible to directly use the four data points closest to the data point O, and there is no limitation on this.

[0065] In a specific embodiment, the following formula is used to calculate the first possibility that the current data point in the respiratory data set initially belongs to a noise point:

[0066] ;

[0067] Among them, represents the first possibility that the data point O initially belongs to a noise point, represents the respiratory depth of the data point O, represents the respiratory depth of the two data points on the left side of the data point O, represents the respiratory depth of the two data points on the right side of the data point O, represents the maximum value function.

[0068] For the data point O, the greater the fluctuation in the respiratory depth between it and the surrounding data points, the more likely it is to be initially classified as a noise point, showing a positive proportional relationship. Since there may be noise points around normal heart failure data points, affecting their degree of fluctuation, and noise points cannot continuously appear around normal heart failure data points, here two data points are selected on each of the left and right sides, the maximum difference is removed and the average value is calculated, so as to calculate the degree of fluctuation between the data point O and the surrounding pixel points. This degree of fluctuation is used to characterize the first possibility that the data point O is initially a noise point.

[0069] Step S12: Calculate the second possibility that all data points in the respiratory data set belong to the Chebyshev respiratory model; among them, the Chebyshev respiratory model represents the respiratory characteristics of heart failure patients during normal breathing.

[0070] Specifically, the respiratory characteristics of heart failure patients during normal breathing follow the Chebyshev respiratory model. In this breathing mode, the respiratory depth also fluctuates. Therefore, the fluctuation of the respiratory depth alone cannot be used as the basis for determining noise data. Since the characteristic of the fluctuating respiratory depth is not only a characteristic of noise data but also a characteristic that appears in the Chebyshev respiratory mode, in this step, a feature analysis of the Chebyshev respiratory model is required. In the Chebyshev respiratory model, as the respiratory frequency increases, the respiratory depth first becomes deeper and then shallower, then stagnates for a period of time, and then becomes deeper and shallower again in turn, specifically showing the characteristics of a periodic pattern of a cosine function. Therefore, the second possibility that all data points belong to the Chebyshev respiratory model can be calculated, and this second possibility is used to correct the weight of the first possibility that the obtained data points are initially noise data points, so as to obtain the final possibility that all data points are noise data points.

[0071] Specifically, please combine Figure 3 , step S12 specifically includes:

[0072] Step S21: Determine the third possibility that any two data points in the respiratory data set are adjacent upper and lower vertices of the cosine function corresponding to the Chebyshev respiratory model.

[0073] First of all, since the Chebyshev respiratory model all shows the characteristics of a periodic pattern of a cosine function where the respiratory depth first becomes deeper and then shallower, then stagnates for a period of time, and then appears to become deeper and then shallower and then pause for a period of time, the second possibility can be calculated by first finding the cosine function of the Chebyshev respiratory model.

[0074] Specifically, based on the curvature radius direction of any two data points in the respiratory data set, the number of data points between any two data points, the first possibility that the data points between any two data points are initially noise points, the number of data points with the same values as any two data points respectively, and the first weight correction coefficient, determine the preliminary possibility that any two data points are adjacent upper and lower vertices of the cosine function corresponding to the Chebyshev respiratory model.

[0075] For all data points on the respiratory data, select the data points closest to the 90-degree and 270-degree directions in the direction of the radius of curvature of all data points. Moreover, the range of the radius of curvature directions of all data points between these two vertices lies between these two points, and the respiratory rate also increases. To calculate the curvature of each data point, the quadratic fitting of the curve can be first performed through each data point and its two adjacent data points on the left and right to obtain the radius of curvature directions of all data points.

[0076] In a specific embodiment, the following formula is used to determine the preliminary possibility that any two data points are adjacent upper and lower vertices of the cosine function corresponding to the Cheyne-Stokes respiration model:

[0077] ;

[0078] ;

[0079] ;

[0080] where, represents the preliminary possibility that any two data points OA and OB are adjacent upper and lower vertices of the cosine function corresponding to the Cheyne-Stokes respiration model, represents the exponential function with the natural constant e as the base, and max represents taking the maximum value, represents the radius of curvature direction of data point OA, represents the radius of curvature direction of data point OB, represents the number of data points with the same value as data point OA, represents the number of data points with the same value as data point OB, represents the first possibility that the j-th data point between data points OA and OB is preliminarily a noise point, m represents the number of data points between data points OA and OB, represents the first weight correction coefficient of the first possibility that the j-th data point between data points OA and OB is preliminarily a noise point, represents the relationship rationality under the angle rationality that any two data points OA and OB are adjacent upper and lower vertices of the corresponding cosine function, represents the angle rationality that any two data points OA and OB are adjacent upper and lower vertices of the corresponding cosine function.

[0081] It should be noted that in the above formula, the first weight correction coefficient is determined based on the radius of curvature direction of the j-th data point between any two data points. Specifically, if the radius of curvature direction of the j-th data point between any two data points, such as data points OA and OB, is in If it is between them, the first weight correction coefficient is 1; otherwise, the first weight correction coefficient is 0. It is expressed by the formula as:

[0082] ;

[0083] It represents the curvature radius direction of the j-th data point between data points OA and OB.

[0084] For the adjacent upper and lower vertices of the cosine function, one curvature radius direction approaches 90° and one curvature radius direction approaches 270°. Therefore and The difference between and and The larger the absolute value of the difference between and, the greater the initial possibility of belonging to adjacent upper and lower vertices, showing an inverse proportional relationship. For the cosine function, the upper and lower vertices correspond to each other. The number of upper vertices should be the same as the number of lower vertices, and the difference is small. Therefore, the smaller the difference in the number of data points with the same values as data points OA and OB on the time axis, the greater the possibility of belonging to adjacent upper and lower vertices, showing an inverse proportional relationship. In addition, for all data points between OA and OB, the possibility of showing suspected noise should be greater, that is The larger it is, because adjacent upper and lower vertices need to be selected, and the curvatures of the upper and lower vertices are both between these two curvatures. And there are two possible points that must not be within these two curvatures. One is that there are multiple periods within the interval, resulting in not being within the interval, and the other means that if there are abnormal points, it may also cause not to be within the angle interval of the curvature radius. Only by selecting the points located between the curvatures of OA and OB and constructing The formula, the average value can be calculated again, that is , the larger this average value is, the greater the possibility of belonging to adjacent upper and lower vertices, showing a direct proportional relationship with .

[0085] Specifically, for the adjacent upper and lower vertices OA and OB of the cosine function, as the Cheyne-Stokes respiration model changes with time, the respiration frequency should increase. Therefore, a scatter plot can be made according to the respiration intervals from left to right. The time difference between each data point and its next data point is recorded as the time interval of each data point. By obtaining the time interval curve of all data points between OA and OB, the scatter plot of the time intervals of all data points between the adjacent upper and lower vertices of the cosine function as shown in Figure 4 is obtained. The overall respiration frequency between points OA and OB can be determined to be increasing or decreasing according to the derivative after linear fitting of these functions.

[0086] The following formula can be constructed to perform a weight correction on the preliminary possibility of the adjacent upper and lower vertices of the cosine function of the Chebyshev breathing model for data points OA and OB through the increase and decrease of the breathing frequency. Specifically, the second weight correction coefficient is used to correct the preliminary possibility, so as to obtain the third possibility that any two data points are the adjacent upper and lower vertices of the corresponding cosine function of the Chebyshev breathing model. The second weight correction coefficient is determined based on the derivative value of the linear function of the time interval constructed by any two data points. In a specific embodiment, if the derivative value of the linear function of the time interval constructed by the any two data points is less than 0, the second weight correction coefficient is 1; if the derivative value of the linear function of the time interval constructed by the any two data points is greater than or equal to 0, the second weight correction coefficient is 0. The formula of the second weight correction coefficient is expressed as:

[0087] ;

[0088] wherein, represents the second weight correction coefficient, represents the derivative value of the linear function of the time interval constructed by data points OA and OB.

[0089] After determining the second weight correction coefficient, the second weight correction coefficient is used to correct the preliminary possibility, so as to obtain the third possibility that any two data points are the adjacent upper and lower vertices of the corresponding cosine function of the Chebyshev breathing model. Since the breathing frequency within the cosine function constructed by the Chebyshev breathing model increases between adjacent upper and lower vertices, a linear function can be constructed according to the time interval between adjacent data points. If the derivative of the linear function is less than 0, it proves that the breathing frequency is increasing; if the derivative of the linear function is greater than or equal to 0, it proves that the breathing frequency is decreasing. Based on this, a weight correction is made on the possibility that data points OA and OB are the adjacent upper and lower vertices of the cosine function of the Chebyshev breathing model. The calculation formula for constructing the third possibility is: .

[0090] wherein: represents the third possibility that data points OA and OB are the adjacent upper and lower vertices of the corresponding cosine function of the Chebyshev breathing model, represents the preliminary possibility that data points OA and OB are the adjacent upper and lower vertices of the corresponding cosine function of the Chebyshev breathing model.

[0091] Step S22: Determine the cosine function corresponding to the Chebyshev breathing model based on the breathing depth of the two data points with the greatest third possibility and the time of the two data points with the greatest third possibility.

[0092] Combined with Figure 5 , step S22 specifically includes:

[0093] Step S41: Determine the breathing depth difference between the two data points with the third greatest probability as the amplitude of the cosine function, and determine the period of the cosine function using the time difference between the two data points with the third greatest probability.

[0094] Specifically, the two data points with the third largest probability are recorded as points A and B, and the breathing depth of data point A is expressed as , the breathing depth of data point B is expressed as , the breathing depth is different That is, the amplitude of the cosine function, the time difference between data point A and data point B That is one quarter of the period of the cosine function, which is used to determine the amplitude and period of the cosine function.

[0095] Step S42: construct a reference formula using the amplitude of the cosine function, the period of the cosine function, the breathing depth of the two data points with the third greatest probability, and the time of the two data points with the third greatest probability.

[0096] Specifically, , Substituting into the cosine function formula, α represents the amplitude of the cosine function, and β represents the period of the cosine function, that is, the following reference formula can be constructed by combining them:

[0097]

[0098]

[0099]

[0100] Step S43: Obtaining the phase parameter and offset parameter of the cosine function corresponding to the Che-Kessler breathing model based on the reference formula.

[0101] By calculating the above reference formula, the phase parameter of the cosine function corresponding to the Che-Kessler breathing model can be obtained: , offset parameter .

[0102] Step S44: constructing a cosine function corresponding to the Che-Kessler breathing model based on the phase parameter and the offset parameter.

[0103] Specifically, the phase parameter , offset parameter Substituting into the cosine function, the cosine function corresponding to the Che-Kessler breathing model can be constructed as:

[0104] ;

[0105] in, Indicates the breathing depth of data point A The absolute value of the difference in respiratory depth from data point B Indicates the time of data point A And the time of data point B The absolute value of the time difference, T represents the time of any data point, M represents the respiratory depth of the data point calculated by substituting the time of any data point into the cosine function corresponding to the Che-Sarkar respiratory model, α represents the amplitude of the cosine function, and β represents the period of the cosine function.

[0106] Step S23: Determine the second possibility that all data points in the respiratory data set belong to the Che-Sarkar respiratory model based on the cosine function corresponding to the Che-Sarkar respiratory model.

[0107] In a specific embodiment, based on the cosine function corresponding to the Che-Sarkar respiratory model, determine the fourth possibility that the current data point in the respiratory data set preliminarily belongs to the Che-Sarkar respiratory model, so as to determine the fourth possibility that all data points preliminarily belong to the Che-Sarkar respiratory model.

[0108] Specifically, for data point O, substitute the time To where data point O is located into the above cosine function corresponding to the Che-Sarkar respiratory model, replace T in the cosine function, and calculate the respiratory depth M of data point O To .

[0109] Based on the calculated respiratory depth M of data point O To And the respiratory depth of data point O in the respiratory data set Calculate the fourth possibility that data point O preliminarily belongs to the Che-Sarkar respiratory model. In one embodiment, the formula for the fourth possibility is:

[0110] ;

[0111] Wherein,[[]] Represents the fourth possibility that the data point Preliminarily belongs to the Che-Sarkar respiratory model,[[]] Represents a hyperparameter such that the denominator is not zero. For example,[[]] Can be 0.01. In the formula,[[]] Represents the fitting error between data point O and the cosine function. The larger the error, the lower the possibility of the data point belonging to the Che-Sarkar respiratory model, showing an inverse proportional relationship.

[0112] Furthermore, based on the characteristic of the increasing respiratory frequency in the Che-Sarkar respiratory model, determine the correction coefficient of the fourth possibility based on the moment corresponding to the current data point and the moments corresponding to the current data point and its left and right adjacent data points. Specifically, the formula for the correction coefficient of the fourth possibility is expressed as:[[]]

[0113] ;

[0114] Among them, represents the correction coefficient corresponding to the fourth possibility of the data point O. represents the time corresponding to the data point O, represents the time corresponding to the data point adjacent to the left of the data point O, represents the time corresponding to the data point adjacent to the right of the data point O.

[0115] Based on the correction coefficient of the fourth possibility, correct the second possibility, so as to obtain the second possibility that the current data point belongs to the Chebyshev breathing model, and thus determine the second possibility that all data points belong to the Chebyshev breathing model.

[0116] In a specific embodiment, the calculation formula for the second possibility that the current data point belongs to the Chebyshev breathing model is: ; Among them, represents the data point the second possibility of belonging to the data point of the Chebyshev breathing model, represents the data point the fourth possibility of initially belonging to the data point of the Chebyshev breathing model, represents a correction coefficient for the fourth possibility.

[0117] Since only reflects whether the data point O is on the cosine function of the Chebyshev breathing model and cannot reflect whether it conforms to the characteristic of the accelerating breathing frequency of the Chebyshev breathing model, so construct to correct whether its breathing frequency is accelerating. Only if it is accelerating can it possibly belong to the Chebyshev breathing model.

[0118] Step S13: Determine the final possibility that the data point belongs to the noise point based on the first possibility and the second possibility.

[0119] Specifically, perform normalization processing on the second possibility Q that all data points O belong to the Chebyshev breathing model: .

[0120] Use the normalized second possibility and the first possibility to determine the final possibility that the data point belongs to the noise point. Specifically:

[0121] ;

[0122] Among them, represents the final possibility that the data point O belongs to the noise point, represents the data point the first possibility of initially belonging to the noise data, is the second possibility that the data point O belongs to the data points of the Cheyne-Stokes respiration model, represents the absolute value of the natural logarithm function.

[0123] Since the greater the possibility that the data point O belongs to the Cheyne-Stokes respiration model, the lower the weight that it belongs to a noise point should be, and the smaller the possibility that it belongs to the Cheyne-Stokes respiration model, the higher the weight that it belongs to a noise point should be. It is planned to use the function to perform an amplified differentiation process on this result, the closer the function is to 1, the closer the weight is to 0; the closer it is to 0, the greater the weight.

[0124] For normal heart failure patients, the breathing is relatively stable, the breathing depth is consistent, presenting a Cheyne-Stokes respiration pattern. In this pattern, the breathing rate increases, the breathing depth first becomes deeper and then shallower, then stagnates for a period of time, and then appears a pattern of first becoming deeper, then shallower, and then pausing for a period of time. Based on these characteristics, the influence of noise data caused by movement amplitude and improper wearing can be excluded first, and then the heart rate of heart failure patients can be monitored through the normal breathing data set of heart failure patients by respiratory sinus arrhythmia (RSA) to perform more accurate real-time heart rate monitoring on heart failure patients.

[0125] Step S14: Determine whether the data point is noise data based on the final possibility.

[0126] Perform maximum-minimum normalization processing on the final possibility that all the above data points belong to noise first, and then set a threshold. In this application, taking the threshold as 0.8 as an example, if it exceeds this threshold, it is considered noise data.

[0127] Step S15: If the data point is a noise data point, remove the noise data point from the breathing data set, and use the remaining data points to determine the heart rate change of the heart failure patient.

[0128] The noise data caused by movement amplitude and improper wearing has been determined. By removing this noise data, the breathing data set showing normal heart failure of each group of heart failure patients within five minutes can be obtained.

[0129] Furthermore, the heart rate of heart failure patients can be monitored more accurately through the normal breathing data set of heart failure patients by respiratory sinus arrhythmia (RSA). Through real-time heart rate monitoring, the change of heart rate can be observed. An increase in heart rate may indicate an increase in cardiac load or a problem in the cardiovascular system, while a decrease in heart rate may be related to a decline in cardiac function. By monitoring the change trend of heart rate, the health of heart failure patients can be preliminarily judged.

[0130] The method of the present application monitors the real-time heart rate by determining the respiratory data set showing normal heart failure in heart failure patients. This technology can avoid the influence of noise data caused by movement amplitude and improper wearing in heart rate monitoring by the PPG technology, making the real-time monitoring more accurate. Because the respiration of normal heart failure patients has a certain depth otherwise, or presents a Cheyne-Stokes respiration pattern, with an increased respiratory rate and a periodic pattern of first deep, then shallow, and then pausing for a period of time in the depth of respiration. Based on these characteristics, noise data is judged and removed to obtain normal respiratory data for more effective monitoring of the heart rate condition of heart failure patients.

[0131] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0132] Each embodiment in this specification is described in a progressive manner, and the same or similar parts among the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A heart rate detection method based on a wearable sensor, characterized in that: include: Calculate the first probability that all data points in the respiratory data set are initially noise points; Calculating a second probability that all data points in the respiratory data set belong to a Che-Kessler respiratory model; wherein the Che-Kessler respiratory model characterizes the respiratory characteristics of a heart failure patient during normal breathing; Determine a final probability that the data point belongs to a noise point based on the first probability and the second probability; Determining whether the data point is noise data based on the final probability; If the data point is a noise data point, the noise data point is removed from the respiratory data set, and the remaining data points are used to determine the heart rate change of the heart failure patient.

2. A heart rate detection method based on a wearable sensor according to claim 1, characterized in that: Calculate the first possibility that all data points in the respiratory data set are initially noise points, including: Calculating a first possibility that the current data point in the respiratory data set preliminarily belongs to a noise point based on a difference in respiratory depth between the current data point and data points surrounding the current data point in the respiratory data set, thereby determining a first possibility that all data points in the respiratory data set preliminarily belong to noise points; The surrounding data points of the current data point include two data points on the left side of the current data point and two data points on the right side of the current data point.

3. A heart rate detection method based on a wearable sensor according to claim 2, characterized in that: Calculating a first possibility that a current data point in the respiratory data set is initially a noise point comprises: The first possibility that the current data point in the respiratory data set is initially a noise point is calculated using the following formula: ； in, Indicates the first possibility that the data point O is initially a noise point, represents the breathing depth of data point O, represents the breathing depth of the two data points to the left of data point O, represents the breathing depth of the two data points to the right of data point O, Represents the maximum value function.

4. A heart rate detection method based on a wearable sensor according to claim 1, characterized in that: Calculating a second likelihood that all data points in the respiratory data set belong to the Che-Kessler respiratory model comprises: Determine a third possibility that any two data points in the respiratory data set are adjacent upper and lower vertices of a cosine function corresponding to a Che-Kessler respiratory model; Determine a cosine function corresponding to the Che-Kessler breathing model based on the breathing depth of the two data points with the third greatest probability and the time of the two data points with the third greatest probability; A second possibility that all data points in the respiratory data set belong to the Che-Kerk respiratory model is determined based on a cosine function corresponding to the Che-Kerk respiratory model.

5. A heart rate detection method based on a wearable sensor according to claim 4, characterized in that: Determining that any two data points in the respiratory data set are adjacent upper and lower vertices of the cosine function corresponding to the Che-Kessler respiratory model is a third possibility, including: Determine the preliminary possibility that the arbitrary two data points are adjacent upper and lower vertices of the cosine function corresponding to the Che-Kessler breathing model based on the direction of the curvature radius of any two data points in the respiratory data set, the number of data points between the arbitrary two data points, the first possibility that the data points between the arbitrary two data points are initially noise points, the number of data points having the same value as the arbitrary two data points, and the first weight correction coefficient; Correcting the preliminary possibility using a second weight correction coefficient, thereby obtaining a third possibility that the arbitrary two data points are adjacent upper and lower vertices of the cosine function corresponding to the Che-Kessler breathing model; Among them, the first weight correction coefficient is determined based on the direction of the curvature radius of the i-th data point between the any two data points; the second weight correction coefficient is determined based on the derivative value of the linear function of the time interval constructed by the any two data points.

6. A heart rate detection method based on a wearable sensor according to claim 5, characterized in that: The method comprises: If the curvature radius of the i-th data point between any two data points is in the direction If , the first weight correction coefficient is 1, otherwise, the first weight correction coefficient is 0; If the derivative value of the linear function of the time interval constructed by any two data points is less than 0, the second weight correction coefficient is 1; if the derivative value of the linear function of the time interval constructed by any two data points is greater than or equal to 0, the second weight correction coefficient is 0.

7. A heart rate detection method based on a wearable sensor according to claim 5, characterized in that: Determining the preliminary possibility that the two arbitrary data points are adjacent upper and lower vertices of the cosine function corresponding to the Che-Kessler breathing model includes: The preliminary probability is determined using the following formula: ； ； ； in, It indicates the preliminary possibility that any two data points OA and OB are adjacent upper and lower vertices of the cosine function corresponding to the Che-Schmidt breathing model. represents an exponential function with the natural constant e as the base, and max represents the maximum value. Indicates the direction of the curvature radius of the data point OA, Indicates the direction of the curvature radius of the data point OB, Indicates the number of data points with the same value as data point OA, Indicates the number of data points with the same value as data point OB. represents the first possibility that the jth data point between data points OA and OB is a noise point, m represents the number of data points between data points OA and OB, The first weight correction coefficient representing the first possibility that the jth data point between data points OA and OB is initially a noise point, It means that any two data points OA and OB are the corresponding upper and lower vertices of the cosine function under the reasonableness of the angle relationship. It indicates the reasonableness of the angle between any two data points OA and OB, which are the adjacent upper and lower vertices of the corresponding cosine function.

8. A heart rate detection method based on a wearable sensor according to claim 4, characterized in that: Determining a cosine function corresponding to the Che-Kessler breathing model based on the breathing depth of the two data points with the third greatest probability and the time of the two data points with the third greatest probability includes: Determine the breathing depth difference between the two data points with the third greatest probability as the amplitude of the cosine function, and determine the period of the cosine function using the time difference between the two data points with the third greatest probability; Constructing a reference formula using the amplitude of the cosine function, the period of the cosine function, the breathing depth of the two data points with the third greatest probability, and the time of the two data points with the third greatest probability; Based on the reference formula, a phase parameter and an offset parameter of a cosine function corresponding to the Che-Kessler breathing model are obtained; A cosine function corresponding to the Che-Kerk breathing model is constructed based on the phase parameter and the offset parameter.

9. A heart rate detection method based on a wearable sensor according to claim 4, characterized in that: Determining a second possibility that all data points in the respiratory data set belong to the Che-Kerk respiratory model based on the cosine function corresponding to the Che-Kerk respiratory model includes: Determining a fourth possibility that the current data point in the respiratory data set preliminarily belongs to the Che-Kessler respiratory model based on the cosine function corresponding to the Che-Kessler respiratory model, thereby determining a fourth possibility that all data points preliminarily belong to the Che-Kessler respiratory model; Determine the fourth possible correction coefficient based on the time corresponding to the current data point and the time corresponding to the current data point and the left and right adjacent data points; The second possibility is corrected based on the correction coefficient of the fourth possibility to obtain the second possibility that the current data point belongs to the Che-Kessler breathing model, thereby determining the second possibility that all data points belong to the Che-Kessler breathing model.

10. A heart rate detection method based on a wearable sensor according to claim 8, characterized in that: The cosine function corresponding to the Che-Kerk breathing model is constructed based on the phase parameter and the offset parameter, including: The cosine function corresponding to the Che-Schmidt breathing model is: ； in, Indicates the breathing depth of data point A The absolute value of the difference between the breathing depth at data point B and Indicates the time of data point A The time with data point B The absolute value of the time difference, T represents the time of any data point, M represents the breathing depth of the data point calculated by substituting the time of any data point into the cosine function corresponding to the Che-Kess breathing model, represents the phase parameter, Indicates the offset parameter.

Images