A heart rate monitoring method, device, apparatus and storage medium

Abstract

The application discloses a heart rate monitoring method, device, equipment and storage medium, which is used for anti-interference processing and feature recognition on collected digital artery pulse wave signals and outputs a heart rate result. The method comprises the following steps: monitoring an original digital artery pulse wave signal of a user, wherein the original digital artery pulse wave signal is a vibration signal generated by digital artery pulse pressure pulsation and directly conducted to a wearable device through soft tissue; performing signal enhancement and motion artifact suppression processing on the original digital artery pulse wave signal to obtain a usable digital artery pulse wave signal; identifying a pulse wave feature point from the usable digital artery pulse wave signal; and calculating a heart rate result based on the pulse wave feature point, wherein the heart rate result comprises a heart rate variability result.

Description

Technical Field

[0001] This application relates to the field of wearable device technology, and in particular to a heart rate monitoring method, device, equipment and storage medium. Background Technology

[0002] With the rapid popularization of smart wearable devices, heart rate monitoring has become an important part of daily health management for the public, providing users with a basic reference for their cardiovascular health status. Wearable devices are suitable as wearable carriers for long-term heart rate monitoring due to their unique advantages such as convenient wearing, non-interference with daily activities, high skin-to-finger fit, and all-weather monitoring capability.

[0003] Currently, the mainstream heart rate monitoring technology in the wearable device field is photoplethysmography (PPG) technology. Its core principle is that the wearable device emits light signals of a specific wavelength through a built-in light emitting module. After the light signal penetrates the skin of the finger, it is absorbed and reflected by the subcutaneous arteries. The light receiving module then captures the intensity changes of the reflected light signal and converts it into a pulse wave signal. Finally, the user's heart rate is calculated based on the pulse wave signal. In addition, some technologies attempt to use accelerometers to replace or assist PPG technology. The accelerometer captures the vibration signal generated by the pulse pressure of the finger artery transmitted to the ring shell, thereby obtaining pulse wave information and realizing heart rate monitoring.

[0004] However, for PPG technology, the acquired signals are easily affected by ambient light, the state of skin contact with the fingers, and the wearing posture. Furthermore, when the user is in motion, motion artifacts can severely interfere with the accuracy of the pulse wave signal, leading to distorted heart rate monitoring results. At the same time, its calculation accuracy for heart rate variability parameters is low. For technologies using accelerometers, they have failed to effectively solve the interference problem in the original vibration signal. Useless components such as environmental vibration and motion interference are mixed in with the pulse wave signal, making it difficult to extract a usable pulse wave signal. Moreover, the identification of pulse wave feature points lacks specificity and cannot adapt to the differences in physiological characteristics of different users, resulting in insufficient stability and accuracy of heart rate and heart rate variability results. Summary of the Invention

[0005] This application discloses a heart rate monitoring method, device, equipment, and storage medium, which is used to perform anti-interference processing and feature recognition on the collected digital artery pulse wave signal, and output heart rate and heart rate variability monitoring results.

[0006] The first aspect of this application discloses a method for heart rate monitoring, including:

[0007] The device monitors the user's original digital artery pulse wave signal, which is a vibration signal generated by the pulse pressure of the digital artery being directly transmitted to the wearable device through soft tissue. The original digital artery pulse wave signal is processed with signal enhancement and motion artifact suppression to obtain a usable digital artery pulse wave signal; Identify pulse wave feature points from the available digital artery pulse wave signals; Heart rate results are calculated based on the pulse wave feature points, and the heart rate results include heart rate variability results.

[0008] Optionally, the step of performing signal enhancement and motion artifact suppression processing on the original digital artery pulse wave signal to obtain a usable digital artery pulse wave signal includes: A covariance matrix is ​​constructed on the original digital artery pulse wave signal and principal component decomposition is performed. The target principal component is extracted as the principal direction component. The target principal component is the part with the highest energy proportion in the original digital artery pulse wave signal after principal component decomposition. The main direction component is bandpass filtered to retain the effective frequency band in the original digital artery pulse wave signal that matches the physiological characteristics of the pulse wave, thus completing the signal enhancement. Spectral analysis was performed on the enhanced original digital artery pulse wave signal to identify environmental vibration interference components. The environmental vibration interference components are suppressed. Based on the changes in high-frequency components and signal variance in the original finger artery pulse wave signal after signal enhancement, it is determined whether the user is currently in motion. If so, artifact labeling and removal are performed on the original digital artery pulse wave signal under motion to obtain a usable digital artery pulse wave signal.

[0009] Optionally, identifying pulse wave feature points from the available finger artery pulse wave signal includes: The signal envelope of the available finger artery pulse wave signal is extracted, and a dynamic peak detection threshold is determined based on the amplitude of the signal envelope. The dynamic peak detection threshold is recalculated and automatically updated every few heartbeat cycles. The waveform peaks in the available finger artery pulse wave signals that are higher than the dynamic peak detection threshold are matched with a preset personal pulse wave template, and the peaks that are matched are retained as pulse wave feature points.

[0010] Optionally, the calculation of heart rate based on the pulse wave feature points includes: The heartbeat interval sequence between adjacent pulse wave feature points is converted into an instantaneous heart rate sequence; Perform sliding median filtering on the instantaneous heart rate sequence to remove abnormal heart rate values ​​caused by interference, and output the heart rate result; Based on the intercardia sequence, the corresponding time-domain and frequency-domain indices are calculated respectively. The heart rate variability results in the heart rate results are obtained based on the time domain index and the frequency domain index.

[0011] Optionally, after obtaining a usable digital artery pulse wave signal, the method further includes: Morphological analysis was performed on the available digital artery pulse wave signals; Extract systolic rise time, dicrotic notch location and depth, pulse wave area ratio, and second derivative waveform characteristics, and assess arterial compliance, peripheral resistance, and vascular health status.

[0012] Optionally, the method further includes: During the monitoring of the user's raw digital artery pulse wave signal, when the user's heart rate abnormality, heart rate interval abnormality, or the preset calibration conditions are detected, the PPG monitoring mode is activated, and the monitoring results are checked and calibrated. The monitoring of the user's raw digital artery pulse wave signal adopts the accelerometer monitoring mode.

[0013] Optionally, the method further includes: The stability of the signal determines whether the user is at rest or in motion. If in a resting state, the accelerometer monitoring mode is maintained; If the device is in motion, switch to the PPG monitoring mode.

[0014] A second aspect of this application provides a heart rate monitoring device, comprising: The monitoring unit is used to monitor the user's original digital artery pulse wave signal, which is the vibration signal generated by the digital artery pulse pressure pulsation directly transmitted to the wearable device through soft tissue. The processing unit is used to perform signal enhancement and motion artifact suppression processing on the original digital artery pulse wave signal to obtain a usable digital artery pulse wave signal; The identification unit is used to identify pulse wave feature points from the available finger artery pulse wave signals; A calculation unit is used to calculate heart rate results based on the pulse wave feature points, the heart rate results including heart rate variability results.

[0015] A third aspect of this application provides a heart rate monitoring device, comprising: Processor, memory, input / output units, and bus; The processor is connected to memory, input / output units, and a bus; The memory holds a program, which the processor calls to execute, as in the first aspect and any optional method of the first aspect.

[0016] The fourth aspect of this application provides a computer-readable storage medium on which a program is stored, which, when executed on a computer, performs the methods of the first aspect and any optional methods of the first aspect.

[0017] As can be seen from the above technical solutions, this application has the following advantages: 1. By acquiring the raw digital artery pulse wave signal, and clarifying that the raw digital artery pulse wave signal refers to the vibration signal generated by the arterial pulse pressure pulsation directly transmitted to the wearable device through soft tissue, interference factors related to light signals are fundamentally avoided. At the same time, the vibration essence of the pulse pulsation is directly captured, reducing the influence of wearing posture and skin contact state on signal acquisition, avoiding the misacquisition of environmentally irrelevant vibrations, and ensuring that the acquired raw digital artery pulse wave signal is more targeted and accurate.

[0018] 2. The original digital artery pulse wave signal is enhanced and motion artifacts are suppressed. The signal enhancement highlights the effective signal component of the pulse wave, and the motion artifact suppression removes useless components such as motion interference and environmental vibration. This solves the problem of signal distortion caused by interference in the existing technology and ensures that the final usable digital artery pulse wave signal has high stability and high effectiveness.

[0019] 3. It identifies pulse wave feature points from the available finger artery pulse wave signals after interference suppression, improving the targeting and accuracy of feature point identification; at the same time, it calculates heart rate results based on accurately identified pulse wave feature points, significantly improving the real-time performance and accuracy of heart rate results, meeting users' needs for heart rate monitoring accuracy. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A schematic flowchart of an embodiment of a heart rate monitoring method provided in this application; Figure 2 A schematic flowchart of an embodiment for obtaining and analyzing usable digital artery pulse wave signals provided in this application; Figure 3 A schematic flowchart of an embodiment for identifying pulse wave feature points provided in this application; Figure 4 A schematic flowchart of an embodiment for calculating heart rate results provided in this application; Figure 5A schematic flowchart of an embodiment of the process for monitoring the raw digital artery pulse wave signal of a user provided in this application; Figure 6 A structural diagram of an embodiment of a heart rate monitoring device provided in this application; Figure 7 This is a structural diagram of another embodiment of a heart rate monitoring device provided in this application. Detailed Implementation

[0022] This application provides a heart rate monitoring method, device, equipment, and storage medium for performing anti-interference processing and feature recognition on the acquired digital artery pulse wave signal and outputting heart rate results.

[0023] It should be noted that the heart rate monitoring method provided in this application is implemented by a wearable device, such as a smart ring.

[0024] Please see Figure 1 This application provides an embodiment of a heart rate monitoring method, comprising: 101. Monitor the user's original digital artery pulse wave signal. The original digital artery pulse wave signal is the vibration signal generated by the digital artery pulse pressure pulsation being directly transmitted to the wearable device through soft tissue. The core objective of this step is to acquire the user's raw digital artery pulse wave signal, which is unprocessed. This raw digital artery pulse pressure signal originates from the pulse pressure changes caused by blood flow within the user's digital arteries. Specifically, after the user wears the relevant wearable device, the device acquires the raw digital artery pulse wave signal via a triaxial accelerometer embedded within the device, using the following physical coupling link: digital artery pulse pressure (30-50 mmHg) → radial expansion of the arterial wall (20-50 μm) → conduction through surrounding soft tissue → vibration of the rigid housing of the wearable device → detection by the accelerometer. The triaxial accelerometer then converts this mechanical vibration into a collectable and processable electrical signal, i.e., the raw digital artery pulse wave signal. This raw signal directly reflects the pulsating state of the user's digital arteries and contains core information related to heart rate. The accelerometer is installed on the inner surface of the wearable device near the direction of the digital artery (i.e., on both palmar sides of the fingers) to maximize mechanical coupling efficiency.

[0025] 102. Perform signal enhancement and motion artifact suppression processing on the original digital artery pulse wave signal to obtain usable digital artery pulse wave signal; Because the raw digital artery pulse wave signal acquired in step 101 has weak signal strength and is easily interfered with, especially when the user moves their limbs while wearing the device, motion artifacts will occur. Motion artifacts will seriously interfere with the authenticity of the pulse wave signal and affect the accuracy of heart rate calculation. Therefore, signal enhancement technology can be used to amplify the effective components related to heart rate in the raw pulse wave signal, improve the signal-to-noise ratio, and make the fluctuation characteristics of the pulse wave clearer. At the same time, a motion artifact suppression algorithm is used to identify and filter out irrelevant interference signals caused by user limb movement, device friction, etc., and eliminate the influence of artifacts on the effective signal, finally obtaining a usable digital artery pulse wave signal with clear signal and less interference.

[0026] 103. Identify pulse wave feature points from available finger artery pulse wave signals; The usable finger artery pulse wave signal contains complete pulse cycle information, and pulse wave feature points are key identifiers reflecting heart rate and heart rate variability. This step accurately identifies the core feature points from the processed usable finger artery pulse wave signal. Specifically, using a preset feature point recognition algorithm, the waveform of the usable finger artery pulse wave signal is analyzed segment by segment to identify the feature points within each pulse cycle, including but not limited to the rising edge start point, peak point, and falling edge inflection point of the pulse wave. The distribution pattern of these feature points directly corresponds to the contraction and relaxation process of the heart, and the accuracy of feature point recognition directly determines the reliability of the final heart rate result.

[0027] 104. Calculate heart rate results based on pulse wave feature points. The heart rate results include heart rate variability results.

[0028] This step counts the number of pulse beats per unit time based on the identified feature points, and then calculates the user's baseline heart rate. Based on this, by analyzing the time interval between adjacent pulse cycles and the changing patterns of feature point morphology, heart rate variability is calculated—heart rate variability reflects the fluctuations in the heart's rhythm and is an important indicator for assessing the state of the cardiovascular system. The final output heart rate result includes both baseline heart rate and heart rate variability, two core data points.

[0029] In this embodiment, by acquiring the original finger artery pulse wave signal, and clarifying that the original finger artery pulse wave signal refers to the vibration signal generated by the direct transmission of the finger artery pulse pressure through soft tissue to the wearable device, interference factors related to the light signal are fundamentally avoided. At the same time, the vibration essence of the pulse is directly captured, reducing the influence of wearing posture and skin contact state on signal acquisition, avoiding the misacquisition of environmentally irrelevant vibrations, and ensuring that the acquired original finger artery pulse wave signal is more targeted and accurate.

[0030] The original digital artery pulse wave signal is enhanced and motion artifacts are suppressed. The signal enhancement highlights the effective signal component of the pulse wave, and the motion artifact suppression removes useless components such as motion interference and environmental vibration. This solves the problem of signal distortion caused by interference in the existing technology and ensures that the final usable digital artery pulse wave signal has high stability and high effectiveness.

[0031] Identifying pulse wave feature points from available finger artery pulse wave signals after interference suppression improves the targeting and accuracy of feature point identification. At the same time, calculating heart rate results based on accurately identified pulse wave feature points significantly improves the real-time performance and accuracy of heart rate results, meeting users' needs for heart rate monitoring precision.

[0032] In step 102 above, the raw digital artery pulse wave signal is subjected to signal enhancement and motion artifact suppression processing to obtain a usable digital artery pulse wave signal. Please refer to [link to relevant documentation]. Figure 2 , Figure 2 One embodiment of this application for obtaining and analyzing usable digital artery pulse wave signals includes: 201. Construct a covariance matrix for the original digital artery pulse wave signal and perform principal component decomposition. Extract the target principal component as the principal direction component. The target principal component is the part with the highest energy proportion in the original digital artery pulse wave signal after principal component decomposition. The core of this step is to first organize the acquired raw digital artery pulse wave signals and construct a corresponding covariance matrix. This covariance matrix can accurately reflect the correlation between the data of each dimension of the raw digital artery pulse wave signal and filter out redundant information between signal dimensions. Then, principal component decomposition is performed on the covariance matrix to obtain multiple principal components. Each principal component is sorted from high to low according to its energy proportion. The target principal component usually has an energy proportion of more than 80%, which is the part with the most concentrated core information in the original signal and can preserve the physiological characteristics of the pulse wave to the greatest extent. Therefore, this target principal component is extracted as the principal direction component to initially filter out the scattered weak energy interference in the signal and focus on the core of the effective signal.

[0033] 202. Bandpass filtering is applied to the main directional component to retain the effective frequency bands in the original digital artery pulse wave signal that match the physiological characteristics of the pulse wave, thus completing the signal enhancement. The main directional component still contains frequency band signals unrelated to the physiological characteristics of the pulse wave. This step combines the physiological characteristics of the human pulse wave. The effective frequency band of a normal pulse wave is 0.5-15Hz. The main directional component is input into a 0.5-15Hz fourth-order Butterworth filter. The filter frequency band of 0.5-15Hz is set to filter out low-frequency interference below 0.5Hz (such as breathing fluctuations, slight equipment drift) and high-frequency interference above 15Hz (such as electronic noise, slight vibration). Only the effective frequency band matching the pulse wave is retained to enhance the signal and make the pulse wave waveform characteristics clearer.

[0034] 203. Perform spectral analysis on the original finger artery pulse wave signal after signal enhancement to identify environmental vibration interference components; This step performs spectral analysis on the raw digital artery pulse wave signal after bandpass filtering in step 202. A Fast Fourier Transform (FFT) is used to convert the time-domain signal to a frequency-domain signal, focusing on detecting the energy proportion of 50Hz and its harmonics (such as 100Hz, 150Hz, etc.). A preset energy proportion threshold (e.g., 5%) is used. When the detected energy proportion of 50Hz and its harmonics exceeds this threshold, significant environmental vibration interference is identified, and the specific frequency corresponding to the interference (50Hz and its corresponding harmonic frequency) is recorded. If the threshold is not exceeded, no significant environmental vibration interference is considered, and no further suppression processing is required. This targeted detection accurately identifies common power frequency vibration interference in the environment, avoiding misclassification of irrelevant signals as interference, while ensuring the efficiency and accuracy of interference detection.

[0035] 204. Suppress environmental vibration interference components; Based on the detection results of step 203, when it is determined that there is 50Hz and its harmonic interference (energy ratio exceeds the threshold), the corresponding frequency IIR notch filter is activated. Notch filter parameters are set for 50Hz and the detected harmonic frequencies respectively to accurately filter the interference components corresponding to these frequencies, while retaining the effective frequency band signal of the pulse wave from 0.5 to 15Hz to avoid damage to the effective pulse wave signal when suppressing interference. After processing, the signal is initially verified to confirm that the 50Hz and its harmonic interference has been effectively suppressed and the signal waveform has no obvious distortion.

[0036] 205. Based on the changes in high-frequency components and signal variance in the original digital artery pulse wave signal after signal enhancement, determine whether the current user is in motion. Motion artifacts generated during user movement are one of the main interference factors affecting the quality of pulse wave signals. Their interference intensity is usually much greater than that of environmental vibration interference, which can lead to severe distortion of the pulse wave waveform and blurring of feature points, thereby affecting the accuracy of subsequent feature extraction and heart rate calculation.

[0037] This step utilizes the high-frequency components (>15Hz) of the acceleration signal and the sudden changes in signal variance to determine the motion state in two ways: real-time monitoring of the energy of the high-frequency components with frequencies greater than 15Hz in the acceleration signal, and monitoring of the overall variance change of the signal. When the energy of the high-frequency components increases significantly and the signal variance shows a sudden increase (i.e., a sudden change in variance), it is determined that the current user is in motion. During the motion period, the signal of the corresponding time period is directly marked as an unreliable signal.

[0038] If the energy of the high-frequency component (>15Hz) in the acceleration signal increases by more than 40% compared to when it is stationary, and the signal variance suddenly increases from 0.02 when it is stationary to more than 0.1, it can be clearly determined to be in motion, and the signal during this period should be marked as unreliable.

[0039] 206. If so, then the original digital artery pulse wave signal under motion is subjected to artifact marking and removal to obtain a usable digital artery pulse wave signal. If step 205 detects that the user is in motion, then the signals marked as unreliable during the motion period are specifically processed to remove interference segments, obtaining usable finger artery pulse wave signals. Specifically, using the unreliable signal segments marked in step 205 as the core, and combining the time-domain and frequency-domain characteristics of motion artifacts, the signal after bandpass filtering and environmental vibration suppression is reviewed segment by segment to further confirm the range of artifact segments—in the time domain, artifact segments typically exhibit severe waveform distortion, abnormally high or low peak values, and periodic instability; in the frequency domain, the frequency of artifact segments exceeds the effective pulse wave frequency band of 0.5-15Hz, and the energy of high-frequency components is significantly higher than that of normal pulse wave signals. After marking, a removal algorithm is used to remove the marked artifact segments (including unreliable signals during motion) from the signal. At the same time, the signal after artifact removal is smoothed (e.g., by moving average filtering) to fill in the missing parts of the signal, ensuring the continuity and integrity of the signal.

[0040] After processing, the original digital artery pulse wave signal is quality checked to confirm that there are no obvious motion artifacts and the waveform period is stable. Finally, a usable digital artery pulse wave signal is obtained that can be used for subsequent morphological analysis, feature extraction and heart rate calculation.

[0041] 207. Perform morphological analysis on available digital artery pulse wave signals; Using a pre-defined morphological analysis algorithm, the time-domain waveform of usable finger artery pulse wave signals is analyzed cycle by cycle to clarify the complete waveform outline within each pulse cycle, including the morphological characteristics of the rising segment (the rising portion of the waveform formed by the rapid influx of blood into the peripheral arteries during cardiac systole), the peak segment (the peak value during systole, corresponding to the moment of strongest cardiac contraction), and the falling segment (the falling portion of the waveform formed by the return of blood during cardiac diastole). Simultaneously, key inflection points and extreme points in the waveform are accurately identified (with a focus on identifying J-wave peaks, dicrotic notches, etc.). Furthermore, the waveform's symmetry, continuity, and stability are analyzed to determine the quality of the usable signal—if the waveform is symmetrical, continuous without breaks, and periodically stable, the signal quality is good and can be used for subsequent analysis; if the waveform has slight distortion and small periodic fluctuations, secondary smoothing is required; if the waveform is severely distorted, the preceding processing steps need to be re-verified to ensure that the subsequently extracted features and calculated heart rate results accurately reflect the human physiological state.

[0042] 208. Extract systolic rise time, dicrotic notch location and depth, pulse wave area ratio, and second derivative waveform characteristics, and assess arterial compliance, peripheral resistance, and vascular health status.

[0043] This step consists of two parts: initially extracting key morphological features of the pulse wave, and then performing secondary verification on the available signals. The first part is the preliminary extraction of morphological features. Based on the waveform inflection points and extreme points identified in step 207, the approximate range of the systolic rise time, dicrotic notch position and depth is preliminarily extracted. At the same time, the area ratio of the systolic to diastolic phases of the pulse wave is preliminarily calculated to avoid positioning deviations during subsequent precise extraction. Simultaneously, the second derivative of the pulse wave signal is preliminarily calculated to capture the approximate waveform trend of the acceleration pulse wave (APG) and to preliminarily determine the rationality of the vascular elasticity-related features.

[0044] The second part is a secondary verification of signal quality. Based on the initially extracted morphological features, usable finger artery pulse wave signals are further screened. If the initially extracted features such as the pulse wave systolic rise time and dicrotic notch are within the normal physiological range, and the second derivative waveform shows no significant distortion, the signal quality meets the standard, and dynamic peak detection can continue. If the initially extracted features significantly deviate from the normal range, or the second derivative waveform is severely distorted, it indicates that there are still unremoved artifacts or interference in the signal. In this case, it is necessary to return to step 206 to re-perform artifact removal and smoothing until the signal quality meets the standard. This initial feature extraction and secondary verification further ensures signal quality.

[0045] In this embodiment, by performing principal component decomposition of the covariance matrix on the original digital artery pulse wave signal, the principal direction component with optimal energy can be automatically extracted, adaptively matching different wearing angles and postures. Combined with bandpass filtering, signal enhancement, spectrum analysis to identify and suppress environmental vibration interference are achieved. At the same time, based on high-frequency components and signal variance abrupt changes, motion state is accurately detected and motion artifacts are eliminated, which can reduce the distortion effect of environmental interference and limb movement on the pulse wave signal and improve signal purity. Furthermore, by performing morphological analysis on the available signal and extracting multiple pulse wave physiological features, arterial compliance, peripheral resistance, and overall vascular health can be assessed intuitively and accurately, achieving a digital artery pulse wave detection and analysis effect with strong anti-interference ability, high stability, and comprehensive physiological information interpretation.

[0046] In step 103 above, pulse wave feature points are identified from the available finger artery pulse wave signals. Please refer to [link to relevant documentation]. Figure 3 , Figure 3 One embodiment of the pulse wave feature point identification provided in this application includes: 301. Extract the signal envelope of the usable finger artery pulse wave signal, determine the dynamic peak detection threshold based on the amplitude of the signal envelope, and recalculate and automatically update the dynamic peak detection threshold every few heartbeat cycles. First, the signal envelope of the usable finger artery pulse wave signal obtained in step 206 is extracted. Through signal envelope analysis, the overall amplitude change pattern of the pulse wave signal is captured. Based on the amplitude of the signal envelope, the dynamic peak detection threshold is calculated. This dynamic peak detection threshold is not a fixed value, but will adaptively adjust with the amplitude change of the pulse wave signal. That is, every few heartbeat cycles (such as every 5 beats or every 8 beats), the signal envelope amplitude is recalculated and the dynamic peak detection threshold is updated to ensure that the threshold always matches the actual amplitude of the current pulse wave signal and adapts to the fluctuation of the pulse wave signal under different physiological states of the user (such as resting or after light activity).

[0047] 302. Match the peak values ​​of the available finger artery pulse wave signals that are higher than the dynamic peak detection threshold with the preset personal pulse wave template, and retain the qualified peak values ​​as pulse wave feature points.

[0048] This step accurately identifies pulse wave feature points through threshold filtering and template matching, while simultaneously establishing and continuously updating a personal pulse wave template to further improve the accuracy and personalization of feature point identification. The specific operation is divided into two parts: template establishment and updating, and peak matching and feature point filtering. The first part concerns the establishment and updating of a personal pulse wave template: When a user first wears the device, the device simultaneously collects accelerometer and PPG signals within a preset time (e.g., 5 minutes). Combining the data from both, it filters out pulse wave signal segments with good quality and no obvious interference, and establishes an initial personal pulse wave template based on this. The template includes personalized information such as the waveform characteristics, peak shape, and periodicity of the user's own pulse wave. During subsequent use, the personal pulse wave template is continuously updated based on each actual pulse wave signal collected, correcting the waveform parameters in the template to adapt to changes in the user's vascular condition (such as changes in vascular elasticity, blood pressure fluctuations, etc.), ensuring that the template always matches the user's current pulse wave characteristics.

[0049] The second part involves peak matching and feature point selection: Waveform peaks above the dynamic peak detection threshold in the available finger artery pulse wave signals are matched one by one with a preset personal pulse wave template. The correlation coefficient between each candidate peak and the template is calculated (e.g., a correlation coefficient > 0.7 is used as a reference threshold). When the correlation coefficient between a candidate peak and the personal pulse wave template is greater than the preset threshold, the match is considered successful, and the peak is retained as a pulse wave feature point (i.e., the core J-wave peak). If the correlation coefficient is less than or equal to the preset threshold, it is considered noise or an artifact and is discarded to ensure the accuracy of pulse wave feature point identification. This combination of dynamic threshold selection and personalized template matching avoids misjudgments caused by interference signals while adapting to the user's personalized pulse wave characteristics.

[0050] In this embodiment, a method of dynamically extracting the signal envelope and adaptively updating the peak detection threshold is adopted. This method can flexibly adjust the detection standard in real time according to the changes in pulse wave amplitude, solving the problem of missed detection and false detection that are easy to occur when the fixed threshold is used. At the same time, the candidate peaks are matched and screened by combining the personal pulse wave template, which can accurately distinguish the real pulse wave main peak from noise and motion artifact interference peaks, improve the accuracy of pulse wave feature point recognition and anti-interference ability, and adapt to pulse wave detection scenarios under different wearing states and different physiological conditions.

[0051] For the heart rate calculation result based on pulse wave feature points in step 104 above, please refer to [link / reference needed]. Figure 4 , Figure 4 One embodiment of the calculation of heart rate results provided in this application includes: 401. Convert the heartbeat interval sequence between adjacent pulse wave characteristic points into an instantaneous heart rate sequence; Based on pulse wave feature points, the time interval between two adjacent effective pulse wave feature points is extracted to obtain a continuous heartbeat interval sequence, with the unit of heartbeat interval being seconds. Then, through a preset conversion formula, each heartbeat interval value is converted into the corresponding instantaneous heart rate value, ultimately forming an instantaneous heart rate sequence that corresponds one-to-one with the heartbeat interval sequence.

[0052] For example, if the time interval between two adjacent pulse wave feature points is 0.8 seconds, the corresponding instantaneous heart rate is 60 ÷ 0.8 = 75 beats / minute; if the heartbeat interval sequence is [0.78s, 0.80s, 0.82s, 0.79s], the converted instantaneous heart rate sequence is [76.9 beats / minute, 75.0 beats / minute, 73.2 beats / minute, 75.9 beats / minute], ensuring that there is corresponding instantaneous heart rate data for each heartbeat cycle, fully reflecting the real-time frequency changes of the heartbeat.

[0053] 402. Perform sliding median filtering on the instantaneous heart rate sequence to remove abnormal heart rate values ​​caused by interference, and output the heart rate result; By employing sliding median filtering, abnormal heart rate values ​​caused by interference (such as incompletely removed motion artifacts and environmental disturbances) in the instantaneous heart rate sequence are eliminated, resulting in a stable and accurate heart rate output. Specifically, a sliding window of 3-5 heart rate data points is selected, which can be adjusted according to the actual signal quality. The instantaneous heart rate sequence is moved segment by segment along the sliding window, and the median of the instantaneous heart rate values ​​within each window is calculated. The calculated median is used as the filtered heart rate value at the center of that window. In this way, isolated abnormal heart rate values ​​can be suppressed while preserving the overall trend of the heart rate sequence, avoiding over-filtering that could distort the heart rate data.

[0054] 403. Based on the intercardiac interval sequence, calculate the corresponding time-domain and frequency-domain indices respectively; Based on the heart rate interval sequence, calculate the time-domain and frequency-domain indices related to heart rate variability: 1. Time-domain index calculation: Based on the complete heartbeat interval sequence, multiple time-domain parameters reflecting heart rate fluctuations are calculated. Core indices include: SDNN (standard deviation of all normal sinus intervals), reflecting the overall fluctuation of all heartbeat intervals. Abnormal heartbeat interval values ​​need to be removed during calculation, and then the standard deviation of the remaining heartbeat interval values ​​is calculated; RMSSD (root mean square of the difference between adjacent normal sinus intervals), reflecting the short-term fluctuation of adjacent heartbeat intervals. During calculation, the difference between adjacent heartbeat interval values ​​is first calculated, the difference is squared, the average value is taken, and then the square root is calculated; pNN50 (percentage of the number of adjacent normal sinus intervals with a difference greater than 50ms out of the total number of heartbeats), reflecting the proportion of significant fluctuations in adjacent heartbeat intervals.

[0055] 2. Frequency domain index calculation: First, Fourier transform is performed on the heartbeat interval sequence to convert the time-domain heartbeat interval sequence into a frequency domain signal, and the energy distribution of different frequency components is obtained. Then, the low-frequency band (LF) and high-frequency band (HF) are divided, and the energy values ​​of the two frequency bands are calculated. Finally, the core frequency domain index is obtained, that is, the ratio of low-frequency components to high-frequency components. This index can reflect the regulatory function of the autonomic nervous system on the heart.

[0056] 404. Obtain the heart rate variability results from the heart rate results based on the time domain index and the frequency domain index.

[0057] The various time-domain and frequency-domain indicators were standardized and combined with normal physiological reference ranges to determine whether each indicator was within the normal range. Then, based on the abnormalities of each indicator, the overall level of heart rate variability was comprehensively assessed—for example, decreased SDNN and RMSSD indicate decreased heart rate variability and reduced cardiac rhythm stability; an increased LF / HF ratio indicates sympathetic nerve dominance; and a decreased LF / HF ratio indicates parasympathetic nerve dominance. Finally, the analysis results of all indicators were integrated to output a complete heart rate variability result, specifying the level of heart rate variability (e.g., normal, mildly decreased, moderately decreased).

[0058] Optionally, based on the signal processing and feature extraction (including HRV results) described above, cardiovascular parameters can be further extracted to provide users with a more comprehensive and accurate reference for cardiovascular health management. The specific extraction methods and core significance of each parameter are as follows: 1. Pulse wave transit time (PTT): Based on the accelerometer pulse wave J-peak and PPG pulse wave peak identified in step 302, the time difference between the two peaks is calculated, i.e., the pulse wave transit time. This parameter is a core reference indicator for cuffless blood pressure estimation. Through the correlation model between pulse wave transit time and blood pressure, non-invasive blood pressure monitoring can be achieved, providing a convenient reference for daily blood pressure management.

[0059] 2. Arterial Stiffness Index (ASI): Based on the pulse wave systolic rise time extracted in step 208 and combined with the pulse wave waveform features identified in step 207, the ASI is calculated using a preset calculation algorithm (such as a fitting algorithm based on waveform slope and peak change rate). This index directly reflects the stiffness of arteries; a higher ASI value indicates greater arterial stiffness and poorer elasticity, making it an important screening indicator for arteriosclerosis.

[0060] 3. Vascular age assessment: Extract the peak values ​​of the a wave and e wave from the acceleration pulse wave (APG) calculated in step 208, and calculate the ratio between the two (ae wave ratio); Combined with the clinically recognized vascular age reference standard, compare the a wave ratio with the standard threshold to complete the assessment of the user's vascular age and intuitively reflect the degree of vascular aging.

[0061] 4. Autonomic Nervous System Function Assessment: Based on the HRV frequency domain index LF / HF ratio calculated in step 403, and combined with the results of HRV nonlinear analysis (such as fractal dimension, entropy value, etc.), an autonomic nervous system function assessment model is constructed to comprehensively assess the user's autonomic nervous system regulatory function. An increased LF / HF ratio indicates sympathetic nervous system dominance, while a decreased ratio indicates parasympathetic nervous system dominance, reflecting the homeostatic state of the cardiovascular system.

[0062] 5. Respiratory rate extraction: Based on the available finger artery pulse wave signal obtained in step 206, the respiratory modulation law of the pulse wave amplitude is analyzed. Respiratory movement can cause periodic fluctuations in the pulse wave amplitude. Through signal filtering and period recognition algorithm, the fluctuation period is captured and then converted into the user's respiratory rate, realizing synchronous and non-invasive monitoring of respiratory status.

[0063] In this embodiment, by converting the inter-heart rate sequence into an instantaneous heart rate sequence and using a sliding median filter to remove abnormal heart rate values ​​caused by interference, the heart rate distortion caused by artifact interference can be effectively suppressed, significantly improving the stability and accuracy of heart rate detection. At the same time, the time-domain and frequency-domain indices of heart rate variability are calculated simultaneously and the heart rate variability results are obtained by combining them. This can reflect the heart rhythm fluctuations and autonomic nervous system regulation state in a multi-dimensional and complete manner, realizing high-precision, automated integrated analysis of heart rate and heart rate variability. The output results are stable and reliable, and can provide accurate data support for subsequent comprehensive cardiovascular health assessment.

[0064] In step 101 above, the user's raw digital artery pulse wave signal is monitored. Please refer to [link to relevant documentation]. Figure 5 , Figure 5 An embodiment of the process for monitoring a user's raw digital artery pulse wave signal provided in this application includes: 501. During the monitoring of the user's original finger artery pulse wave signal, when the user's heart rate abnormality, heartbeat interval abnormality or the preset calibration conditions are detected, the PPG monitoring mode is activated, and the monitoring results are checked and calibrated. The monitoring mode of the accelerometer is used during the monitoring of the user's original finger artery pulse wave signal. This step, based on the monitoring results from the accelerometer monitoring mode described above, activates the PPG monitoring mode for verification and calibration through anomaly identification and preset condition judgment, thus achieving proactive switching of the monitoring mode. Specifically, in accelerometer monitoring mode, the user's raw finger artery pulse wave signal is continuously monitored, and heart rate results and heart rate interval sequences are acquired simultaneously. Anomalies in both are judged in real time, and it is also monitored whether preset calibration conditions are met. The triggering logic is as follows: 1. Abnormal trigger: When any of the following abnormal situations are detected, the PPG monitoring mode will be started immediately: (1) Abnormal heart rate: The heart rate result exceeds the preset normal physiological range (e.g., heart rate <60 beats / minute or >100 beats / minute in the resting state, heart rate >180 beats / minute in the exercise state), or the heart rate fluctuation amplitude exceeds 20 beats / minute in a short period of time (e.g., within 10 seconds); (2) Abnormal heartbeat interval: The heartbeat interval sequence shows obvious abnormal fluctuations, or the heartbeat interval value continues to deviate from the normal range, indicating that there may be an abnormality in the heart rhythm, which needs to be checked through the PPG monitoring mode.

[0065] 2. Preset calibration conditions trigger: To ensure the accuracy of long-term monitoring, preset calibration conditions are set. When any of the following conditions are met, the PPG monitoring mode is started for calibration: (1) Time calibration: The acceleration sensor monitoring mode runs continuously for a preset duration (e.g., 30 minutes) and calibration is started periodically to avoid monitoring deviation caused by sensor drift; (2) Signal quality calibration: If the quality of the usable finger artery pulse wave signal is in a "slightly abnormal" state for three consecutive times in steps 207-208, the signal processing parameters need to be calibrated through the PPG monitoring mode; (3) User operation trigger: When the user manually initiates the calibration command, the PPG monitoring mode is started.

[0066] After activating the PPG monitoring mode, the wearable device simultaneously turns on its built-in PPG sensor to collect the user's finger artery PPG signal (photoplethysmography pulse wave signal), while retaining the monitoring data from the accelerometer. The monitoring results (heart rate, pulse wave feature points, etc.) of the two modes are compared and verified to calibrate the signal processing parameters, feature point recognition threshold, and heart rate calculation algorithm in the accelerometer monitoring mode, ensuring the monitoring accuracy of the subsequent accelerometer monitoring mode. After verification, it automatically returns to the accelerometer monitoring mode, realizing a calibration closed loop.

[0067] 502. Determine whether the user is at rest or in motion based on the stability of the signal; Using the available finger artery pulse wave signal in the triaxial accelerometer monitoring mode as the core judgment basis, and combined with the motion signal collected by the triaxial accelerometer, the signal stability is analyzed from three dimensions: (1) Signal amplitude stability: calculate the standard deviation of the pulse wave signal amplitude for 10 consecutive heartbeat cycles. If the standard deviation is less than the preset threshold (such as 10% of the average signal amplitude), the signal amplitude is determined to be stable; (2) Signal period stability: calculate the coefficient of variation of 10 consecutive heartbeat intervals. If the coefficient of variation is less than 5%, the signal period is determined to be stable; (3) High frequency interference ratio: calculate the energy ratio of the high frequency component >15Hz in the signal through spectrum analysis. If the ratio is less than 3%, the signal is determined to be less affected by motion interference.

[0068] If the signal amplitude and period are stable and the proportion of high-frequency interference is below the threshold, the user is determined to be in a resting state; if the signal amplitude is unstable, the period fluctuates greatly, and the proportion of high-frequency interference is above the threshold, the user is determined to be in a moving state; if a single-dimensional anomaly occurs (such as only amplitude instability), the high-frequency component changes and variance mutations in step 205 are combined for further verification.

[0069] 503. If in a resting state, maintain the accelerometer monitoring mode; When the user is determined to be in a resting state, the accelerometer monitoring mode remains unchanged. In the resting state, the user's limb movement is minimal, and motion artifact interference is weak. The signal processing flow in steps 201-208 can effectively filter out slight interference such as environmental vibrations and obtain high-quality usable finger artery pulse wave signals. At the same time, compared with the PPG monitoring mode, the accelerometer monitoring mode has the advantages of low power consumption, strong resistance to ambient light interference, and low wearing requirements (no need for strict skin contact), making it suitable for long-term resting monitoring.

[0070] During the accelerometer monitoring mode, the device monitors the raw digital artery pulse wave signal in real time, extracts feature points, calculates heart rate and heart rate variability results, and continuously monitors signal stability and various abnormal indicators. If heart rate abnormality, heartbeat interval abnormality, or the preset calibration conditions are met, the PPG monitoring mode is still started according to step 501 for verification and calibration to ensure the long-term accuracy of monitoring results in the resting state.

[0071] 504. If in motion, switch to PPG monitoring mode.

[0072] To address the interference characteristics during motion, a more suitable PPG monitoring mode is switched to resolve the issue of insufficient accuracy in the accelerometer monitoring mode during motion. When the user is detected to be in motion, signal acquisition and processing in the accelerometer monitoring mode are immediately stopped, and the system switches to PPG monitoring mode. Specific operation and advantages are as follows: 1. Mode Switching Operation: The device automatically shuts down the signal acquisition of the accelerometer, starts the PPG sensor, adjusts the sensor sampling frequency (to be consistent with the accelerometer sampling frequency, such as 100Hz), and acquires the user's finger artery PPG signal; at the same time, it calls the PPG signal processing algorithm adapted to the motion state to filter motion artifacts (such as using an adaptive filtering algorithm to specifically suppress light signal interference caused by limb movement), extracts PPG pulse wave feature points, and calculates heart rate and heart rate variability results to ensure the validity of the monitoring data under motion conditions.

[0073] 2. Key Advantages of Switching: During exercise, the intense vibrations generated by limb movement can cause severe distortion of the pulse wave signal collected by the accelerometer. Even after artifact removal processing, it is still difficult to completely eliminate the interference, which can easily lead to inaccurate heart rate calculation. In contrast, the PPG monitoring mode collects pulse waves through optical signals, which is less affected by limb vibrations. Combined with signal processing algorithms adapted to the exercise state, it can improve the monitoring accuracy during exercise and avoid the problems of missed detections and false detections in the accelerometer monitoring mode during exercise.

[0074] After switching to PPG monitoring mode, the device continuously monitors the user's motion status and signal stability. When it determines that the user has switched back to resting mode, it automatically switches back to accelerometer monitoring mode, realizing intelligent and seamless switching between the two modes. This balances the low power consumption of resting mode with the high accuracy of motion mode, ensuring the reliability and accuracy of finger artery pulse wave monitoring in all scenarios.

[0075] In this embodiment, the intelligent switching logic of the monitoring mode achieves the complementary advantages of the accelerometer monitoring mode and the PPG monitoring mode. PPG calibration is initiated through abnormal triggering and preset calibration conditions, effectively correcting the drift and deviation of the accelerometer monitoring mode and improving the accuracy of long-term monitoring. Motion state is accurately judged based on signal stability, avoiding misjudgments during mode switching, ensuring low power consumption and high stability in the resting state, and high accuracy and anti-interference in the moving state. Seamless switching between the two modes adapts to different usage scenarios, solving the problem that a single monitoring mode cannot meet the monitoring needs of all scenarios, and further improving the finger artery pulse wave monitoring system.

[0076] Please see Figure 6 This application provides an embodiment of a heart rate monitoring device, comprising: The monitoring unit 601 is used to monitor the user's original digital artery pulse wave signal, which is the vibration signal generated by the digital artery pulse pressure pulsation directly transmitted to the wearable device through soft tissue. Optionally, a starting unit 602 is also included, for: During the monitoring of the user's raw digital artery pulse wave signal, when the user's heart rate abnormality, heart rate interval abnormality, or the preset calibration conditions are detected, the PPG monitoring mode is activated, and the monitoring results are checked and calibrated. The monitoring of the user's raw digital artery pulse wave signal is carried out using the accelerometer monitoring mode.

[0077] Optionally, a judgment unit 603 is also included, used for: Determine whether the user is at rest or in motion based on the stability of the signal; Optionally, a holding unit 604 is also included for: If in a resting state, the accelerometer monitoring mode will remain active. Optionally, a switching unit 605 is also included, for: If the device is in motion, switch to PPG monitoring mode.

[0078] The processing unit 606 is used to perform signal enhancement and motion artifact suppression processing on the original digital artery pulse wave signal to obtain a usable digital artery pulse wave signal. Optionally, the processing unit 606 is also used for: A covariance matrix is ​​constructed from the original digital artery pulse wave signal and principal component decomposition is performed. The target principal component is extracted as the principal direction component. The target principal component is the part with the highest energy proportion in the original digital artery pulse wave signal after principal component decomposition. The main direction component is bandpass filtered to retain the effective frequency band in the original digital artery pulse wave signal that matches the physiological characteristics of the pulse wave, thus completing the signal enhancement. Spectral analysis was performed on the enhanced original digital artery pulse wave signal to identify environmental vibration interference components. Suppress environmental vibration interference components; Based on the changes in high-frequency components and signal variance in the original finger artery pulse wave signal after signal enhancement, it is determined whether the current user is in motion. If so, artifact labeling and removal are performed on the original digital artery pulse wave signal under motion to obtain a usable digital artery pulse wave signal.

[0079] Optionally, an analysis unit 607 is also included, for: Morphological analysis of available digital artery pulse wave signals; Optionally, an extraction unit 608 is also included, for: Extract systolic rise time, dicrotic notch location and depth, pulse wave area ratio, and second derivative waveform characteristics, and assess arterial compliance, peripheral resistance, and vascular health status.

[0080] The identification unit 609 is used to identify pulse wave feature points from available finger artery pulse wave signals; Optionally, the identification unit 609 is also used for: Extract the signal envelope of the usable finger artery pulse wave signal, determine the dynamic peak detection threshold based on the amplitude of the signal envelope, and recalculate and automatically update the dynamic peak detection threshold every few heartbeat cycles. The waveform peaks above the dynamic peak detection threshold in the available finger artery pulse wave signals are matched with a preset personal pulse wave template, and the qualified matching peaks are retained as pulse wave feature points.

[0081] The calculation unit 610 is used to calculate heart rate results based on pulse wave feature points, and the heart rate results include heart rate variability results.

[0082] Optionally, the computing unit 610 is also used for: The heartbeat interval sequence between adjacent pulse wave feature points is converted into an instantaneous heart rate sequence; Perform sliding median filtering on the instantaneous heart rate sequence to remove abnormal heart rate values ​​caused by interference, and output the heart rate result; Based on the intercardia sequence, the corresponding time-domain and frequency-domain indices are calculated respectively. The heart rate variability results are obtained from the time-domain and frequency-domain indicators.

[0083] For detailed implementation methods, please refer to... Figures 1 to 5 Examples are not detailed here.

[0084] Please see Figure 7 This application provides a heart rate monitoring device, comprising: Processor 701, memory 702, input / output unit 703, and bus 704.

[0085] The processor 701 is connected to the memory 702, the input / output unit 703, and the bus 704.

[0086] The memory 702 stores a program, and the processor 701 calls the program to execute it, such as... Figure 1 , Figure 2 , Figure 3 , Figure 4 as well as Figure 5 The method in the middle.

[0087] This application provides a computer-readable storage medium on which a program is stored, and when the program is executed on a computer, it performs the following... Figure 1 , Figure 2 , Figure 3 , Figure 4 as well as Figure 5 The method in the middle.

[0088] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0089] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0090] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0091] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0092] If the integrated unit is implemented as 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 technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A heart rate monitoring method, applied to a wearable device, characterized in that, include: The device monitors the user's original digital artery pulse wave signal, which is a vibration signal generated by the pulse pressure of the digital artery being directly transmitted to the wearable device through soft tissue. The original digital artery pulse wave signal is processed with signal enhancement and motion artifact suppression to obtain a usable digital artery pulse wave signal; Identify pulse wave feature points from the available digital artery pulse wave signals; Heart rate results are calculated based on the pulse wave feature points, and the heart rate results include heart rate variability results.

2. The heart rate monitoring method according to claim 1, characterized in that, The process of enhancing and suppressing motion artifacts in the original digital artery pulse wave signal to obtain a usable digital artery pulse wave signal includes: A covariance matrix is ​​constructed on the original digital artery pulse wave signal and principal component decomposition is performed. The target principal component is extracted as the principal direction component. The target principal component is the part with the highest energy proportion in the original digital artery pulse wave signal after principal component decomposition. The main direction component is bandpass filtered to retain the effective frequency band in the original digital artery pulse wave signal that matches the physiological characteristics of the pulse wave, thus completing the signal enhancement. Spectral analysis was performed on the enhanced original digital artery pulse wave signal to identify environmental vibration interference components. The environmental vibration interference components are suppressed. Based on the changes in high-frequency components and signal variance in the original finger artery pulse wave signal after signal enhancement, it is determined whether the user is currently in motion. If so, artifact labeling and removal are performed on the original digital artery pulse wave signal under motion to obtain a usable digital artery pulse wave signal.

3. The heart rate monitoring method according to claim 1, characterized in that, The step of identifying pulse wave feature points from the available finger artery pulse wave signal includes: The signal envelope of the available finger artery pulse wave signal is extracted, and a dynamic peak detection threshold is determined based on the amplitude of the signal envelope. The dynamic peak detection threshold is recalculated and automatically updated every few heartbeat cycles. The waveform peaks in the available finger artery pulse wave signals that are higher than the dynamic peak detection threshold are matched with a preset personal pulse wave template, and the peaks that are matched are retained as pulse wave feature points.

4. The heart rate monitoring method according to claim 1, characterized in that, The calculation of heart rate based on the pulse wave feature points includes: The heartbeat interval sequence between adjacent pulse wave feature points is converted into an instantaneous heart rate sequence; Perform sliding median filtering on the instantaneous heart rate sequence to remove abnormal heart rate values ​​caused by interference, and output the heart rate result; Based on the intercardia sequence, the corresponding time-domain and frequency-domain indices are calculated respectively. The heart rate variability results in the heart rate results are obtained based on the time domain index and the frequency domain index.

5. The heart rate monitoring method according to claim 1, characterized in that, After obtaining a usable digital artery pulse wave signal, the method further includes: Morphological analysis was performed on the available digital artery pulse wave signals; Extract systolic rise time, dicrotic notch location and depth, pulse wave area ratio, and second derivative waveform characteristics, and assess arterial compliance, peripheral resistance, and vascular health status.

6. The heart rate monitoring method according to any one of claims 1 to 5, characterized in that, The method further includes: During the monitoring of the user's raw digital artery pulse wave signal, when the user's heart rate abnormality, heart rate interval abnormality, or the preset calibration conditions are detected, the PPG monitoring mode is activated, and the monitoring results are checked and calibrated. The monitoring of the user's raw digital artery pulse wave signal adopts the accelerometer monitoring mode.

7. The heart rate monitoring method according to claim 6, characterized in that, The method further includes: The stability of the signal determines whether the user is at rest or in motion. If in a resting state, the accelerometer monitoring mode is maintained; If the device is in motion, switch to the PPG monitoring mode.

8. A heart rate monitoring device, characterized in that, The device includes: The monitoring unit is used to monitor the user's original digital artery pulse wave signal, which is the vibration signal generated by the digital artery pulse pressure pulsation directly transmitted to the wearable device through soft tissue. The processing unit is used to perform signal enhancement and motion artifact suppression processing on the original digital artery pulse wave signal to obtain a usable digital artery pulse wave signal; The identification unit is used to identify pulse wave feature points from the available finger artery pulse wave signals; A calculation unit is used to calculate heart rate results based on the pulse wave feature points, the heart rate results including heart rate variability results.

9. A heart rate monitoring device, characterized in that, The device includes: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to execute the heart rate monitoring method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a program stored thereon, the program performing the heart rate monitoring method as described in any one of claims 1 to 7 when executed on a computer.

Images