Physiological signal processing method and device, computer equipment and storage medium

Abstract

The invention is suitable for the technical field of signal processing, and provides a physiological signal processing method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a first physiological signal and a first motion signal, and carrying out the iterative optimization of a preset objective function until convergence, according to the first physiological signal, the first motion signal and the scaling vector, a target physiological signal is obtained, so that the influence of motion artifacts on the physiological signal is effectively inhibited, the noiseless target physiological signal is obtained, and a more accurate heart rate monitoring result can be conveniently obtained subsequently.

Description

Technical Field

[0001] The present application belongs to the field of signal processing technology, and in particular, relates to a physiological signal processing method, device, computer equipment and storage medium. Background Art

[0002] The photoplethysmogram (PPG) signal is a periodic signal obtained by recording the periodic changes in blood caused by the periodic beating of the heart through a photoelectric sensor on the wrist. The heart rate can be analyzed and monitored based on this signal. However, in practical applications, the original PPG signal is often interfered by various noises, including low-frequency noise, high-frequency noise, motion artifacts, and baseline drift, especially motion artifacts (MA). For example, in activities such as running, skiing, or mountaineering, these motion artifacts may be very significant, even exceeding the effective PPG signal of the heart rate, resulting in challenges to the accuracy of heart rate monitoring. Summary of the invention

[0003] The embodiments of the present application provide a physiological signal processing method, apparatus, computer equipment and storage medium, which can solve the technical problem in the prior art that the original PPG signal has noise such as motion artifacts, resulting in inaccurate heart rate monitoring.

[0004] In a first aspect, an embodiment of the present application provides a physiological signal processing method, comprising:

[0005] Acquiring a first physiological signal and a first motion signal;

[0006] Iteratively optimizing a preset objective function until convergence, to obtain a scaling vector for the first motion signal;

[0007] A target physiological signal is obtained according to the first physiological signal, the first motion signal and the scaling vector.

[0008] In a possible implementation manner of the first aspect, the acquiring the first physiological signal and the first motion signal also includes:

[0009] acquiring a second physiological signal;

[0010] Processing the first motion signal based on a principal component analysis method to obtain a reference signal;

[0011] The first physiological signal is obtained according to the second physiological signal and the reference signal.

[0012] In a possible implementation manner of the first aspect, obtaining the first physiological signal according to the second physiological signal and the reference signal includes:

[0013] Performing filtering processing on the second physiological signal and the reference signal based on a filter to obtain an output result;

[0014] obtaining an error signal based on the output result and the second physiological signal;

[0015] The filter parameters are adjusted based on the error signal, and when a preset convergence criterion is met, the first physiological signal is output; or the filter parameters and step size parameters are adjusted based on the error signal, and when the preset convergence criterion is met, the first physiological signal is output.

[0016] In a possible implementation manner of the first aspect, the acquiring the second physiological signal also includes:

[0017] acquiring a third physiological signal and a third motion signal;

[0018] The third physiological signal and the third motion signal are band-pass filtered to obtain the second physiological signal and the first motion signal.

[0019] In a possible implementation manner of the first aspect, the objective function includes a fitting function and a cost function; the fitting function includes a scaling vector to be optimized; and the iterative optimization of the preset objective function until convergence to obtain the scaling vector about the first motion signal includes:

[0020] Determine, by iterative optimization, a scaling vector when the cost function takes a minimum value;

[0021] The scaling vector that makes the cost function take the minimum value is used as the scaling vector of the first motion signal.

[0022] In a possible implementation manner of the first aspect, the objective function is constructed based on an asymmetric least squares method and an adaptive iterative reweighting method.

[0023] In a possible implementation manner of the first aspect, obtaining a target physiological signal according to the first physiological signal, the first motion signal, and the scaling vector includes:

[0024] determining a product value of the first motion signal and the scaling vector;

[0025] The difference between the first physiological signal and the product value is used as the target physiological signal.

[0026] In a possible implementation manner of the first aspect, the method further includes:

[0027] Extracting features of the target physiological signal to obtain a spectrum signal;

[0028] Performing spectrum peak tracking on the spectrum signal to obtain a heart rate spectrum peak value;

[0029] A heart rate value is obtained based on the heart rate spectrum peak value.

[0030] In a second aspect, an embodiment of the present application provides a physiological signal processing device, comprising:

[0031] A first acquisition module, used to acquire a first physiological signal and a first motion signal;

[0032] An iterative optimization module, used for iteratively optimizing a preset objective function until convergence, and obtaining a scaling vector for the first motion signal;

[0033] A signal determination module is used to obtain a target physiological signal according to the first physiological signal, the first motion signal and the scaling vector.

[0034] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the physiological signal processing method described in any one of the first aspects above when executing the computer program.

[0035] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the physiological signal processing method described in any one of the first aspects above is implemented.

[0036] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is executed on a computer device, the computer device executes the physiological signal processing method described in any one of the above-mentioned first aspects.

[0037] The beneficial effects of the embodiments of the present application compared with the prior art are: iteratively optimizing a preset objective function until convergence to obtain a scaling vector for the first motion signal, and obtaining a target physiological signal based on the first physiological signal, the first motion signal and the scaling vector, thereby fully utilizing the correlation between the physiological signal and the motion signal, avoiding the phenomenon of introducing negative values ​​and false peaks by directly subtracting the spectral components extracted from the motion signal from the spectrum of the physiological signal, effectively suppressing the influence of motion artifacts on the physiological signal, and obtaining a noise-free target physiological signal, so as to facilitate subsequent more accurate heart rate monitoring results.

[0038] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is a schematic diagram of an application scenario of a physiological signal processing method provided by an embodiment of the present application;

[0041] Figure 2 is a flowchart of a physiological signal processing method provided by an embodiment of the present application;

[0042] Figure 3 It is a flowchart of an AS-LMS filtering method provided by an embodiment of the present application;

[0043] Figure 4 is a flowchart of a physiological signal processing method provided in yet another embodiment of the present application;

[0044] Figure 5 is a structural schematic diagram of a physiological signal processing device provided in one embodiment of the present application;

[0045] Figure 6 It is a structural diagram of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0046] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0047] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0048] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0049] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0050] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0051] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0052] The present application provides a physiological signal processing method which can be applied to Figure 1 In the application environment shown, the smart wearable device 101 is used to obtain physiological signals and motion signals, and send the physiological signals and motion signals to the physiological signal processing device 102 for analysis and processing, so as to determine the target physiological signal.

[0053] Among them, the smart wearable device 101 is a portable electronic device, which is mainly used to monitor various physiological health indicators of the human body. Currently, common smart wearable devices 101 include sports bracelets, smart watches, etc. The smart wearable device 101 is usually connected to a terminal device or a server via Bluetooth, and the collected data is transmitted to the terminal device or the server for analysis and processing.

[0054] Figure 2 A schematic flow chart of a physiological signal processing method provided by the present application is shown.

[0055] S201, acquiring a first physiological signal and a first motion signal.

[0056] The first physiological signal may be a PPG signal. Due to the periodic beating of the heart, the blood presents periodic fluctuations. The photoelectric sensor on the wrist records the periodic changes of the blood. The periodic image depicted by the acquired signal is the PPG signal. The original PPG signal contains a lot of noise, such as low frequency, high frequency, motion artifacts and baseline drift.

[0057] The first motion signal may be an acceleration signal.

[0058] S202 : Iteratively optimize a preset objective function until convergence, and obtain a scaling vector for the first motion signal.

[0059] In the embodiment of the present application, the objective function includes a variable to be optimized, the scaling vector, and the required scaling vector can be obtained by iteratively optimizing and solving the objective function. The scaling vector is used as a coefficient of the first motion signal. By solving the scaling vector, the phenomenon of introducing negative values ​​and false peaks by directly subtracting the spectral components extracted from the motion signal from the spectrum of the physiological signal can be avoided, and the influence of motion artifacts on the physiological signal can be effectively suppressed.

[0060] Optionally, the objective function is pre-constructed based on an asymmetric least squares method (ALS) and an adaptive iterative reweighting method, which can also be called an adaptive asymmetric least squares method.

[0061] In real life, movement not only changes the acceleration signal, but also affects the synchronized PPG signal. Therefore, the spectral components of the MA in the PPG signal are correlated with the spectral components of the acceleration signal. Spectral Subtraction (SS) can be used to remove the MA. However, directly subtracting the spectrum extracted from the acceleration signal from the spectrum of the PPG signal may produce many negative residuals and false peaks, thereby reducing the performance of the subsequent peak selection process.

[0062] In the embodiment of the present application, ALS is used to implement MA filtering. ALS is a global optimization method independent of the reference peak, which can greatly reduce the generation of false peaks in the spectrum. Optionally, ALS may have deviations when there are strong peaks. In order to solve this problem, an adaptive iterative reweighting method can be used to better suppress false peaks.

[0063] S203: Obtain a target physiological signal according to the first physiological signal, the first motion signal, and the scaling vector.

[0064] Among them, the target physiological signal is a valid PPG signal with noise interference such as MA eliminated.

[0065] In the embodiment of the present application, the first physiological signal, the first motion signal and the scaling vector are solved to obtain the target physiological signal.

[0066] In an embodiment of the present application, a preset objective function is iteratively optimized until convergence to obtain a scaling vector for the first motion signal, and a target physiological signal is obtained based on the first physiological signal, the first motion signal and the scaling vector, thereby fully utilizing the correlation between the physiological signal and the motion signal, avoiding the phenomenon of introducing negative values ​​and false peaks by directly subtracting the spectral components extracted from the motion signal from the spectrum of the physiological signal, and effectively suppressing the influence of motion artifacts on the physiological signal, thereby obtaining a noise-free target physiological signal, which is convenient for subsequent more accurate heart rate monitoring results.

[0067] In a possible implementation, the step of acquiring the first physiological signal and the first motion signal also includes: acquiring a second physiological signal; processing the first motion signal based on a principal component analysis method to obtain a reference signal; and obtaining the first physiological signal based on the second physiological signal and the reference signal.

[0068] Among them, the second physiological signal can be a PPG signal that has undergone preliminary noise reduction preprocessing (such as preliminary noise reduction based on a Butterworth bandpass filter). However, there are still some problems in the signal after the preliminary noise reduction preprocessing, such as residual noise. These residual noises may affect the accuracy of subsequent signal analysis, especially in the spectrum analysis of the signal, the residual noise may cause erroneous spectral component identification. Therefore, it is necessary to process the second physiological signal and the first motion signal based on principal component analysis (PCA) to obtain a more accurate and reliable first physiological signal.

[0069] Among them, PCA is a feature dimensionality reduction method that maps the original high-dimensional data to a low-dimensional space through linear transformation, while retaining the main change characteristics and information in the data as much as possible. In signal processing, PCA can be used to extract the most important features in the signal, reduce the impact of noise, and improve the quality of the signal.

[0070] Among them, in PCA, the reference signal refers to the signal component extracted from the original data, which can represent the main change characteristics of the data. These components are the main components obtained by PCA processing. They are linear combinations of the original data and are orthogonal to each other (ie, unrelated). The reference signal can be used for tasks such as signal denoising, dimensionality reduction and reconstruction. In the present application, the above-mentioned reference signal refers to the signal component extracted from the first motion signal, which can represent the main change characteristics of the first motion signal. Based on the reference signal, the noise caused by motion can be more accurately identified, and the reference signal is used to assist in removing motion artifacts in the second physiological signal, so that the final first physiological signal is more suitable for subsequent physiological parameter monitoring and analysis.

[0071] In a possible implementation, the step of acquiring the second physiological signal also includes: acquiring a third physiological signal and a third motion signal; and performing band-pass filtering on the third physiological signal and the third motion signal to obtain the second physiological signal and the first motion signal.

[0072] The third physiological signal is an original physiological signal, and the third motion signal is an original motion signal.

[0073] In the embodiment of the present application, the third physiological signal and the third motion signal are first obtained. Since they are both original signals, there are a lot of noises and noise reduction preprocessing is required.

[0074] Among them, the heart rate of ordinary people (except infants and young children) is between 0.4-3.5Hz, and the frequencies outside 0.4-3.5Hz are interference frequencies. In the embodiment of the present application, a Butterworth bandpass filter is designed to filter the third physiological signal and the third motion signal, and remove the frequencies outside 0.4-3.5Hz in the third physiological signal and the third motion signal, so as to obtain the second physiological signal and the first motion signal after noise reduction. Among them, the Butterworth bandpass filter ensures that the signal frequency of the passband is the smoothest.

[0075] Optionally, after obtaining the second physiological signal and the first motion signal, further noise reduction processing may be performed.

[0076] In the prior art, the acceleration signals in three directions (X, Y, Z) contain a lot of motion noise. The motion noise composed of different periodic components also makes the acceleration signal extremely complex. If the acceleration signal is directly used as the reference signal related to the motion noise of the adaptive filter (Least Mean Squares, LMS), the convergence rate and filtering performance of the adaptive filter will be reduced.

[0077] In order to solve the above problems, in the embodiment of the present application, the acceleration signal is first processed by the principal component analysis method (PCA) to obtain the signal component of the main noise as the reference signal of the adaptive filter. The second physiological signal and the reference signal are filtered based on the adaptive filter to obtain an output result. Based on the output result and the second physiological signal, an error signal is obtained. After the error signal is obtained, the filter parameters are iteratively adjusted based on the error signal, and the first physiological signal is output when the preset convergence criterion is met.

[0078] Among them, the LMS algorithm is a classic adaptive filtering algorithm, which adjusts the filter coefficients by minimizing the mean square error. However, the convergence speed of the LMS algorithm may be slow, especially when the signal characteristics change. Based on this problem, the embodiment of the present application optionally adopts an adaptive step size least mean square (AS-LMS) algorithm. AS-LMS solves the above problem by dynamically adjusting the step size, thereby accelerating the convergence speed and improving the filtering performance.

[0079] Optionally, when the AS-LMS algorithm is used, it is necessary to iteratively adjust the filter parameters based on the error signal, and adaptively adjust the step size parameters based on the error signal, and output the first physiological signal when a preset convergence criterion is met.

[0080] For ease of understanding, here we use Figure 3 The filtering process of AS-LMS is described below. The English in the attached figure is explained as follows:

[0081] S(n)+N(n) is the input signal, i.e., the second physiological signal, which is a PPG signal contaminated by motion artifacts; MA corrupted PPG signal is a PPG signal contaminated by motion artifacts; Noise Reference signal Generation is noise reference signal generation; N Ris the reference signal obtained after PCA processing; AdaptiveFilter is the adaptive filter; CS-LMS / TVS-LMS / AS-LMS algorithm is the cyclostationary least mean squares (CS-LMS) algorithm / time-varying step size least mean squares (Time-Varying Step Size Least Mean Squares) algorithm / adaptive step size least mean squares (Adaptive StepSize Least Mean Squares) algorithm; N^(n) is the output of the tap filter; e(n) is the error signal; s^(n) is the output signal, that is, the first physiological signal.

[0082] In the embodiment of the present application, after PCA processing is performed on the input signal (S(n)+N(n)), a reference signal N is generated. R , based on the adaptive filter for the input signal (S(n)+N(n)) and the reference signal N R Perform filtering processing to obtain an output result N^(n), obtain an error signal e(n) based on the output result N^(n) and the input signal (S(n)+N(n)), adjust the filter parameters based on the error signal e(n), and output the first physiological signal S^(n) when the preset convergence criterion is met; or, adjust the filter parameters w based on the error signal e(n) i (n) and step size parameter μ(n), when a preset convergence criterion is met, a first physiological signal S^(n) is output.

[0083] The relevant formula is as follows:

[0084] e(n)=S(n)+N(n)-N^(n);

[0085]

[0086] w i (n+1)=w i (n)+μ n S(n)N R (ni), which is used to update the filter coefficients;

[0087] μ(n+1)=μ(n)+ρe(n)γ H (n)u(n), this formula is used to update the step size parameter;

[0088] Among them, ρ is the learning rate; γ(n) is the gradient vector; w i (n) is the weight value, which can also be called the filter parameter; μ is the step size.

[0089] In the embodiment of the present application, the quality of the final output signal is guaranteed by continuously iteratively optimizing the filter coefficients, and the convergence speed of the algorithm is accelerated by continuously adaptively adjusting the step size parameters, thereby better eliminating the noise in the PPG signal.

[0090] Alternatively, if Figure 3 As shown, in addition to using AS-LMS to accelerate convergence and improve filtering performance, CS-LMS or TVS-LMS algorithms can also be used.

[0091] Optionally, a convergence criterion is pre-set, and the convergence criterion may be: an error threshold convergence criterion, an iteration number convergence criterion, an error change rate convergence criterion, or a comprehensive convergence criterion. Specifically:

[0092] Error threshold convergence criterion, set a target error threshold, when the error signal is less than the threshold, it is considered to have converged to a satisfactory result and stop adjustment. For example, when the error signal e(n) is less than 0.01, it is considered to have converged.

[0093] Iteration number convergence criterion, set a maximum number of iterations, when the number of iterations reaches this value, stop adjusting.

[0094] The error change rate convergence criterion detects the size of the error signal change rate. When the change rate is less than a certain threshold, it is considered to have converged. For example, when |e(n)-e(n-1)|<0.0001, it is considered to have converged.

[0095] Comprehensive convergence criteria: convergence is considered to have occurred only when multiple convergence criteria mentioned above are met at the same time.

[0096] In one possible implementation, the objective function includes a fitting function and a cost function; the fitting function includes a scaling vector to be optimized; the iterative optimization of the preset objective function until convergence to obtain a scaling vector for the first motion signal includes: determining the scaling vector when the cost function takes a minimum value through iterative optimization; and using the scaling vector when the cost function takes a minimum value as the scaling vector of the first motion signal.

[0097] In a possible implementation, obtaining a target physiological signal based on the first physiological signal, the first motion signal and the scaling vector includes: determining a product value of the first motion signal and the scaling vector; and taking a difference between the first physiological signal and the product value as the target physiological signal.

[0098] In the embodiment of the present application, the objective function can be expressed as follows:

[0099] z=ya×x (Formula 1)

[0100] S=z Twz+μ||Dz|| 2 (Formula 2)

[0101]

[0102] Among them, Formula 1 is the fitting function, and Formula 2 and Formula 3 are cost functions.

[0103] Wherein, z is the difference spectrum, that is, the target physiological signal after noise removal; y is the spectrum of the first physiological signal, y=(y1; y2; ...; y N ) T , x is the spectrum of the first motion signal, x=(x1;x2;...;x N ) T , N is the spectrum length; a is the scaling vector, that is, each element of x is multiplied by a different scaling factor; D is the differential matrix; w is the diagonal matrix, that is, w = diag (w); μ is the regularization parameter; z T wz represents the asymmetric fitting degree between the fitting function and the first physiological signal, μ||Dz|| 2 It is to ensure the smoothness of the fitting function, and μ plays a role in balancing the degree of asymmetric approximation and smoothness; the goal is to optimize the solution of the scaling vector a to make the negative peak in z as weak as possible.

[0104] In one possible implementation, Figure 4 A schematic flow chart of another physiological signal processing method provided by the present application is shown.

[0105] S401, acquiring a first physiological signal and a first motion signal.

[0106] S402 : Iteratively optimize a preset objective function until convergence, and obtain a scaling vector for the first motion signal.

[0107] S403: Obtain a target physiological signal according to the first physiological signal, the first motion signal, and the scaling vector.

[0108] Among them, the relevant descriptions of S401-S403 in the embodiment of the present application are consistent with the relevant descriptions of S201-S203 in the above embodiment, and will not be repeated here.

[0109] S404: extract features of the target physiological signal to obtain a spectrum signal.

[0110] S405, performing spectrum peak tracking on the spectrum signal to obtain a heart rate spectrum peak value.

[0111] In an embodiment of the present application, the spectrum peak of the spectrum signal can be tracked based on the XGBoost model to obtain the heart rate spectrum peak value.

[0112] Among them, XGBoost is an efficient gradient boosting tree algorithm. In XGBoost, models are trained one after another in sequence, and each new model is trained on the basis of the previous model with the goal of correcting the residuals of the previous model. The prediction results of each model are added to the prediction results of the previous model to form the final prediction output. In addition, XGBoost performs a second-order Taylor expansion on the loss function, that is, it locally approximates the loss function near the predicted value of the current model. Through the first and second derivatives of the loss function, XGBoost can find the minimum value of the loss function and thus optimize the model parameters.

[0113] S406: Obtain a heart rate value based on the heart rate spectrum peak value.

[0114] Among them, it is necessary to first extract features from the target physiological signal. These features may include but are not limited to the signal's time domain features, frequency domain features, statistical characteristics, coefficients, etc. The purpose of feature extraction is to convert the target physiological signal into a set of parameters that can characterize the signal characteristics. These parameters will be used for subsequent spectrum peak tracking and heart rate calculation.

[0115] First, the following features need to be extracted:

[0116] Maximum coefficient feature: In the heart rate spectrum, approximately 75% of the correct heart rate spectrum peaks have the largest coefficient within their corresponding time window.

[0117] Minimum peak spacing feature: Among the heart rate spectrum peaks, about 84% of the correct heart rate spectrum peaks have the smallest spacing with the correct heart rate spectrum peaks in the previous time window. This indicates that the heart rate peaks tend to be closely arranged in consecutive time windows.

[0118] Combining the maximum coefficient and minimum peak spacing features: Among the heart rate spectrum peaks, approximately 96% of the correct heart rate spectrum peaks have both the maximum coefficient and the minimum peak spacing.

[0119] Among them, since the heart rate changes slowly, the real heart rate frequency position L in the previous time window can be pre The peak value is searched within a certain range, which is determined by the basic fluctuation range of the heart rate frequency, including the following two fluctuation ranges:

[0120] The first fluctuation range of heart rate frequency L rang1 :

[0121] L rang1 =[L pre -Δs,...,L pre +Δs];

[0122] The second fluctuation range of heart rate frequency L rang2 :

[0123] L rang2 =[2(L pre -Δs-1), ..., [2(L pre +Δs+1)];

[0124] Among them, Δs is a constant.

[0125] In addition to extracting the above features, it is also necessary to determine the peak spacing, peak coefficient, acceleration time domain power, acceleration frequency domain power, acceleration and PPG correlation coefficient and other features. Specifically:

[0126] Peak distance: calculated in the first fluctuation range L rang1 The positions of the three largest peaks within the second fluctuation range L rang2 The positions of the three largest peaks in the heart rate wave are 6 candidate peaks in total. If there are less than 6, the corresponding value is 0. The distance between each peak and the previous heart rate wave peak is calculated through the 6 candidate peaks as follows: i =|f i -f pre |; where f i is a candidate peak, i is a positive integer from 1 to 6, f pre is the position of the previous heart rate peak of the current peak;

[0127] Peak coefficient, also known as the coefficient ratio of the heart rate spectrum peak C i : Among them, coe i represents the i-th candidate peak coefficient, i is a positive integer from 1 to 6, coe max Indicates the largest coefficient in the candidate peak set.

[0128] Acceleration time domain power: Among them, x i represents the acceleration value of the i-th sampling point, and N is the total number of acceleration signal samples.

[0129] Acceleration frequency domain power: Among them, x(f k ) is at the frequency point f k The complex value of the Fourier transform at ; M is the length of the discrete Fourier transform DFT of the signal.

[0130] Correlation coefficient between acceleration and PPG: Among them, ρ m is the correlation coefficient between the acceleration of the X, Y, and Z axes and the PPG; cov(X, Y) is the covariance of the acceleration and the PPG signal; σ X is the standard deviation of acceleration; σ Y is the standard deviation of the PPG signal.

[0131] After obtaining the above features, you need to set labels. Take the frequency position of the standard electrocardiogram (ECG) heart rate value as the standard position L oc , in L rang1 There is a peak value with L oc closest, the label is marked as 1; in L rang2 There is a peak value with L oc closest, the label is 2; if in L rang1 and L rang2 If no peak is found, it is marked as label 3.

[0132] The XGBoost model is used to learn the above features to identify and track the peaks of the heart rate spectrum. The extracted features are used as input and the labels are used as output Y to train the XGBoost model so that it can predict the peaks of the heart rate spectrum. When calculating the heart rate, the corresponding features are extracted and passed into the trained XGBoost model to obtain Y.

[0133] Among them, XGBoost is an additive model composed of k base models. Assume that the tree model to be trained in the tth iteration is f t (input), then

[0134] If Y is label 1 or label 2, the input f i is the predicted peak position f loc If Y is label 3, it is considered that the PPG signal in the current time window has been severely contaminated by noise, and the heart rate spectrum peak is predicted by the following strategy.

[0135]

[0136] h=f predict -f pre-predict

[0137] Among them, f predict is the frequency corresponding to the predicted heart rate spectrum peak in the current time window; f pre-predict represents the frequency corresponding to the predicted heart rate spectrum peak of the previous time window, and h is the difference between the frequency corresponding to the predicted heart rate spectrum peak of the current time window and the frequency corresponding to the predicted heart rate spectrum peak of the previous time window.

[0138] In the embodiment of the present application, a smoothing algorithm is used to obtain f predict and f pre-predict The algorithm uses the frequency corresponding to the estimated heart rate spectrum peak in the last count time windows to calculate the heart rate through the following formula: hr = f loc ×60; count is a positive integer and is set according to actual needs.

[0139] In the embodiments of the present application, motion artifacts are effectively removed through the methods in the above embodiments, and the exercise heart rate can be accurately calculated in sports with large motion artifacts such as elliptical machines and running.

[0140] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0141] Corresponding to the physiological signal processing method described in the above embodiment, Figure 5 A structural block diagram of a physiological signal processing device provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0142] Reference Figure 5 , the physiological signal processing device comprises:

[0143] A first acquisition module 501, used to acquire a first physiological signal and a first motion signal;

[0144] An iterative optimization module 502, configured to iteratively optimize a preset objective function until convergence, and obtain a scaling vector for the first motion signal;

[0145] The signal determination module 503 is configured to obtain a target physiological signal according to the first physiological signal, the first motion signal and the scaling vector.

[0146] In a possible implementation, the physiological signal processing device further includes:

[0147] A second acquisition module, used for acquiring a second physiological signal;

[0148] A reference signal acquisition module, used for processing the first motion signal based on a principal component analysis method to obtain a reference signal;

[0149] The first physiological signal acquisition module is used to obtain the first physiological signal according to the second physiological signal and the reference signal.

[0150] In a possible implementation, the first physiological signal acquisition module is used to:

[0151] Performing filtering processing on the second physiological signal and the reference signal based on a filter to obtain an output result;

[0152] obtaining an error signal based on the output result and the second physiological signal;

[0153] The filter parameters are adjusted based on the error signal, and when a preset convergence criterion is met, the first physiological signal is output; or the filter parameters and step size parameters are adjusted based on the error signal, and when the preset convergence criterion is met, the first physiological signal is output.

[0154] In a possible implementation, the physiological signal processing device further includes:

[0155] A third acquisition module, used for acquiring a third physiological signal and a third motion signal;

[0156] The bandpass filtering module is used to perform bandpass filtering on the third physiological signal and the third motion signal to obtain the second physiological signal and the first motion signal.

[0157] In a possible implementation, the objective function includes a fitting function and a cost function; the fitting function includes a scaling vector to be optimized; then the iterative optimization module 502 is used to:

[0158] Determine, by iterative optimization, a scaling vector when the cost function takes a minimum value;

[0159] The scaling vector that makes the cost function take the minimum value is used as the scaling vector of the first motion signal.

[0160] In a possible implementation, the objective function is constructed based on an asymmetric least squares method and an adaptive iterative reweighting method.

[0161] In a possible implementation, the signal determination module 503 is configured to:

[0162] determining a product value of the first motion signal and the scaling vector;

[0163] The difference between the first physiological signal and the product value is used as the target physiological signal.

[0164] In a possible implementation, the physiological signal processing device further includes:

[0165] A feature extraction module, used to extract features of the target physiological signal to obtain a spectrum signal;

[0166] A spectrum peak tracking module, used for tracking the spectrum peak of the spectrum signal to obtain the peak value of the heart rate spectrum;

[0167] The heart rate value calculation module is used to obtain the heart rate value based on the heart rate spectrum peak value.

[0168] It should be noted that the information interaction, execution process and other contents between the above-mentioned modules are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0169] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0170] An embodiment of the present application also provides a computer device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps of any of the above-mentioned method embodiments when executing the computer program.

[0171] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0172] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0173] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0174] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0175] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0176] In the embodiments provided in the present application, it should be understood that the disclosed apparatus / computer equipment and methods can be implemented in other ways. For example, the apparatus / computer equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the apparatus or unit, which can be electrical, mechanical or other forms.

[0177] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0178] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0179] Figure 6 This is a schematic diagram of the structure of a computer device provided in one embodiment of the present application. Figure 6 As shown, the computer device of this embodiment includes: at least one processor 20 ( Figure 6 Only one is shown in the figure), a memory 21, and a computer program 22 stored in the memory 21 and executable on the at least one processor 20, wherein the processor 20 implements the steps in any of the above-mentioned physiological signal processing method embodiments when executing the computer program 22.

[0180] The computer device may include, but is not limited to, a processor 20 and a memory 21. Those skilled in the art will appreciate that Figure 6 It is only an example of a computer device and does not constitute a limitation of the computer device. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0181] The processor 20 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0182] In some embodiments, the memory 21 may be an internal storage unit of the computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory 21 may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Further, the memory 21 may also include both an internal storage unit of the computer device and an external storage device. The memory 21 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 21 may also be used to temporarily store data that has been output or is to be output.

Claims

1. A physiological signal processing method, characterized in that: include: Acquiring a first physiological signal and a first motion signal; Iteratively optimizing a preset objective function until convergence, to obtain a scaling vector for the first motion signal; A target physiological signal is obtained according to the first physiological signal, the first motion signal and the scaling vector.

2. The physiological signal processing method according to claim 1, characterized in that: The acquiring of the first physiological signal and the first motion signal also includes: acquiring a second physiological signal; Processing the first motion signal based on a principal component analysis method to obtain a reference signal; The first physiological signal is obtained according to the second physiological signal and the reference signal.

3. The physiological signal processing method according to claim 2, characterized in that: The step of obtaining the first physiological signal according to the second physiological signal and the reference signal includes: Performing filtering processing on the second physiological signal and the reference signal based on a filter to obtain an output result; obtaining an error signal based on the output result and the second physiological signal; The filter parameters are adjusted based on the error signal, and when a preset convergence criterion is met, the first physiological signal is output; or the filter parameters and step size parameters are adjusted based on the error signal, and when the preset convergence criterion is met, the first physiological signal is output.

4. The physiological signal processing method according to claim 2, characterized in that: The acquiring of the second physiological signal also includes: acquiring a third physiological signal and a third motion signal; The third physiological signal and the third motion signal are band-pass filtered to obtain the second physiological signal and the first motion signal.

5. The physiological signal processing method according to any one of claims 1 to 4, characterized in that: The objective function includes a fitting function and a cost function; the fitting function includes a scaling vector to be optimized; The iterative optimization of a preset objective function until convergence to obtain a scaling vector for the first motion signal includes: Determine, by iterative optimization, a scaling vector when the cost function takes a minimum value; The scaling vector that makes the cost function take the minimum value is used as the scaling vector of the first motion signal.

6. The physiological signal processing method according to claim 5, characterized in that: The objective function is constructed based on an asymmetric least square method and an adaptive iterative reweighting method.

7. The physiological signal processing method according to any one of claims 1 to 4, characterized in that: The step of obtaining a target physiological signal according to the first physiological signal, the first motion signal and the scaling vector includes: determining a product value of the first motion signal and the scaling vector; The difference between the first physiological signal and the product value is used as the target physiological signal.

8. The physiological signal processing method according to any one of claims 1 to 4, characterized in that: The method further comprises: Extracting features of the target physiological signal to obtain a spectrum signal; Performing spectrum peak tracking on the spectrum signal to obtain a heart rate spectrum peak value; A heart rate value is obtained based on the heart rate spectrum peak value.

9. A physiological signal processing device, characterized in that: include: A first acquisition module, used to acquire a first physiological signal and a first motion signal; An iterative optimization module, used for iteratively optimizing a preset objective function until convergence, and obtaining a scaling vector for the first motion signal; A signal determination module is used to obtain a target physiological signal according to the first physiological signal, the first motion signal and the scaling vector.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images