Blood pressure compensation measuring method and device for motion scene and smart watch

Abstract

The invention relates to a blood pressure compensation measurement method and device for a motion scene and a smart watch, and the method comprises the steps: generating a lactic acid accumulation mark through fusing a motion acceleration signal and a heart rate signal, extracting a specific frequency band component of a pressure shock wave signal based on the mark, and generating a time domain envelope line; quantifying envelope line main peak width change and frequency domain energy distribution abnormity to generate tension abnormity, and mapping the tension abnormity into a blood vessel volume dynamic change curve; constructing a biomechanical compensation curve in combination with a static blood vessel volume reference, and reconstructing an envelope line for eliminating blood vessel tension interference through frequency domain convolution; and finally, positioning a blood pressure feature point based on the reconstructed envelope differential feature and outputting a compensated blood pressure value. According to the method, a lactic acid metabolism and vascular biomechanical characteristic compensation model is established for the first time, the problem of systematic offset of feature points caused by abnormal vascular tension after movement is solved, and the blood pressure measurement error is remarkably reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blood pressure compensation measurement, in particular to a blood pressure compensation measurement method and device for a motion scene and a smart watch. BACKGROUND

[0002] In the field of health monitoring, blood pressure measurement in a motion scene is of great significance for evaluating cardiovascular function and ensuring safety during exercise. Traditional oscillometric blood pressure measurement techniques rely on the identification of characteristic points of the cuff pressure oscillation wave to determine systolic and diastolic blood pressure by analyzing the maximum slope point and curvature change point of the envelope curve. This method has good reliability in a resting state and has been widely used in household blood pressure monitors and medical devices. With the development of wearable devices, there is an increasing demand for real-time blood pressure monitoring during exercise, especially in fitness training and athlete physiological monitoring scenarios. Accurate blood pressure data after exercise is crucial for preventing cardiovascular accidents.

[0003] However, existing blood pressure measurement techniques face significant limitations in post-exercise scenarios. When the human body performs anaerobic exercise, the accumulation of lactic acid produced by muscle tissue causes abnormal changes in blood vessel tension, leading to distortion of the envelope curve of the cuff pressure oscillation wave. This physiological interference manifests as characteristic point shifts, envelope curve broadening, and abnormal spectral energy distribution, making it difficult for traditional algorithms to accurately locate blood pressure characteristic points. Existing solutions attempt to suppress limb motion interference through motion artifact filtering or acceleration compensation, but fail to address the core issue of changes in blood vessel biomechanical properties caused by lactic acid metabolism, resulting in systematic deviations in measurement results. More critically, existing techniques lack the ability to quantitatively model dynamic changes in blood vessel tension, making it impossible to establish a compensation mechanism between lactic acid accumulation and blood pressure measurement distortion, and there is a risk of misleading blood pressure data obtained by users after exercise. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a blood pressure compensation measurement method, device and smart watch for a motion scene that can accurately eliminate lactic acid accumulation interference.

[0005] The purpose of the present application is achieved by the following scheme: In a first aspect, the present application provides a blood pressure compensation measurement method for a motion scene, comprising the following steps: S1: processing motion acceleration signals collected by a three-axis acceleration sensor and heart rate signals collected by a heart rate sensor, calculating an acceleration-based exercise intensity index and a heart rate recovery slope based on heart rate variability, and generating a lactic acid accumulation flag based on a pre-set intensity threshold and a pre-set recovery threshold; S2: processing a cuff pressure oscillation wave signal collected by a pressure sensor, extracting an intrinsic mode function in the 0.5-5Hz frequency band and performing Hilbert transform to generate a time-domain envelope curve; S3: processing the time-domain envelope, calculating the change amount of the main peak width of the envelope relative to the static reference and the energy ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generating a tension abnormality degree representing abnormal tension of blood vessels; S4: processing the tension abnormality degree, calculating a blood vessel diastolic volume through a non-linear mapping function, and constructing a blood vessel volume compensation curve in combination with an exponential decay time function; S5: processing the time-domain envelope and the blood vessel volume compensation curve, reconstructing the compensated envelope through frequency-domain convolution operation and positioning the maximum slope point and the second derivative positive point on the reconstructed envelope, and outputting a compensated blood pressure value for eliminating motion interference.

[0006] In one of the embodiments, the S1 of the blood pressure compensation measurement method for a motion scene provided by the application specifically includes the following steps: S11: Cubic root integration is performed on the motion acceleration signal collected by the three-axis acceleration sensor, the integral value of the cubic root of the three-axis combined acceleration in the time window is calculated, and a motion intensity index is generated; S12: Linear regression processing is performed on the heart rate signal collected by the heart rate sensor, the slope value of the heart rate decline rate after the motion stops is calculated, and a heart rate recovery slope is generated; S13: Double-threshold logical judgment is performed on the motion intensity index and the heart rate recovery slope, when the motion intensity exceeds the preset intensity threshold and the heart rate recovery rate is lower than the preset recovery threshold, an activated lactic acid accumulation flag is generated.

[0007] In one of the embodiments, the S2 of the blood pressure compensation measurement method for a motion scene provided by the application specifically includes the following steps: S21: Empirical mode decomposition processing is performed on the cuff pressure oscillation wave signal collected by the pressure sensor, the signal is decomposed into a plurality of intrinsic mode function components, and the intrinsic mode function in the 0.5-5Hz frequency band is extracted; S22: Hilbert transform processing is performed on the intrinsic mode function, the instantaneous amplitude of the analytic signal is calculated, and an initial envelope is generated; S23: Smooth filtering processing is performed on the initial envelope, high-frequency fluctuation interference is eliminated through mean value calculation in the time window, and an optimized time-domain envelope is generated.

[0008] In one of the embodiments, the S3 of the blood pressure compensation measurement method for a motion scene provided by the application specifically includes the following steps: S31: Main peak detection processing is performed on the time-domain envelope, the global maximum value point of the envelope is located and its half-width is measured, and a main peak width change amount is generated; S32: Perform frequency domain energy analysis processing on the time domain envelope, calculate the energy distribution ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generate a frequency domain feature ratio; S33: Perform weighted fusion processing on the main peak width variation and the frequency domain feature ratio, calculate the comprehensive deviation in combination with the static reference value, generate the tension abnormality degree, and the tension abnormality degree is used to indicate the abnormality degree of blood vessel tension caused by lactic acid accumulation.

[0009] In one of the embodiments, the present application provides a calculation formula of the tension abnormality degree of the blood pressure compensation measurement method of the motion scene: ; Wherein, is the tension abnormality degree, representing the abnormality degree of the blood vessel tension, is the main peak width variation, is the static reference half-width, is the frequency domain feature ratio, is the time domain feature weight factor, is the frequency domain feature weight factor.

[0010] In one of the embodiments, the S4 of the blood pressure compensation measurement method of the motion scene provided by the present application specifically includes the following steps: S41: Perform non-linear function mapping processing on the tension abnormality degree, convert the tension abnormality degree into a blood vessel volume change ratio through hyperbolic tangent transformation, and generate a blood vessel diastolic volume; S42: Perform exponential decay modeling processing on the blood vessel diastolic volume, simulate the dynamic recovery process of the blood vessel volume in the lactic acid metabolism process by applying a time decay function, and construct a volume change curve that decreases with time; S43: Perform baseline superposition processing on the volume change curve, add the volume change curve to the preset static blood vessel volume reference value, generate a blood vessel volume compensation curve, and the blood vessel volume compensation curve is used for subsequent frequency domain filtering processing to eliminate the interference of the blood vessel tension abnormality.

[0011] In one of the embodiments, the S5 of the blood pressure compensation measurement method of the motion scene provided by the present application specifically includes the following steps: S51: Perform frequency domain convolution processing on the time domain envelope and the blood vessel volume compensation curve, multiply the envelope spectrum by the compensation filter transfer function in the frequency domain and implement inverse Fourier transform conversion to the time domain, and generate a reconstructed compensation envelope; S52: Perform differential feature extraction processing on the reconstructed compensation envelope, find the maximum value point by calculating the first derivative, and detect the position point where the second derivative first changes from negative to positive, and generate the systolic pressure feature time and the diastolic pressure feature time; S53: Perform pressure value mapping processing on the systolic pressure feature moment and the diastolic pressure feature moment, query the cuff pressure value at the corresponding moment according to the feature moment, and generate a compensated blood pressure value containing a compensated systolic pressure value and a compensated diastolic pressure value.

[0012] In one of the embodiments, the blood pressure compensation measurement method for a motion scene provided by the application specifically comprises the following steps: S521: Perform first derivative calculation processing on the reconstructed compensation envelope, calculate the instantaneous slope value of each sampling point on the envelope by the central difference method, and generate a first derivative sequence; S522: Perform extreme point detection processing on the first derivative sequence, scan the entire sequence to find the maximum positive point and record its corresponding moment, and generate a systolic pressure feature moment; S523: Perform second derivative calculation processing on the reconstructed compensation envelope, calculate the curvature change value of each sampling point on the envelope by the second central difference method, and search for the first point where the curvature changes from negative to positive after the systolic pressure feature moment, and generate a diastolic pressure feature moment.

[0013] In a second aspect, the application provides a blood pressure compensation measurement device for a motion scene, which is configured with the following modules: A lactic acid accumulation sign generation module for processing the motion acceleration signal collected by the three-axis acceleration sensor and the heart rate signal collected by the heart rate sensor, calculating the motion intensity index based on acceleration and the heart rate recovery slope based on the heart rate change rate, and generating a lactic acid accumulation sign based on the preset intensity threshold and the preset recovery threshold; A time domain envelope generation module for processing the cuff pressure oscillation wave signal collected by the pressure sensor, extracting the intrinsic mode function in the 0.5-5Hz frequency band and performing Hilbert transform, and generating a time domain envelope; A tension abnormality calculation module for processing the time domain envelope, calculating the change amount of the envelope main peak width relative to the static reference and the energy ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generating a tension abnormality degree representing the abnormality of the blood vessel tension; A blood vessel volume compensation curve construction module for processing the tension abnormality degree, calculating the blood vessel diastolic volume obtained by a nonlinear mapping function, and constructing a blood vessel volume compensation curve in combination with an exponential decay time function; A compensated blood pressure value output module for processing the time domain envelope and the blood vessel volume compensation curve, reconstructing the compensated envelope by frequency domain convolution operation and positioning the maximum slope point and the second derivative positive change point on the reconstructed envelope, and outputting a compensated blood pressure value for eliminating motion interference.

[0014] In a third aspect, the present application provides a smart watch, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the blood pressure compensation measurement method in any of the above motion scenarios when executing the computer program.

[0015] To sum up, the blood pressure compensation measurement method in a motion scenario provided by the present application is based on a double-threshold judgment mechanism of acceleration signal cubic root integral and heart rate recovery slope, which can accurately identify the physiological state changes caused by lactic acid accumulation, overcoming the misjudgment defects caused by the traditional method which only relies on a single motion sensor. Secondly, by extracting the specific frequency band components of the pressure oscillation wave and combining the Hilbert envelope extraction technology, the signal distortion caused by abnormal vascular tension can be effectively separated, solving the problem of feature waveform submersion under motion interference. Thirdly, the tension abnormality degree is innovatively constructed by combining the time domain envelope line main peak broadening feature and the frequency domain energy distribution abnormal feature, which can realize the quantitative representation of the change of vascular biomechanical properties. Most importantly, the vascular volume dynamic response model established by the nonlinear mapping and the exponential decay function can accurately simulate the recovery law of vascular tension in the lactic acid metabolism process, and the envelope line can be reconstructed by combining the frequency domain convolution compensation technology, which can completely eliminate the systematic deviation of blood pressure feature points caused by the change of vascular volume. Finally, the differential feature joint detection mechanism is used in the positioning link, which can ensure the stable capture of the real blood pressure feature points in the complex interference environment. This method first establishes a complete technical chain from lactic acid metabolism dynamics to blood pressure measurement compensation, which can maintain the measurement accuracy of millimeters of mercury level in the scene of abnormal vascular tension after exercise, and solves the core pain point of the inaccuracy of blood pressure monitoring of wearable devices in fitness rehabilitation, sports training and other scenarios.

[0016] For better understanding and implementation, the present application is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of a blood pressure compensation measurement method in a motion scenario provided by an embodiment of the present application; Figure 2 A flowchart of generating a tension abnormality degree provided by an embodiment of the present application; Figure 3 A flowchart of generating a compensation blood pressure value containing a compensation systolic pressure value and a compensation diastolic pressure value provided by an embodiment of the present application; Figure 4 A structural diagram of a blood pressure compensation measurement device in a motion scenario provided by another embodiment of the present application. DETAILED DESCRIPTION

[0018] For the purposes of the present invention, a more complete understanding can be obtained by reference to the following description taken in connection with the accompanying drawings. The drawings are intended to illustrate preferred embodiments of the present invention and are not intended to limit the scope of the invention. It is understood that the invention can be realized in many different forms and is not limited to the embodiments described herein. Rather, the purpose of the drawings is to illustrate the principles of the present invention.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0020] In one embodiment, as shown in Figure 1 In one embodiment, as shown in S1: processing the motion acceleration signal collected by the three-axis acceleration sensor and the heart rate signal collected by the heart rate sensor, calculating the motion intensity index based on acceleration and the heart rate recovery slope based on heart rate variability, and generating the lactic acid accumulation flag based on the preset intensity threshold and the preset recovery threshold.

[0021] Specifically, the system collects the motion acceleration signal through the three-axis acceleration sensor integrated in the smart watch, collects the heart rate signal through the optical heart rate sensor integrated in the smart watch, and performs filtering processing on the motion acceleration signal collected by the smart watch to remove limb tremor noise; performs adaptive filtering processing on the heart rate signal collected by the smart watch to eliminate motion artifacts, wherein the reference signal of the adaptive filtering is taken from the output of the three-axis acceleration sensor integrated in the smart watch.

[0022] Preferably, the system performs vector synthesis operation on the preprocessed three-axis acceleration signal, and then performs average value calculation on the synthesized signal using a sliding window to obtain the motion intensity index based on acceleration. The system determines the preset intensity threshold through a clinical experiment. During the clinical experiment, the system synchronously acquires the blood lactic acid concentration data of the subject through the monitoring unit associated with the smart watch, and determines the preset intensity threshold according to the corresponding relationship between the blood lactic acid concentration and the motion intensity index.

[0023] The system identifies the motion stop moment through the change of the acceleration signal, and determines that the motion stops when the motion intensity index decreases from above a certain level to below a certain level. The system extracts a heart rate sequence collected by the smart watch in a period after the motion stop moment, performs linear fitting operation on the heart rate sequence, and obtains a heart rate recovery slope. The system determines a preset recovery threshold value through clinical data, and determines the preset recovery threshold value according to the corresponding relationship between the heart rate recovery slope and the lactic acid metabolism state. When the motion intensity index reaches the preset intensity threshold value and the heart rate recovery slope reaches the preset recovery threshold value, the system sets the lactic acid accumulation flag to a specific state; otherwise, the system sets the lactic acid accumulation flag to another state.

[0024] S2: Process the cuff pressure oscillatory wave signal collected by the pressure sensor, extract the intrinsic mode function in the 0.5-5Hz frequency band and perform Hilbert transform to generate the time domain envelope.

[0025] Specifically, the system controls the piezoresistive pressure sensor of the smart watch to collect the pressure oscillatory wave signal in the process of cuff inflation and deflation, and the sensor measurement range covers the full range of cuff pressure change. The system performs Butterworth high-pass filtering on the original pressure oscillatory wave signal to remove the baseline drift caused by slow deflation of the cuff, and then performs low-pass filtering to eliminate high-frequency electromagnetic interference to obtain the preprocessed signal. The system performs empirical mode decomposition on the preprocessed signal, repeatedly calculates the upper and lower envelope lines of the signal and takes the mean value as the trend item, removes the trend item from the original signal, until the remaining signal meets the intrinsic mode function judgment condition, and separates the multi-order intrinsic mode function.

[0026] Illustratively, the system selects the frequency band corresponding to the intrinsic component of the blood pressure oscillatory wave, and excludes respiratory interference and muscle tremor interference, wherein the respiratory interference is lower than the intrinsic component frequency band, and the muscle tremor interference is higher than the intrinsic component frequency band; further, the system performs Hilbert transform on the screened intrinsic mode function, respectively, constructs an analytical signal with the real part being the original intrinsic mode function and the imaginary part being the Hilbert transform result, calculates the modulus value of each analytical signal, and takes the arithmetic mean of all instantaneous amplitudes to obtain the time domain envelope line representing the amplitude change characteristics of the cuff pressure oscillatory wave.

[0027] S3: Process the time domain envelope line, calculate the change amount of the main peak width of the envelope line relative to the static reference and the energy ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generate the tension abnormality degree representing the abnormality of vascular tension.

[0028] Specifically, the system repeatedly performs multiple blood pressure measurements in a resting state of the user to be detected, extracts a time-domain envelope line after each measurement, calculates the half-height width of the main peak of the envelope line, and takes the average of the multiple half-height widths as a static reference of the main peak width of the envelope line. When in a motion scene, the system calculates the first derivative of the real-time time-domain envelope line, locates the main peak position by finding the extreme value point of the first derivative changing from positive to negative, calculates the real-time half-height width of the main peak, and obtains the change amount relative to the static reference through difference and ratio operations. The system performs fast Fourier transform on the pre-processed pressure oscillation wave signal, divides two characteristic frequency bands after obtaining the frequency domain distribution, wherein the first frequency band corresponds to the high-frequency vibration of the vascular smooth muscle, and the energy changes when the tension is abnormal, and the second frequency band corresponds to the low-frequency vibration of the basic blood flow; the system performs integral operation on the two frequency bands to obtain energy values, obtains an energy ratio through ratio operation, and constructs a weighted summation calculation model based on the change amount and the energy ratio. The weight coefficient is calibrated through multivariate linear regression of multiple motion after blood pressure measurement samples, the change amount and the energy ratio are substituted into the model, and the tension abnormality degree representing the degree of abnormality of the vascular tension is obtained. The tension abnormality degree value range corresponds to different abnormality levels, and the higher the level, the stronger the compensation processing required.

[0029] S4: processing the tension abnormality degree, calculating the vascular diastolic volume through a non-linear mapping function, and constructing a vascular volume compensation curve in combination with an exponential decay time function.

[0030] Specifically, the present embodiment establishes the correlation between the vascular tension abnormality degree and the vascular diastolic volume through experiments, measures the change of the arterial inner diameter of the subject after exercise by using an ultrasonic Doppler device, synchronously records the vascular tension abnormality degree, and forms an experimental database; fitting operation is performed on the database data to obtain a non-linear mapping function, the function includes parameters corresponding to the maximum vascular diastolic volume of the radial artery of a healthy adult, and parameters ensuring that the diastolic volume tends to the maximum diastolic volume when the tension abnormality degree reaches the upper limit. Based on the above processing, the system substitutes the real-time tension abnormality degree into the function to obtain the additional diastolic volume of the blood vessel caused by lactic acid accumulation.

[0031] Specifically, the system defines an exponential decay time function based on the time dependence of lactic acid metabolism, the function includes an initial decay time constant corresponding to the lactic acid metabolism rate at the moment when the exercise stops, and a decay coefficient determined based on the experimental measurement results of the lactic acid half-life, the input is the time after the exercise stops, and the output is the decay coefficient reflecting the recovery trend of the vascular diastolic volume. A frequency parameter corresponding to the periodic characteristics of the cuff deflation rate is introduced, the vascular diastolic volume, the output value of the exponential decay time function, and the frequency-related periodic function value are combined through multiplication operation to construct a vascular volume compensation curve that dynamically adjusts with time, which is used to offset the influence of abnormal vascular volume on the time-domain envelope line.

[0032] S5: processing the time domain envelope and the blood vessel volume compensation curve, reconstructing the compensated envelope through a frequency domain convolution operation, positioning the maximum slope point and the second derivative positive change point on the reconstructed envelope, and outputting the compensated blood pressure value for eliminating motion interference.

[0033] Specifically, the system performs fast Fourier transform on the time domain envelope and the blood vessel volume compensation curve respectively, converts the time domain signal into a frequency domain signal, and obtains the frequency domain signal corresponding to the time domain envelope and the frequency domain signal corresponding to the blood vessel volume compensation curve. The system performs a product operation on the two frequency domain signals in the frequency domain, which is equivalent to the convolution operation in the time domain, and can avoid the edge effect in the time domain convolution process. The system performs inverse fast Fourier transform on the frequency domain signal after the product operation, converts the frequency domain signal back to the time domain signal, and obtains the reconstructed compensation envelope.

[0034] Further, the system performs a first derivative operation on the compensation envelope to obtain a first derivative signal of the compensation envelope, and then locates the maximum value point in the first derivative signal. The maximum value point is a systolic pressure feature point, corresponding to the oscillation wave amplitude change rate peak when the cuff pressure is balanced with the arterial systolic pressure. The system records the cuff static pressure at this time as the systolic pressure. The system performs a second derivative operation on the compensation envelope to obtain a second derivative signal of the compensation envelope, and then locates the zero-crossing point from negative to positive in the second derivative signal. The zero-crossing point is a diastolic pressure feature point, corresponding to the oscillation wave amplitude change rate inflection point when the cuff pressure is lower than the arterial diastolic pressure. The system records the cuff static pressure at this time as the diastolic pressure. Finally, the systolic pressure and the diastolic pressure are integrated to generate a compensation blood pressure value. The compensation blood pressure value can effectively eliminate motion interference, provide more accurate and reliable blood pressure measurement results for the user, and thus achieve the purpose of blood pressure compensation measurement in the motion scene.

[0035] In summary, the motion scene blood pressure compensation measurement method provided by the application is based on a double threshold judgment mechanism of acceleration signal cubic root integral and heart rate recovery slope, can accurately identify the physiological state change caused by lactic acid accumulation, and overcomes the misjudgment defect caused by the traditional method which only relies on a single motion sensor. Secondly, by extracting the specific frequency band component of the pressure oscillation wave and combining the Hilbert envelope extraction technology, the signal distortion caused by abnormal vascular tension can be effectively separated, and the problem of feature waveform submersion under motion interference is solved. Thirdly, the tension abnormality degree is innovatively constructed by combining the main peak broadening feature of the time domain envelope line and the abnormal feature of the frequency domain energy distribution, which can realize the quantitative representation of the change of the vascular biomechanical characteristics. Most importantly, the vascular volume dynamic response model is established by nonlinear mapping and exponential decay function, which can accurately simulate the recovery law of the vascular tension in the lactic acid metabolism process, and the envelope line is reconstructed by combining the frequency domain convolution compensation technology, which can completely eliminate the systematic deviation of the blood pressure feature points caused by the change of the vascular volume. Finally, the differential feature joint detection mechanism is used in the positioning link, which can ensure the stable capture of the real blood pressure feature points in the complex interference environment. The method first establishes a complete technical chain from lactic acid metabolism dynamics to blood pressure measurement compensation, can maintain millimeter mercury level measurement accuracy in the scene of abnormal vascular tension after exercise, and solves the core pain point of blood pressure monitoring inaccuracy of wearable devices in fitness rehabilitation, exercise training and other scenes.

[0036] In one of the embodiments, the S1 of the motion scene blood pressure compensation measurement method provided by the application specifically includes the following steps: S11: Cubic root integral is performed on the motion acceleration signal collected by the three-axis acceleration sensor, the integral value of the cubic root of the three-axis combined acceleration in the time window is calculated, and a motion intensity index is generated.

[0037] Specifically, the motion acceleration signal collection of the system is realized by a smart watch. The three-axis acceleration sensor integrated in the smart watch establishes a real-time data link with the embedded processor. After the sensor collects the motion acceleration signals corresponding to the wearing positions of the X-axis, Y-axis and Z-axis, the signals are directly transmitted to the processor. The system performs preprocessing on the three-axis acceleration signals collected by the smart watch, removes the limb tremor noise generated by the wrist non-purpose shaking by using a filtering algorithm, which will interfere with the effectiveness of the acceleration signal. After filtering, only the signal components related to the motion intensity are retained. After preprocessing, the system performs vector synthesis operation on the three-axis acceleration signals, integrates the signals of the three axes into a single combined acceleration signal through vector synthesis, avoids the strength misjudgment caused by the change of the motion direction of the single axis signal, and ensures that the combined signal can reflect the overall motion amplitude.

[0038] The system performs a cubic root operation on the synthetic acceleration signal, which can compress the numerical range of the synthetic acceleration signal while preserving the signal differences under different motion intensities, preventing excessive amplification of the signal at high acceleration values and masking of the signal at low acceleration values, and providing a stable signal basis for subsequent integral operations. The system sets a fixed time window, the window length of which is determined according to the signal change period under the motion state, to ensure that enough samples can be collected within the window to reflect the current motion intensity. The system performs an integral operation on the synthetic acceleration signal subjected to the cubic root operation within the set time window, and the integral result is the motion intensity index, which is positively correlated with the motion intensity.

[0039] S12: Linear regression processing is performed on the heart rate signal collected by the heart rate sensor to calculate the slope value of the heart rate decline rate after the motion stops, and a heart rate recovery slope is generated.

[0040] Specifically, the optical heart rate sensor integrated in the smart watch emits a light signal of a specific wavelength, and after the light signal penetrates the wrist skin, the sensor receives the light reflection signal generated by the blood flow and converts it into a heart rate electrical signal. The electrical signal is transmitted in real time to the embedded processor of the smart watch. Preferably, the system pre-processes the heart rate electrical signal collected by the smart watch, which can use an adaptive filtering algorithm to eliminate motion artifacts. Motion artifacts are caused by light reflection interference caused by the relative displacement of the wrist skin and the sensor and muscle contraction, which can cause false fluctuations in the heart rate signal. When filtering, the motion signal collected by the three-axis acceleration sensor of the smart watch is used as a reference to adjust the filtering parameters in real time, and the signal reflecting the true heart rate is obtained after removing the artifacts.

[0041] Illustratively, the system identifies the motion stop time according to the motion intensity index. When the motion intensity index decreases from above a certain dynamic threshold to below another dynamic threshold and this state is maintained for a certain length of time, the system determines that the motion has stopped, records and marks the time node of the stop time. After the motion stops, the system extracts the heart rate signal within a certain time period after the stop time from the local storage unit of the smart watch to form a continuous heart rate sequence, which covers the initial stage of the heart rate decline from the post-exercise peak to ensure that the sequence can reflect the initial trend of heart rate recovery. The system performs linear regression processing on the heart rate sequence with time as the independent variable and heart rate value as the dependent variable, establishes a linear regression model, calculates the slope value of the model to obtain the heart rate recovery slope, and the negative slope value indicates the heart rate decline, and the absolute value size reflects the recovery rate.

[0042] S13: Double-threshold logic judgment is performed on the motion intensity index and the heart rate recovery slope. When the motion intensity exceeds the preset intensity threshold and the heart rate recovery rate is lower than the preset recovery threshold, an activated lactic acid accumulation flag is generated.

[0043] Specifically, in one aspect, the system collects exercise intensity indexes under different exercise intensities through a smart watch, simultaneously monitors blood lactic acid concentration changes of the subject, records blood lactic acid concentration values corresponding to each exercise intensity index, and determines a preset intensity threshold value when the exercise intensity index reaches a certain value and the blood lactic acid concentration exceeds the base concentration in the resting state and continuously rises. In another aspect, the system collects heart rate recovery slopes of different subjects after exercise through a smart watch, simultaneously monitors corresponding blood lactic acid metabolism states, and determines a preset recovery threshold value when the heart rate recovery slope is lower than a certain value and the blood lactic acid metabolism rate significantly decreases and the accumulation state is difficult to quickly relieve.

[0044] Exemplarily, the system acquires the generated real-time exercise intensity index, compares it with the preset intensity threshold value, and judges whether it exceeds the threshold value; simultaneously acquires the generated heart rate recovery slope, compares it with the preset recovery threshold value, and judges whether it is lower than the threshold value. When the exercise intensity index exceeds the preset intensity threshold value and the heart rate recovery slope is lower than the preset recovery threshold value, the system determines that there is lactic acid accumulation, and generates a lactic acid accumulation marker in an activated state; if either condition is not met, the system determines that there is no significant lactic acid accumulation, and generates a lactic acid accumulation marker in a non-activated state.

[0045] In one of the embodiments, the blood pressure compensation measurement method for exercise scenes provided by the present application comprises the following steps: S21: The pressure sensor acquires the cuff pressure oscillation wave signal, and the signal is decomposed into multiple intrinsic mode function components through empirical mode decomposition processing, and the intrinsic mode function in the 0.5-5Hz frequency band is extracted.

[0046] Specifically, the system controls the pressure resistance pressure sensor inside the cuff to collect the pressure oscillation wave signal in the inflation and deflation process of the cuff, and the sensor measurement range covers the full range of cuff pressure changes. After the signal is output by the sensor, it is transmitted to the system main control unit. Further, the system pre-processes the original pressure oscillation wave signal, eliminates the baseline drift caused by slow deflation of the cuff and high-frequency electromagnetic interference through filtering operation, and obtains the pre-processed signal. At the same time, the system performs empirical mode decomposition on the pre-processed signal, first calculates the upper and lower envelope lines of the signal, fits the extreme points to generate the upper and lower envelope lines using the interpolation algorithm, and then takes the mean value of the upper and lower envelope lines as the trend item to remove the trend item from the original signal. The system repeats the above screening process until the remaining signal meets the intrinsic mode function judgment condition, i.e. the number of extreme points is equal to or differs by one from the number of zero-crossing points, and finally decomposes the pre-processed signal into multiple intrinsic mode function components. The system screens the component corresponding to the 0.5-5Hz frequency band from the multiple intrinsic mode function components obtained by decomposition according to the frequency band characteristics of the blood pressure oscillation wave intrinsic component. This frequency band is the main energy distribution interval of the blood pressure oscillation wave, and can exclude respiratory interference and muscle tremor interference, and is used as the target signal for subsequent processing.

[0047] S22: Hilbert transform is performed on the intrinsic mode function to calculate the instantaneous amplitude of the analytic signal and generate an initial envelope.

[0048] Specifically, the system calls the screened intrinsic mode function (IMF) data in the 0.5-5Hz frequency band, performs Hilbert transform on the IMF component, performs fast Fourier transform (FFT) on the time series data of the IMF component, and converts the time domain signal into a frequency domain signal. During the conversion process, the system supplements zero values according to the data length of the IMF component to ensure that the frequency domain resolution covers all frequency points in the 0.5-5Hz frequency band, avoiding loss of frequency information. Then, the converted frequency domain signal is multiplied by a sign function, which determines the value according to the positive and negative of the frequency value. When the frequency value is greater than zero, the value is 1, and when the frequency value is less than zero, the value is -1. This operation adjusts the phase characteristics of the frequency domain signal, so that the subsequent inverse transform can obtain a result that meets the definition of Hilbert transform. Finally, inverse fast Fourier transform (IFFT) is performed on the multiplied frequency domain signal to convert the frequency domain signal back to the time domain signal, which is the Hilbert transform result of the IMF component.

[0049] Further, the system combines the time series data of the original IMF component as the real part and the corresponding Hilbert transform result as the imaginary part to form an analytic signal. Each time point of the analytic signal contains two values of real and imaginary parts. The system performs instantaneous amplitude calculation on the analytic signal: first, calculate the square value of the real part and the square value of the imaginary part of each time point of the analytic signal, then add the two square values to obtain the square sum; then perform square root operation on the square sum, and the operation result is the instantaneous amplitude of the time point. The instantaneous amplitude reflects the amplitude of the cuff pressure oscillation wave at the corresponding time point, and is directly related to the strength of the arterial pulse. The system arranges the instantaneous amplitude of each time point in sequence according to the time sequence of the signal collected by the smart watch to form continuous time series data, which is the initial envelope.

[0050] S23: Smooth filtering is performed on the initial envelope to eliminate high-frequency fluctuation interference through mean value calculation in the time window, and an optimized time domain envelope is generated.

[0051] Specifically, the system acquires initial envelope timing data, and finds that the initial envelope has high-frequency fluctuation interference through timing change analysis of the data. The interference sources include two aspects. One is that the small control error of the deflation valve during the deflation of the cuff causes a small fluctuation in the cuff pressure, which is transmitted to the pressure sensor to form pressure fluctuation interference. The other is the electronic noise of the pressure sensor itself, which is generated by the current change of the internal circuit of the sensor and appears as high-frequency random signal fluctuation. These interferences cause irregular jitter in the shape of the initial envelope, and if directly used for subsequent feature calculation, it will cause calculation result deviation. Therefore, the system performs smoothing filtering processing on the initial envelope, and the filtering method uses mean value calculation in the time window.

[0052] For example, the system analyzes the timing change period of the initial envelope, determines the change period of the envelope by counting the duration of the main peak of the initial envelope, and sets the length of the fixed time window according to the change period. The length needs to meet two conditions. One is less than half the period of the main peak of the initial envelope to avoid excessive smoothing of the main peak shape and loss of key features due to too long window. The other is greater than the period of high-frequency fluctuation interference to ensure that each window contains multiple high-frequency fluctuation periods so that the fluctuation interference can be offset by mean value calculation. The system slides the time window with a fixed step according to the time sequence, divides the timing data of the initial envelope into multiple consecutive windows, and each window contains a fixed number of timing data points. For edge windows, if the number of data points is insufficient, the system uses edge data point repetition padding to make the number of data points in the edge window reach the set value.

[0053] The system performs mean value calculation on all data points in each time window, adds the values of all data points in the window to obtain a total sum, divides the total sum by the number of data points in the window to obtain the mean value of the window, and replaces the values of all original data points in the window with the mean value to complete the smoothing processing of one window. The system repeats the sliding window, mean value calculation, and data replacement operations until all timing data points of the initial envelope are smoothed to obtain the optimized time-domain envelope. The optimized time-domain envelope retains the main peak, secondary peak, and other main shape features related to arterial pulsation in the initial envelope, while eliminating high-frequency fluctuation interference, and the timing data change is more consistent with the real amplitude change of arterial pulsation.

[0054] In one embodiment, as shown in FIG. 1, Figure 2 The S3 of the motion scene blood pressure compensation measurement method provided by the application specifically includes the following steps: S31: Perform main peak detection processing on the time-domain envelope, locate the global maximum value point of the envelope and measure its half-width, and generate the main peak width variation.

[0055] Specifically, the system first performs baseline drift correction on the time domain envelope, constructs a baseline model using a linear fitting algorithm, and the formula is: ; Where B(t) is the baseline value at time t, k is the baseline slope (determined by the linear decline characteristic of the cuff deflation pressure), b is the baseline initial value at t=0, and t is the time variable. The system calculates the baseline value at each time point through the formula, subtracts the baseline value at the corresponding time point from the original envelope line data, completes the drift correction, and ensures that the envelope line only reflects the characteristics of the arterial pulse. After correction, the system performs main peak detection: traverses the envelope line time series data points, locates the global maximum value point , is the maximum value point time coordinate, is the corresponding amplitude. Calculate the half value , traverse before and after, locate the left endpoint t1 and right endpoint t2 where the amplitude is equal to , and calculate the half width: ; Where W is the current main peak half width, is the half left endpoint time, is the half right endpoint time. The system calls the stored static reference half width (the smart watch collects the resting signal and obtains it through the same process when the lactate accumulation marker is not activated), calculates the main peak width change: ; Where ΔW is the main peak width change, W is the current half width, is the static reference half width. The change quantity is related to the change of vascular tension, and the data is collected and calculated by the smart watch.

[0056] S32: Perform frequency domain energy analysis on the time domain envelope, calculate the energy distribution ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generate the frequency domain feature ratio.

[0057] Specifically, the system performs discrete Fourier transform on the time domain envelope to convert it into a frequency domain signal, and the formula is: ; Where X(k) is the frequency domain amplitude of the kth frequency point, is the nth time domain envelope data point, N is the total number of data points, n is the time domain index (0≤n<N), k is the frequency domain index (0≤k<N), and j is the imaginary unit. The system completes the conversion from time domain to frequency domain through the formula to obtain the amplitude of each frequency point. Subsequently, the system calculates the power spectral density: ; wherein, is the power spectral density of the kth frequency point, is the modulus of X(k), and N is the total number of data points, is the frequency value of the kth frequency point. The system performs energy integration on the 8-12 Hz and 1-4 Hz frequency bands respectively according to the preset frequency band: ; wherein, E is the total energy of the target frequency band, is the starting frequency of the frequency band, is the ending frequency of the frequency band, is the power spectral density function. The energy of the 8-12 Hz frequency band and the energy of the 1-4 Hz frequency band are calculated, and the frequency domain feature ratio is generated: ; wherein, R is the frequency domain feature ratio, is the total energy of the 8-12 Hz frequency band, is the total energy of the 1-4 Hz frequency band, and the ratio change reflects the energy deviation caused by abnormal vascular tension.

[0058] S33: Perform weighted fusion processing on the main peak width change amount and the frequency domain feature ratio, calculate the comprehensive deviation amount in combination with the static reference value, and generate the tension abnormality degree, which is used to indicate the degree of abnormal vascular tension caused by lactic acid accumulation.

[0059] Specifically, the system calls the stored static reference value, including the static reference half-width and the resting frequency domain feature ratio , which are obtained by the smart watch collecting resting signals and going through the same process when the lactic acid accumulation marker is not activated, as the reference. The system performs weighted fusion on ΔW and R, first determines the time domain feature weight factor and the frequency domain feature weight factor , which are obtained by fitting clinical data, and the system substitutes the preferred formula to calculate the tension abnormality degree: ; wherein, is the tension abnormality degree, representing the degree of abnormal vascular tension, is the main peak width change amount, is the static reference half-width, is the frequency domain feature ratio, is the time domain feature weight factor, is the frequency domain feature weight factor.

[0060] In one of the embodiments, the application provides a motion scene blood pressure compensation measurement method S4, which specifically comprises the following steps: S41: Nonlinear function mapping processing is performed on the tension abnormality, and the tension abnormality is converted into a blood vessel volume change ratio through hyperbolic tangent transformation to generate a blood vessel diastolic volume.

[0061] Specifically, the system converts the tension abnormality into a quantifiable blood vessel volume change ratio through nonlinear mapping. The hyperbolic tangent transformation is selected as the mapping function, which can compress the arbitrary value range of the tension abnormality to the interval [0, 1], avoid the overflow of the volume change ratio caused by extreme values, and at the same time, retain the positive correlation characteristics of the volume change. The mapping process can adopt the following formula: ; Wherein, V is the blood vessel diastolic volume (i.e. the blood vessel volume change ratio, dimensionless), is the maximum blood vessel volume change ratio (determined based on the actual measured value of the blood vessel volume in the resting state, reflecting the upper limit of the blood vessel diastolic volume, and the data comes from the resting state blood vessel related signal collected by the smart watch), k is the mapping adjustment coefficient (obtained by fitting clinical data, to ensure that can linearly reflect the blood vessel tension change when taking the intermediate value), is the tension abnormality generated by the step. The system calls the V and k parameters stored in the smart watch, substitutes them to obtain , and calculates ΔV through the above formula. When ΔV increases,

[0062] tends to 1, tends to 0, tends to 0, realizing accurate mapping of the tension abnormality to the blood vessel diastolic volume, and all operation data come from the pre-acquisition and processing results of the smart watch. S42: Exponential decay modeling processing is performed on the blood vessel diastolic volume, a time decay function is applied to simulate the dynamic recovery process of the blood vessel volume in the lactic acid metabolism process, and a volume change curve decreasing with time is constructed.

[0063] S42: Exponential decay modeling processing is performed on the blood vessel diastolic volume, a time decay function is applied to simulate the dynamic recovery process of the blood vessel volume in the lactic acid metabolism process, and a volume change curve decreasing with time is constructed.

[0064] ​​​​​​​​Specifically, the system simulates the dynamic recovery of blood vessel volume in the lactate metabolism process based on the generated vasodilation amount, the lactate metabolism gradually weakens over time, the blood vessel tension gradually recovers to normal, and the corresponding blood vessel volume change ratio needs to decrease over time. Preferably, the system can use an exponential decay function to model, and the modeling process uses the formula: ; wherein, is the dynamic volume change amount at time t, λ is the decay coefficient, which is determined based on the lactate metabolism rate, and is obtained by monitoring the lactate concentration and blood vessel volume change relationship at different time points, and the data correlation smartwatch records the motion stop time, t is the time variable after the motion stops. The system obtains the value of t in real time through the smartwatch, and calculates the corresponding by substituting the above formula, and arranges the volume change curve in time sequence. The curve shows an exponential decreasing trend as t increases: in the early stage of motion stop, is small, tends to 1, is close to ; as t increases, increases, tends to 0, and Vdecay(t) tends to 0, accurately simulating the dynamic recovery process of blood vessel volume caused by lactate metabolism.

[0065] S43: Baseline superposition processing is performed on the volume change curve, the volume change curve is added to the preset static blood vessel volume baseline value, a blood vessel volume compensation curve is generated, and the blood vessel volume compensation curve is used for subsequent frequency domain filtering processing to eliminate abnormal interference of blood vessel tension.

[0066] Specifically, the system needs to introduce a static blood vessel volume baseline value to ensure that the compensation curve can reflect the absolute change of blood vessel volume. The baseline value is the blood vessel volume reference value in the resting state, which is obtained by the smartwatch through the collection of resting state cuff pressure oscillation wave signals when the lactate accumulation sign is not activated, combined with the blood vessel volume estimation algorithm, and stored in the local storage unit of the smartwatch. Preferably, the compensation curve construction can use the following formula: wherein, is the blood vessel volume compensation curve (dimensionless) at time t, is the static blood vessel volume baseline value, is the generated dynamic volume change amount. The formula combines the dynamic volume change amount with the static baseline through baseline superposition, so that can reflect not only the blood vessel volume basis in the resting state, but also the dynamic changes in the lactate metabolism process after exercise.

[0067] The system calls the stored in the smartwatch, and combines the real-time calculated , substituting the above formula . The generated blood vessel volume compensation curve changes with time trends consistent with , the initial moment , gradually approaches as t increases. The curve is used for subsequent frequency domain filtering processing, providing quantitative compensation basis for eliminating the interference of abnormal blood vessel tension on blood pressure measurement.

[0068] In one embodiment, as shown in Figure 3 , the present application provides a motion scene blood pressure compensation measurement method S5 specifically includes the following steps: S51: Perform frequency domain convolution processing on the time domain envelope and the blood vessel volume compensation curve, multiply the envelope spectrum with the compensation filter transfer function in the frequency domain, and perform inverse Fourier transform to convert to the time domain, to generate a reconstructed compensation envelope.

[0069] Specifically, the system calls the optimized time domain envelope and the blood vessel volume compensation curve to ensure that the time axes of the two curves are synchronized. The system performs discrete Fourier transform on the optimized time domain envelope to convert the time domain signal to the frequency domain spectrum, obtaining the frequency component distribution of the envelope; at the same time, performs frequency domain analysis on the blood vessel volume compensation curve to extract its corresponding compensation filter transfer function, which is determined by the correlation between the blood vessel volume change and the blood pressure oscillation wave frequency domain response, and is used to adjust the envelope spectrum.

[0070] The system performs element-wise multiplication operation on the envelope spectrum and the compensation filter transfer function in the frequency domain, which adjusts the amplitude of each frequency component of the envelope to offset the frequency domain distortion caused by abnormal blood vessel tension. The system performs inverse discrete Fourier transform on the operated frequency domain signal to convert the signal from the frequency domain back to the time domain, obtaining a continuous curve that eliminates lactic acid accumulation interference and is close to the resting state, which is defined as the reconstructed compensation envelope.

[0071] S52: Perform differential feature extraction processing on the reconstructed compensation envelope to find the maximum value point by calculating the first derivative and detect the position point where the second derivative first changes from negative to positive to generate the systolic pressure feature moment and the diastolic pressure feature moment.

[0072] Specifically, the system performs differential feature extraction processing on the reconstructed compensation envelope, and calculates the first derivative of the envelope using a sliding window algorithm. The system traverses all data points of the first derivative, locates the data point with the maximum amplitude, which corresponds to the moment when the reconstructed compensation envelope changes in amplitude the fastest, and defines this moment as the systolic pressure feature moment. The system continues to calculate the second derivative of the first derivative, and also uses a sliding window algorithm to ensure calculation accuracy. The system traverses the data points of the second derivative, detects the first position point where the value changes from negative to positive, and defines the time point corresponding to this position point as the diastolic pressure feature moment. Preferably, the systolic pressure feature moment and the diastolic pressure feature moment are obtained by the following steps: S521: Perform first derivative calculation processing on the reconstructed compensation envelope, calculate the instantaneous slope value of each sampling point on the envelope by central difference method, and generate a first derivative sequence.

[0073] Specifically, the system first calls the reconstructed compensation envelope data in the cache unit, determines the sampling interval of the data, which is consistent with the sampling frequency of the smart watch pressure sensor, to ensure the accuracy of the time dimension of the subsequent difference calculation. Preferably, the system can perform first derivative calculation on the reconstructed compensation envelope using the central difference method. For each sampling point on the envelope, the system selects the two adjacent sampling points before and after the point, calculates the amplitude difference between the two sampling points, and divides the difference by twice the sampling interval to obtain the instantaneous slope value of the point. During the calculation, for the sampling points at the beginning and end of the envelope, the system uses one-sided difference method to supplement the calculation, selects one sampling point after the first point and one sampling point before the last point, respectively, to calculate the amplitude difference, and then divides the difference by the sampling interval to ensure that each sampling point on the entire envelope has an instantaneous slope value. All instantaneous slope values of the sampling points are arranged in time sequence to form a first derivative sequence.

[0074] S522: Perform extreme point detection processing on the first derivative sequence, scan the entire sequence to find the maximum positive point and record its corresponding time, and generate the systolic pressure feature moment.

[0075] Specifically, the system traverses each sampling point in the first derivative sequence in time sequence, starting from the beginning of the sequence, and extracts the instantaneous slope value of each sampling point in turn. The system compares the instantaneous slope value of the current sampling point with the instantaneous slope value of the adjacent sampling point, and records the maximum value and the corresponding time stamp in the current traversal process. The traversal process continues until the end of the first derivative sequence, and the system finally determines the maximum positive point in the entire sequence. The instantaneous slope value of this point is the maximum value of all sampling points in the sequence, and the value is positive, reflecting the fastest rising rate of the envelope at this moment. The system extracts the time stamp corresponding to the maximum positive point, which is the systolic pressure feature moment, representing the time node when the cuff pressure is balanced with the arterial systolic pressure.

[0076] S523: Perform second derivative calculation processing on the reconstructed compensation envelope, calculate the curvature change value of each sampling point on the envelope by the second central difference method, and search for the first point where the curvature changes from negative to positive after the systolic pressure feature time, to generate the diastolic pressure feature time.

[0077] Specifically, the system calls the generated reconstructed compensation envelope again, calculates the curvature change value of each sampling point by the second central difference method, and determines the search range as all sampling points after the systolic pressure feature time, and excludes the envelope rising stage before the systolic pressure feature time and the critical change of the amplitude drop stage, and traverses the curvature change value in the range. The system detects the first sampling point where the value changes from negative to positive, extracts the time stamp corresponding to the sampling point, and defines the time stamp as the diastolic pressure feature time, wherein the negative value corresponds to the accelerated decline of the envelope amplitude, the positive value corresponds to the decelerated decline of the amplitude, and the conversion point is the critical time when the decline rate slows down, which matches the physiological process of the cuff pressure being lower than the diastolic pressure.

[0078] S53: Perform pressure value mapping processing on the systolic pressure feature time and the diastolic pressure feature time, query the cuff pressure value at the corresponding time according to the feature time, and generate a compensation blood pressure value containing a compensation systolic pressure value and a compensation diastolic pressure value.

[0079] Specifically, the system synchronously acquires the cuff static pressure signal collected by the piezoresistive pressure sensor, which records the static pressure values at different times during the inflation and deflation process of the cuff. The system queries the static pressure value corresponding to the systolic pressure feature time in the cuff static pressure signal, and the pressure value is the compensation systolic pressure value after eliminating the motion interference. The system queries the static pressure value corresponding to the diastolic pressure feature time in the cuff static pressure signal, and the pressure value is the compensation diastolic pressure value after eliminating the motion interference. The system integrates the compensation systolic pressure value and the compensation diastolic pressure value into a data group to form a compensation blood pressure value containing two parameters, which can be directly used for output display and provides accurate data for blood pressure monitoring in a motion scene.

[0080] It should be understood that although the steps in the flowcharts involved in the embodiments described above are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the embodiments described above can include multiple steps or multiple stages, which are not necessarily executed at the same time but can be executed at different times, and the execution of the steps or stages is not necessarily sequential but can be executed alternately or in rotation with at least some of the other steps or the steps or stages in the other steps.

[0081] Based on the same inventive concept, the embodiments of the present application also provide a motion scene blood pressure compensation measurement device for implementing the motion scene blood pressure compensation measurement method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more motion scene blood pressure compensation measurement device embodiments provided below can refer to the limitations of the motion scene blood pressure compensation measurement method described above, which will not be repeated here.

[0082] Preferably, as shown in the embodiments of the present application, a motion scene blood pressure compensation measurement device 600 is provided, which is configured with the following modules: Figure 4 The motion scene blood pressure compensation measurement device 600 comprises: A lactic acid accumulation sign generation module 610 is configured to process the motion acceleration signal collected by the three-axis acceleration sensor and the heart rate signal collected by the heart rate sensor, calculate the acceleration-based exercise intensity index and the heart rate-based heart rate recovery slope, and generate a lactic acid accumulation sign based on a preset intensity threshold and a preset recovery threshold; A time domain envelope generation module 620 is configured to process the cuff pressure oscillation wave signal collected by the pressure sensor, extract the intrinsic mode function in the 0.5-5Hz frequency band and perform Hilbert transform to generate a time domain envelope; A tension abnormality calculation module 630 is configured to process the time domain envelope, calculate the change amount of the envelope main peak width relative to the static reference and the energy ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generate a tension abnormality degree representing the abnormality of the blood vessel tension; A blood vessel volume compensation curve construction module 640 is configured to process the tension abnormality degree, calculate the blood vessel distension amount obtained by a nonlinear mapping function, and construct a blood vessel volume compensation curve in combination with an exponential decay time function; The compensation blood pressure value output module 650 is configured to process the time domain envelope and the blood vessel volume compensation curve, reconstruct the compensated envelope through a frequency domain convolution operation, locate the maximum slope point and the second derivative positive point on the reconstructed envelope, and output the compensated blood pressure value for eliminating motion interference.

[0083] Preferably, the lactic acid accumulation sign generation module 610 provided in the application is configured with the following units: The motion intensity index calculation unit is configured to calculate the cubic root integral value of the three-axis synthesized acceleration in the time window by performing cubic root integration on the motion acceleration signal collected by the three-axis acceleration sensor, and generate the motion intensity index. The heart rate recovery slope calculation unit is configured to calculate the slope value of the heart rate decline rate after the motion stops by performing linear regression processing on the heart rate signal collected by the heart rate sensor, and generate the heart rate recovery slope. The lactic acid accumulation sign judgment unit is configured to perform double-threshold value logic judgment on the motion intensity index and the heart rate recovery slope, and generate the lactic acid accumulation sign in the active state when the motion intensity exceeds the preset intensity threshold value and the heart rate recovery rate is lower than the preset recovery threshold value.

[0084] Preferably, the time domain envelope generation module 620 provided in the application is configured with the following units: The intrinsic mode function extraction unit is configured to perform empirical mode decomposition processing on the cuff pressure oscillation wave signal collected by the pressure sensor, decompose the signal into a plurality of intrinsic mode function components, and extract the intrinsic mode function in the 0.5-5Hz frequency band. The initial envelope generation unit is configured to perform Hilbert transform processing on the intrinsic mode function in the 0.5-5Hz frequency band, calculate the instantaneous amplitude of the analytic signal, and generate the initial envelope. The envelope line smoothing optimization unit is configured to perform smoothing filtering processing on the initial envelope, eliminate high-frequency fluctuation interference through mean value calculation in the time window, and generate the optimized time domain envelope.

[0085] Preferably, the tension abnormality degree calculation module 630 provided in the application is configured with the following units: The main peak width detection unit is configured to perform main peak detection processing on the time domain envelope, locate the global maximum value point of the envelope and measure the half-width thereof, and generate the main peak width variation. The frequency domain energy ratio calculation unit is configured to perform frequency domain energy analysis processing on the time domain envelope, calculate the energy distribution ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generate the frequency domain feature ratio. The tension abnormality degree fusion calculation unit is configured to perform weighted fusion processing on the main peak width variation and the frequency domain feature ratio, calculate a comprehensive offset in combination with a static reference value, and generate a tension abnormality degree, which is used to indicate the degree of abnormality of blood vessel tension caused by lactic acid accumulation.

[0086] Preferably, the blood vessel volume compensation curve construction module 640 provided in the present application is configured with the following units: The blood vessel diastolic volume mapping unit is configured to perform non-linear function mapping processing on the tension abnormality degree, convert the tension abnormality degree into a blood vessel volume change ratio through hyperbolic tangent transformation, and generate a blood vessel diastolic volume; The volume change curve modeling unit is configured to perform exponential decay modeling processing on the blood vessel diastolic volume, simulate the dynamic recovery process of blood vessel volume in the lactic acid metabolism process by applying a time decay function, and construct a volume change curve that decreases over time; The blood vessel volume compensation curve generation unit is configured to perform baseline superposition processing on the volume change curve, add the volume change curve to a preset static blood vessel volume reference value, and generate a blood vessel volume compensation curve, which is used for subsequent frequency domain filtering processing to eliminate blood vessel tension abnormality interference.

[0087] Preferably, the compensation blood pressure value output module 650 provided in the present application is configured with the following units: The frequency domain convolution reconstruction unit is configured to perform frequency domain convolution processing on the time domain envelope line and the blood vessel volume compensation curve, multiply the envelope line spectrum by the compensation filter transfer function in the frequency domain and implement inverse Fourier transform conversion to the time domain, and generate a reconstructed compensation envelope line; The differential feature extraction unit is configured to perform differential feature extraction processing on the reconstructed compensation envelope line, find the maximum value point by calculating the first-order derivative, and detect the position point where the second-order derivative first changes from negative to positive, and generate a systolic pressure feature time point and a diastolic pressure feature time point; The compensation blood pressure value mapping unit is configured to perform pressure value mapping processing on the systolic pressure feature time point and the diastolic pressure feature time point, query the cuff pressure value at the corresponding time point according to the feature time point, and generate a compensation blood pressure value containing a compensation systolic pressure value and a compensation diastolic pressure value.

[0088] Preferably, the differential feature extraction unit comprises a first derivative sequence generation subunit, a systolic pressure feature time point detection subunit and a diastolic pressure feature time point detection subunit. The first derivative sequence generation subunit is configured to perform first derivative calculation on the reconstructed compensation envelope, calculate the instantaneous slope value of each sampling point on the envelope by using the central difference method, and generate a first derivative sequence. The systolic pressure feature time point detection subunit is configured to perform extreme value point detection on the first derivative sequence, scan the entire sequence to find the maximum positive point and record the corresponding time point, and generate a systolic pressure feature time point. The diastolic pressure feature time point detection subunit is configured to perform second derivative calculation on the reconstructed compensation envelope, calculate the curvature change value of each sampling point on the envelope by using the second central difference method, and search for the first point where the curvature changes from negative to positive after the systolic pressure feature time point, and generate a diastolic pressure feature time point.

[0089] In one embodiment, the application further provides an intelligent watch, comprising a memory and a processor, the memory stores a computer program, and the processor implements the blood pressure compensation measurement method for a motion scene when executing the computer program.

[0090] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0091] For the device embodiment, since it basically corresponds to the method embodiment, the relevant part is described in the part of the method embodiment. The device embodiment described above is only schematic, wherein the components described as separate components can or can not be physically separate, and the components displayed as a unit can or can not be a physical unit, i.e. can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present disclosure according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0092] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any skilled person in the art can easily think of various changes or replacements within the technical range disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of blood pressure compensated measurement of a motion scenario, characterized by, The method comprises the following steps: S1: processing the motion acceleration signal collected by the triaxial acceleration sensor and the heart rate signal collected by the heart rate sensor, calculating the motion intensity index based on acceleration and the heart rate recovery slope based on heart rate variability, and generating the lactic acid accumulation flag based on the preset intensity threshold and the preset recovery threshold; S2: processing the cuff pressure oscillation wave signal collected by the pressure sensor, extracting the intrinsic mode function in the 0.5-5Hz frequency band and performing Hilbert transform to generate the time domain envelope; S3: processing the time domain envelope, calculating the change amount of the envelope line main peak width relative to the static reference and the energy ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generating the tension abnormality degree representing the abnormality of blood vessel tension; S4: processing the tension abnormality degree, calculating the blood vessel diastolic volume compensation curve obtained by a nonlinear mapping function, and combining an exponential decay time function to construct the blood vessel volume compensation curve; S5: processing the time domain envelope and the blood vessel volume compensation curve, reconstructing the compensated envelope line by frequency domain convolution operation, and positioning the maximum slope point and the second derivative positive point on the reconstructed envelope line, and outputting the compensated blood pressure value for eliminating motion interference.

2. The method of claim 1, wherein, The S1 comprises: S11: cubic root integration of the motion acceleration signal collected by the triaxial acceleration sensor, calculation of the integral value of the cubic root of the three-axis combined acceleration in the time window, and generation of the motion intensity index; S12: linear regression processing of the heart rate signal collected by the heart rate sensor, calculation of the slope value of the heart rate decline rate after stopping motion, and generation of the heart rate recovery slope; S13: double-threshold logical judgment on the motion intensity index and the heart rate recovery slope, generation of the lactic acid accumulation flag in the activated state when the motion intensity exceeds the preset intensity threshold and the heart rate recovery rate is lower than the preset recovery threshold.

3. The method of claim 1, wherein, The S2 comprises: S21: empirical mode decomposition processing of the cuff pressure oscillation wave signal collected by the pressure sensor, decomposition of the signal into multiple intrinsic mode function components, and extraction of the intrinsic mode function in the 0.5-5Hz frequency band; S22: Hilbert transform processing of the intrinsic mode function, calculation of the instantaneous amplitude of the analytical signal, and generation of the initial envelope line; S23: smoothing filter processing of the initial envelope line, elimination of high-frequency fluctuation interference by mean value calculation in the time window, and generation of the optimized time domain envelope.

4. The method of claim 1, wherein, The S3 comprises: S31: main peak detection processing of the time domain envelope, positioning of the global maximum value point of the envelope line and measurement of its half-width, and generation of the main peak width change amount; S32: frequency domain energy analysis processing of the time domain envelope, calculation of the energy distribution ratio of the 8-12Hz frequency band to the 1-4Hz frequency band, and generation of the frequency domain feature ratio; S33: weighted fusion processing of the main peak width change amount and the frequency domain feature ratio, calculation of the comprehensive offset amount combined with the static reference value, generation of the tension abnormality degree, and the tension abnormality degree for indicating the degree of blood vessel tension abnormality caused by lactic acid accumulation.

5. The method of claim 4, wherein, The calculation formula of the tension abnormality degree is: ； wherein, is a tension abnormality degree, representing the degree of abnormality of the tension of the blood vessel, is a main peak width variation amount, is a static reference half-height width, is a frequency domain feature ratio, is a time domain feature weight factor, is a frequency domain feature weight factor.

6. The method of claim 1, wherein, The S4 comprises: S41: Perform nonlinear function mapping processing on the tension abnormality degree, convert the tension abnormality degree into a blood vessel volume change ratio through hyperbolic tangent transformation, and generate a blood vessel diastolic volume; S42: Perform exponential decay modeling processing on the blood vessel diastolic volume, simulate the dynamic recovery process of the blood vessel volume in the lactic acid metabolism process by applying a time decay function, and construct a volume change curve that decreases over time; S43: Perform baseline superposition processing on the volume change curve, add the volume change curve to a preset static blood vessel volume baseline value, and generate a blood vessel volume compensation curve, which is used for subsequent frequency domain filtering processing to eliminate blood vessel tension abnormality interference.

7. The method according to any one of claims 1 to 6, characterized in that, The S5 includes: S51: Perform frequency domain convolution processing on the time domain envelope and the blood vessel volume compensation curve, multiply the envelope spectrum with the compensation filter transfer function in the frequency domain, and implement inverse Fourier transform to convert to the time domain to generate a reconstructed compensation envelope; S52: Perform differential feature extraction processing on the reconstructed compensation envelope, find the maximum value point by calculating the first derivative, and detect the position point where the second derivative first changes from negative to positive to generate a systolic pressure feature time and a diastolic pressure feature time; S53: Perform pressure value mapping processing on the systolic pressure feature time and the diastolic pressure feature time, query the cuff pressure value at the corresponding time according to the feature time, and generate a compensation blood pressure value including a compensation systolic pressure value and a compensation diastolic pressure value.

8. The method of claim 7, wherein, The S52 includes: S521: Perform first derivative calculation processing on the reconstructed compensation envelope, calculate the instantaneous slope value of each sampling point on the envelope by the central difference method to generate a first derivative sequence; S522: Perform extreme value point detection processing on the first derivative sequence, scan the entire sequence to find the maximum positive value point and record its corresponding time to generate a systolic pressure feature time; S523: Perform second derivative calculation processing on the reconstructed compensation envelope, calculate the curvature change value of each sampling point on the envelope by the second central difference method, and search for the first point where the curvature changes from negative to positive after the systolic pressure feature time to generate a diastolic pressure feature time.

9. A motion scene blood pressure compensated measurement device, characterized by The device includes: A lactic acid accumulation sign generation module configured to process motion acceleration signals collected by a three-axis acceleration sensor and heart rate signals collected by a heart rate sensor, calculate a motion intensity index based on acceleration and a heart rate recovery slope based on a heart rate change rate, and generate a lactic acid accumulation sign based on a preset intensity threshold and a preset recovery threshold; A time domain envelope generation module configured to process cuff pressure oscillation wave signals collected by a pressure sensor, extract an intrinsic mode function in a 0.5-5Hz frequency band, and perform Hilbert transform to generate a time domain envelope; A tension abnormality degree calculation module configured to process the time domain envelope, calculate a change amount of an envelope main peak width relative to a static baseline and an energy ratio of an 8-12Hz frequency band to a 1-4Hz frequency band, and generate a tension abnormality degree representing blood vessel tension abnormality; A blood vessel volume compensation curve construction module configured to process the tension abnormality degree, calculate a blood vessel diastolic volume obtained by a nonlinear mapping function, and construct a blood vessel volume compensation curve in combination with an exponential decay time function; The compensation blood pressure value output module is configured to process the time domain envelope and the blood vessel volume compensation curve, reconstruct a compensated envelope through a frequency domain convolution operation, locate a maximum slope point and a second derivative positive variation point on the reconstructed envelope, and output a compensated blood pressure value for eliminating motion interference.

10. A smart watch comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor implements the method in any one of claims 1 to 8 when executing the computer program.

Images