Wearable blood pressure monitoring device and blood pressure monitoring method

Abstract

The invention relates to the technical field of wearable intelligent equipment, and provides a wearable blood pressure monitoring device and a blood pressure monitoring method.The device comprises a pressure sensor, an acceleration sensor and a central processing unit; the pressure sensor is used for collecting a pressure signal of a user in a motion state; the acceleration sensor is used for collecting an acceleration signal of a user in a motion state; the central processing unit is used for performing mode identification based on the motion frequency signal and the motion amplitude signal to obtain a current motion mode, and performing signal identification on the pressure signal based on the current motion mode to obtain a motion interference signal in the pressure signal; performing signal filtering processing on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal; determining blood pressure parameters of the user in the motion state based on the blood pressure oscillation wave signals. According to the embodiment of the invention, the accuracy and reliability of blood pressure measurement in the motion state are improved, and the requirement of a user on accurate monitoring of blood pressure in the motion process is met.

Description

Technical Field

[0001] The present invention relates to the technical field of wearable intelligent devices, and in particular to a wearable blood pressure monitoring device and a blood pressure monitoring method. Background Art

[0002] Currently, oscillometric blood pressure monitoring technology is widely used in various wearable devices. This method analyzes blood pressure readings by measuring the oscillation waves generated by pressure changes within a cuff or wristband. However, during exercise, the movement of human limbs generates a large number of irregular interference signals. These interference signals overlap with the actual blood pressure oscillation waves, making it difficult for existing oscillometric blood pressure monitoring to accurately distinguish between valid and interference signals. This results in large errors in blood pressure measurement results and fails to meet users' needs for accurate blood pressure monitoring during exercise. Therefore, how to effectively eliminate interference signals and obtain accurate blood pressure measurement results during exercise has become a technical problem that needs to be solved urgently. Summary of the Invention

[0003] The present invention provides a wearable blood pressure monitoring device and a blood pressure monitoring method, which aim to improve the accuracy and reliability of blood pressure measurement during exercise and meet the user's demand for precise blood pressure monitoring during exercise.

[0004] In a first aspect, the present invention provides a wearable blood pressure monitoring device, comprising a pressure sensor, an acceleration sensor, and a central processing unit; the central processing unit is connected to the pressure sensor and the acceleration sensor respectively;

[0005] The pressure sensor is used to: collect the pressure signal of the user in motion;

[0006] The acceleration sensor is used to collect acceleration signals of the user in motion; the acceleration signals include motion frequency signals and motion amplitude signals;

[0007] The CPU is used to:

[0008] performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain a current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal;

[0009] performing signal filtering processing on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal;

[0010] The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

[0011] In a second aspect, the present invention further provides a blood pressure monitoring method, which is implemented based on the wearable blood pressure monitoring device described in the first aspect, and the blood pressure monitoring method includes:

[0012] The pressure sensor collects the pressure signal of the user in motion, and the acceleration sensor collects the acceleration signal of the user in motion; the acceleration signal includes a motion frequency signal and a motion amplitude signal;

[0013] performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain a current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal;

[0014] performing signal filtering processing on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal;

[0015] The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

[0016] In a third aspect, the present invention further provides an electronic device comprising: a memory for storing a computer software program; and a processor for reading and executing the computer software program, thereby implementing any of the above-mentioned blood pressure monitoring methods.

[0017] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, wherein the storage medium stores a computer software program, and when the computer software program is executed by a processor, it implements any of the blood pressure monitoring methods described above.

[0018] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which implements any of the blood pressure monitoring methods described above when executed by a processor.

[0019] The wearable blood pressure monitoring device provided by an embodiment of the present invention collects acceleration signals and accurately identifies the user's current motion mode during exercise. It can specifically obtain motion interference signals under different motion modes, then filter the pressure signal based on the motion interference signals to accurately identify and remove the motion interference signals, thereby accurately filtering out the blood pressure oscillation wave signal from the pressure signal. The wearable blood pressure monitoring device further determines the user's blood pressure parameters during exercise based on the blood pressure oscillation wave signal, effectively avoiding the influence of motion interference signals on measurement results, improving the accuracy and reliability of blood pressure measurement during exercise, and meeting the user's need for accurate blood pressure monitoring during exercise. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 1 is a schematic structural diagram of a wearable blood pressure monitoring device provided by the present invention;

[0021] Figure 2 1 is a flow chart of the blood pressure monitoring method provided by the present invention;

[0022] Figure 3 An embodiment diagram of an electronic device provided by an embodiment of the present invention;

[0023] Figure 4 A diagram of an embodiment of a computer-readable storage medium provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0025] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0026] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0027] Optional, see Figure 1 As shown, Figure 1 The figure is a schematic diagram of the structure of the wearable blood pressure monitoring device provided by the present invention. The wearable blood pressure monitoring device includes a pressure sensor, an acceleration sensor, and a central processing unit. The central processing unit is connected to the pressure sensor and the acceleration sensor respectively.

[0028] Optionally, pressure sensors typically utilize piezoresistive or capacitive principles and are installed in contact with the human body, such as on the inner surface of a wristband or cuff. These sensors collect pressure signals generated by changes in blood pressure and body movement during exercise. As the heart contracts and relaxes, changes in pressure within the blood vessels are transmitted to the pressure sensor in contact with the skin. Simultaneously, the squeezing and friction of limbs during exercise also generate pressure changes, which are converted by the pressure sensor into electrical signals for output. This allows the pressure sensor to capture pressure signals from the user during exercise.

[0029] Optionally, the acceleration sensor generally adopts a three-axis accelerometer, which can measure acceleration in three mutually perpendicular directions (X, Y, and Z axes). By installing the acceleration sensor in a position close to the moving part of the human body, such as the wrist, ankle, etc., the acceleration signal of the user in motion can be collected in real time. The motion frequency signal in the acceleration signal reflects the fast and slow rhythm of the user's movement, such as the frequency of steps when running; the motion amplitude signal reflects the intensity of the movement, such as the amplitude of the arm swing. The acceleration sensor detects the change of inertial force and converts it into an electrical signal. Therefore, the acceleration sensor collects the acceleration signal of the user in motion. The acceleration signal includes a motion frequency signal and a motion amplitude signal.

[0030] Optionally, the central processing unit performs pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern, as specifically described in steps 201 to 204 .

[0031] Furthermore, the central processing unit performs signal recognition on the pressure signal according to the current motion mode to obtain a motion interference signal in the pressure signal under the current motion mode, as specifically shown in steps 205 to 217 .

[0032] Furthermore, the central processing unit performs signal filtering on the pressure signal based on the motion interference signal to obtain the blood pressure oscillation wave signal in the pressure signal. Common filtering methods include low-pass filtering, band-pass filtering, and adaptive filtering. Low-pass filtering can remove high-frequency motion interference signals from the pressure signal while retaining low-frequency blood pressure oscillation wave signals, because the frequency of blood pressure oscillation wave signals is generally around 0.5-15Hz; band-pass filtering can only allow signals within a specific frequency range to pass based on the frequency range of the blood pressure oscillation wave signal, further enhancing the extraction effect of the blood pressure oscillation wave signal; adaptive filtering can automatically adjust the filtering parameters according to the changes in the signal, and more effectively remove motion interference signals in complex motion environments.

[0033] In practical applications, multiple filtering methods are often combined for processing. First, a low-pass filter is used to preliminarily filter the pressure signal to remove most high-frequency motion interference signals. The preliminarily filtered signal is then further processed using a bandpass filter to retain the signal within the frequency range of the blood pressure oscillation wave signal. Finally, an adaptive filter is used to fine-tune the signal, further optimizing the filtering effect based on real-time signal changes, resulting in a relatively pure blood pressure oscillation wave signal.

[0034] Furthermore, the central processing unit determines the user's blood pressure parameters during exercise based on the blood pressure oscillation signal. The blood pressure oscillation signal contains a wealth of information related to blood pressure. Determining the user's blood pressure parameters during exercise based on the blood pressure oscillation signal primarily involves calculating systolic, diastolic, and mean arterial pressure. Common calculation methods are based on the oscillometric principle, which determines blood pressure values by analyzing the relationship between the amplitude of the blood pressure oscillation wave and cuff pressure. Specifically, this method extracts features from the blood pressure oscillation signal, such as identifying characteristic parameters such as the peak and valley values of the oscillation wave and the slope of the waveform. Generally speaking, as cuff pressure decreases, the amplitude of the blood pressure oscillation wave gradually increases. When the cuff pressure equals the mean arterial pressure, the amplitude of the oscillation wave reaches its maximum value. As the cuff pressure approaches the systolic pressure, the amplitude of the oscillation wave begins to increase significantly. As the cuff pressure approaches the diastolic pressure, the amplitude of the oscillation wave begins to decrease. By establishing a mathematical model between the characteristic parameters of the blood pressure oscillation wave and the blood pressure value, the characteristic parameters are used to calculate the systolic, diastolic, and mean arterial pressures. For example, an empirical formula or a regression model obtained through machine learning training can be used to take the extracted characteristic parameters as input and output the corresponding blood pressure parameter values.

[0035] In one embodiment, taking a wrist-type wearable blood pressure monitoring device as an example, after obtaining a relatively pure blood pressure oscillation wave signal, the blood pressure oscillation wave signal is feature extracted. Using a peak detection algorithm, each peak in the blood pressure oscillation wave signal is found, and the amplitude and time point corresponding to each peak are recorded. At the same time, the slope change between adjacent peaks is calculated. For example, a blood pressure calculation model is established based on a linear regression algorithm of machine learning through a large amount of clinical trial data. The peak amplitude, slope and other characteristic parameters of the extracted blood pressure oscillation wave signal are input into the model. After calculation, the model outputs the blood pressure parameters of the user in the exercise state. For example, the final systolic pressure is 130 mmHg, the diastolic pressure is 85 mmHg, and the mean arterial pressure is 100 mmHg, providing the user with blood pressure health reference data in the exercise state.

[0036] By collecting acceleration signals and accurately identifying the user's current motion mode during exercise, the embodiments of the present invention can specifically obtain motion interference signals under different motion modes. The pressure signal is then filtered based on the motion interference signals to accurately identify and remove the motion interference signals, thereby accurately filtering out the blood pressure oscillation wave signal from the pressure signal. Furthermore, the blood pressure parameters of the user during exercise are determined based on the blood pressure oscillation wave signal, effectively avoiding the influence of motion interference signals on the measurement results, improving the accuracy and reliability of blood pressure measurement during exercise, and meeting the user's need for accurate blood pressure monitoring during exercise.

[0037] In one embodiment, steps 201 to 204 are described as follows:

[0038] Step 201 : determining a frequency characteristic index based on the number of motion frequencies in the motion frequency signal and the ratio between the square of each motion frequency and the mean of the motion frequencies.

[0039] Optionally, in the motion frequency signal, different motion modes present unique frequency distribution characteristics. The embodiment of the present invention determines the frequency characteristic index by calculating the number of motion frequencies and the ratio between the square of each motion frequency and the mean of the motion frequency, and quantifies the characteristics of the motion frequency signal. Specifically, the number of different motion frequencies in the motion frequency signal is counted, and the square of each motion frequency is calculated. Then, the mean of all motion frequencies is obtained, and finally the square of each motion frequency is compared with the mean to obtain the frequency characteristic index, wherein the frequency characteristic index reflects the discrete degree and distribution characteristics of the motion frequency.

[0040] Continuing with the example of a wrist blood pressure monitor, let's say the accelerometer collects a motion frequency signal within 10 seconds and detects three different motion frequencies: f1 = 2 Hz, f2 = 2.1 Hz, and f3 = 1.9 Hz. First, calculate the mean of the motion frequencies. Then calculate the square of each motion frequency, respectively (f1) 2 =4, (f2) 2 =4.41, (f3) 2 =3.61. Finally, the frequency characteristic index is calculated: k1=4 / 2=2, k2=4.41 / 2=2.205, and k3=3.61 / 2=1.805. These three values constitute the frequency characteristic index of the motion frequency signal of this segment.

[0041] Step 202: Determine a frequency interval identifier of each frequency characteristic indicator based on the frequency interval in which each frequency characteristic indicator is located.

[0042] Furthermore, according to the frequency interval in which each frequency characteristic indicator is located, the frequency interval identifier of each frequency characteristic indicator is determined, and different frequency intervals are pre-divided, such as low-frequency interval, medium-frequency interval, high-frequency interval, etc., and then each frequency characteristic indicator is judged to fall into which interval, and a corresponding identifier is assigned, such as "low", "medium" and "high". Among them, the frequency interval identifier can intuitively reflect the relative size and distribution characteristics of the motion frequency.

[0043] Continuing with the above embodiment, the frequency interval identification rule is pre-set as follows: a frequency characteristic index less than 1.5 is in the "low" interval, 1.5-2.5 is in the "medium" interval, and greater than 2.5 is in the "high" interval. For the calculated frequency characteristic index, k1=2, k2=2.205, k3=1.805, k1, k2, and k3 all fall into the "medium" interval, and therefore, the frequency interval identification is "medium".

[0044] Step 203 : determining an amplitude variation index based on the sum of the absolute values of the amplitude differences between adjacent time points in the motion amplitude signal and the maximum amplitude value.

[0045] Furthermore, the motion amplitude signal reflects the intensity and variability of motion. Embodiments of the present invention determine the amplitude variation index by calculating the ratio of the sum of the absolute values of the amplitude differences between adjacent time points in the motion amplitude signal to the maximum amplitude value. Specifically, the calculation involves first finding the difference between the motion amplitude signals at adjacent time points and taking their absolute values. These absolute values are then summed, and the maximum value in the motion amplitude signal is found. Finally, the ratio of the two values is calculated. The amplitude variation index measures the degree of variability in motion amplitude.

[0046] Continuing with the above embodiment, for example, within a certain period of time, the amplitude values of the motion amplitude signal collected by the acceleration sensor at adjacent time points are A1=5m / s 2 , A2=7m / s 2 , A3=6m / s 2 , A4=8m / s 2 First, calculate the absolute value of the amplitude difference between adjacent time points: |A2-A1|=|7-5|=2, |A3-A2|=|6-7|=1, |A4-A3|=|8-6|=2, and their sum is S=2+1+2=5. The maximum value in the motion amplitude signal is A4=8m / s 2 , therefore, the amplitude variation index I = S / A max =5 / 8=0.625.

[0047] Step 204 : Perform pattern recognition based on the frequency interval identifier and the amplitude change index to obtain the current motion pattern.

[0048] Furthermore, pattern recognition is performed based on the frequency interval identifier and the amplitude change index to obtain the current motion pattern, as specifically described in steps 2041 to 2043 .

[0049] The embodiment of the present invention extracts distinctive characteristic indicators from the motion frequency signal and the motion amplitude signal. Therefore, the characteristic indicators can accurately identify the current motion mode of the user in the exercise state, and can specifically filter the motion interference signal of the pressure signal under different motion modes, thereby accurately filtering out the blood pressure oscillation wave signal in the pressure signal, effectively avoiding the influence of the motion interference signal on the measurement result, improving the accuracy and reliability of blood pressure measurement in the exercise state, and meeting the user's demand for accurate blood pressure monitoring during exercise.

[0050] In one embodiment, steps 2041 to 2043 are described as follows:

[0051] Step 2041: If the frequency interval identifier is the first interval identifier and the amplitude change index is less than the first preset amplitude change threshold, it is determined that the current sports mode is the riding mode.

[0052] Optionally, the first interval identifier, the second interval identifier, and the third interval identifier in the embodiment of the present invention are different identifiers. When the frequency interval identifier is the first interval identifier (indicating that the exercise frequency is in a specific lower range) and the amplitude change index is less than the first preset amplitude change threshold (indicating that the exercise amplitude change is relatively small), the current exercise mode is determined to be the cycling mode. This is because during cycling, the body mainly rotates regularly around the moving parts of the bicycle, the exercise frequency is relatively stable and low, and the body amplitude changes relatively smoothly, which meets the characteristics set by the judgment condition.

[0053] In one embodiment, the first interval identifier corresponds to a "low" frequency interval with a frequency characteristic index less than 1.2, and the first preset amplitude change threshold is set to 0.3. The frequency characteristic indices k1 = 1.1, k2 = 1.05, and k3 = 1.15, all of which are "low" frequency interval identifiers, i.e., the first interval identifier. The amplitude change index I = 0.25, which is less than the first preset amplitude change threshold of 0.3, therefore, the current sports mode is determined to be cycling mode.

[0054] Step 2042: If the frequency interval identifier is the second interval identifier, and the amplitude change index is greater than or equal to the first preset amplitude change threshold and less than the second preset amplitude change threshold, determine that the current exercise mode is the running mode.

[0055] Furthermore, when the frequency interval is identified as the second interval (corresponding to a moderate exercise frequency range), and the amplitude change index is between the first and second preset amplitude change thresholds, the current exercise mode is determined to be running. During running, alternating foot contact produces periodic motion with a moderate frequency. Simultaneously, the body's ups and downs and arm swings cause some variation in the amplitude of the motion, but not as dramatic as in some strenuous exercise.

[0056] In one embodiment, the second interval identifier corresponds to a frequency characteristic index in the "medium" frequency range of 1.2-2.2, the first preset amplitude change threshold is 0.3, and the second preset amplitude change threshold is 0.6. The frequency characteristic indices k1 = 1.8, k2 = 2.0, and k3 = 1.9, and the frequency interval identifier is "medium," i.e., the second interval identifier. The amplitude change index I = 0.45 satisfies the condition of being greater than or equal to the first preset amplitude change threshold of 0.3 and less than the second preset amplitude change threshold of 0.6, thus determining that the current exercise mode is running mode.

[0057] In step 2043, if the frequency interval identifier is the third interval identifier and the amplitude change index is greater than or equal to the second preset amplitude change threshold, it is determined that the current exercise mode is the swimming mode.

[0058] Furthermore, when the frequency interval identifier is the third interval identifier (representing a higher frequency range), and the amplitude change index is greater than or equal to the second preset amplitude change threshold (indicating a significant amplitude change), the current exercise mode is determined to be swimming. Because swimming involves high-frequency movements such as paddling and body swinging, and the amplitude of movement is affected by factors such as water resistance in water, it can change dramatically.

[0059] In one embodiment, the third interval identifier corresponds to a "high" frequency interval with a frequency characteristic index greater than 2.2, and the second preset amplitude change threshold is 0.6. The frequency characteristic indices k1 = 2.5, k2 = 2.6, and k3 = 2.4, and the frequency interval identifier is "high," i.e., the third interval identifier. The amplitude change index I = 0.7, which is greater than or equal to the second preset amplitude change threshold of 0.6, therefore, the current exercise mode is determined to be swimming.

[0060] The embodiment of the present invention constructs a targeted motion pattern recognition rule system based on the frequency interval identification and amplitude change index obtained from motion signal processing, so that the differences in frequency and amplitude change characteristics of different motion patterns can be utilized, and common motion patterns such as cycling, running, and swimming can be quickly and accurately identified in complex motion scenes. Therefore, the pressure signals under different motion modes can be targeted to filter the motion interference signals, so as to accurately filter out the blood pressure oscillation wave signal in the pressure signal, effectively avoiding the influence of the motion interference signal on the measurement results, improving the accuracy and reliability of blood pressure measurement in the exercise state, and meeting the user's demand for accurate blood pressure monitoring during exercise.

[0061] In one embodiment, steps 2044 to 2047 are described as follows:

[0062] Step 2044 : comparing the running direction angle difference between adjacent time points in the running direction signal with a preset direction change rate threshold.

[0063] Optionally, the acceleration signal collected by the embodiment of the present invention also includes a running direction signal. Therefore, the running direction signal is processed to calculate the running direction angle difference between adjacent time points. Specifically, the running direction angle difference between two adjacent time points t is obtained. i and t i+1 The running direction angle θ i and θ i+1 , through the formula Δθ=|θ i+1 -θ i Calculate the angle difference. Then, compare the calculated angle difference with a preset directional change rate threshold. The preset directional change rate threshold is set based on the reasonable range of directional changes during normal motion and is used to determine whether the directional change is abnormal.

[0064] In one embodiment, the preset direction change rate threshold is 30°. When a user rides a bicycle, at two adjacent time points t1 and t2, the collected running direction angles are θ1 = 45° and θ2 = 60°, respectively. The running direction angle difference Δθ = |60° - 45°| = 15°. Comparing this difference with the preset direction change rate threshold of 30°, 15° < 30° is found.

[0065] Step 2045: If the running direction angle difference is greater than the preset direction change rate threshold, it is determined that there is a sudden change in direction at the current time point, and the number of sudden changes is recorded to obtain the number of sudden changes in direction.

[0066] Furthermore, if the angle difference is greater than a preset direction change rate threshold, it indicates that the user's movement direction has suddenly changed at the current time point. This sudden change in direction is recorded, and the number of sudden changes is incremented by 1 each time a direction change that meets the conditions occurs. By continuously recording the number of sudden changes in direction, we can obtain an overall picture of the user's movement direction changes over a period of time.

[0067] Continuing with the above example, if the running direction angle collected at time t3 during the subsequent acquisition process is θ3 = 120°, then the running direction angle difference between t2 and t3 is Δθ = |120° - 60°| = 60°. Since 60° > 30°, it is determined that a sudden change in direction occurred at time t3, and the number of sudden changes in direction is recorded as 1. As the movement continues, if multiple sudden changes in direction occur, the number of sudden changes is counted accordingly.

[0068] Step 2046: If the current motion mode is the cycling mode and the number of sudden changes in direction is greater than the first upper limit of the number of sudden changes in direction in the cycling mode, the current motion mode is corrected from the cycling mode to the running mode.

[0069] Furthermore, if the current motion mode is the cycling mode, the number of directional mutations is further determined, wherein the upper limit of the first directional mutation number in the cycling mode is set according to the frequency and amplitude of the direction change during normal cycling. If the number of directional mutations is greater than the upper limit of the first directional mutation number, it means that the direction change in the current motion process does not conform to the normal characteristics of the cycling mode, and is more in line with the direction change characteristics of the running mode (the direction is more likely to change frequently when running). At this time, the current motion mode is corrected from the cycling mode to the running mode. In one embodiment, the upper limit of the first directional mutation number in the cycling mode is 5 times. During the cycling process, after a period of monitoring, the recorded number of directional mutations reached 7 times. Since 7>5, and the current motion mode has been determined to be the cycling mode, the current motion mode is corrected to the running mode.

[0070] Step 2047: If the current motion mode is the swimming mode and the number of sudden direction changes is greater than the second upper limit of the number of sudden direction changes in the swimming mode, the current motion mode is corrected from the swimming mode to the running mode.

[0071] Furthermore, when the current motion mode is swimming mode, the number of directional mutations is compared with the upper limit of the second number of directional mutations in swimming mode, wherein the upper limit of the second number of directional mutations in swimming mode is set based on the normal direction change pattern of swimming motion. If the number of directional mutations is greater than the upper limit of the second number of directional mutations, it means that the current direction change situation exceeds the normal range of the swimming mode, and the direction changes in the running mode are relatively more frequent during the motion process. Therefore, the current motion mode is corrected from the swimming mode to the running mode. In one embodiment, the upper limit of the second number of directional mutations in the swimming mode is 3 times. During the user's swimming process, the recorded number of directional mutations reaches 4 times. Since 4>3 and the current motion mode is swimming mode, the current motion mode is corrected to the running mode.

[0072] The embodiment of the present invention, based on the preliminary identification of motion patterns based on frequency interval identification and amplitude change index, effectively captures the directional mutation during the motion process by accurately calculating the running direction angle difference and comparing it with the threshold, and corrects the preliminary identified motion pattern in combination with the reasonable range setting of the number of directional mutations under different motion modes. Therefore, it fully takes into account the change in motion direction, can effectively correct the misjudgment of motion patterns caused by special motion states or interference, and improves the accuracy of motion pattern recognition in complex motion scenes. Therefore, it can filter the motion interference signal of the pressure signal under different motion modes in a targeted manner, thereby accurately filtering out the blood pressure oscillation wave signal in the pressure signal, effectively avoiding the influence of the motion interference signal on the measurement results, improving the accuracy and reliability of blood pressure measurement in the motion state, and meeting the user's demand for accurate blood pressure monitoring during exercise.

[0073] In one embodiment, steps 205 to 209 are described as follows:

[0074] In step 205, if the current motion mode is the running mode, the pressure signal is divided based on a time window of a preset size, and a first target time window and a second target time window are determined based on a preset mean threshold and the pressure signal value of each sampling point in the sub-signal within each time window.

[0075] Optionally, after determining that the current motion mode is the running mode, in order to identify the motion interference signal from the pressure signal, the pressure signal is first divided based on a time window of a preset size, wherein the size of the time window is set according to the periodic characteristics of the pressure signal changes during running motion and the sampling frequency, and the continuous pressure signal is divided into multiple sub-signal segments of equal length.

[0076] Furthermore, a preset mean threshold is set, and by comparing the pressure signal value at each sampling point in the sub-signal within each time window with the threshold, time windows that meet specific conditions are screened out. Specifically, time windows in which the mean of the pressure signal value is higher than the preset mean threshold are determined as first target time windows, and time windows in which the mean of the pressure signal value is lower than the preset mean threshold are determined as second target time windows.

[0077] In one embodiment, the pressure signal sampling frequency is 1000 Hz, the preset time window size is 500 sampling points (i.e., 0.5 seconds), and the preset mean threshold is 5 mV. When the user is in running mode, the pressure signal is a series of continuous voltage values. The pressure signal is divided into a time window of 500 sampling points to obtain multiple sub-signal segments. For a sub-signal in one time window, the pressure signal values of its 500 sampling points are {p1, p2, ..., p 500}, calculate the mean value of the pressure signal value in the time window For example, if it is 6mV, since 6mV>5mV, the time window is determined as the first target time window; if in another time window, the calculated average value 4mV<5mV, it is determined as the second target time window.

[0078] Step 206: For each first target time window, determine the local maximum and local minimum, and determine the time interval between adjacent local maximum values.

[0079] Furthermore, for each sub-signal within the first target time window, the signal's fluctuation characteristics are analyzed by searching for local maxima and local minima. A local maximum refers to a point within a sampling point and its neighborhood where the pressure signal value is greater than that of surrounding points; a local minimum refers to a point where the pressure signal value is less than that of surrounding points. After finding the local maxima and local minima, the time intervals between adjacent local maxima are calculated. This time interval reflects the periodic characteristics of the pressure signal's fluctuations.

[0080] Continuing with the above embodiment, for example, if there is a local maximum point p in the sub-signal max1 ,p max2 ,... and local minimum point p min1 ,p min2 ,.... Among them, p max1 The corresponding sampling point numbers are n1, p max2 The corresponding sampling point number is n2. Since the sampling frequency is 1000Hz, the adjacent local maximum value p max1 and p max2 The time interval T 12=(n2-n1) / 1000 (unit: seconds) By traversing the sub-signals in the entire first target time window, the time intervals between all adjacent local maxima are determined.

[0081] Step 207 : For each local maximum, determine a time correlation parameter of each local maximum based on the difference between the rise time from the previous local minimum to each local maximum and the fall time from each local maximum to the subsequent local minimum.

[0082] Furthermore, for each local maximum, the rise time from the previous local minimum to the local maximum and the fall time from the local maximum to the subsequent local minimum are calculated. The difference between these two times is then calculated and used as the time correlation parameter for the local maximum. The time correlation parameter can reflect the asymmetry of the pressure signal during local fluctuations, and different motion interference conditions will cause the time correlation parameter to exhibit different characteristics.

[0083] In one embodiment, a local maximum p max , the local minimum before it is p min1 , the corresponding sampling point numbers are n max and n min1 ; The subsequent local minimum is p min2 , the corresponding sampling point number is n min2 Then from p min1 to p max Rise time t up =(n max -n min1 ) / 1000 (unit: second), from p max to p min2 The fall time t down =(n min2 -n max ) / 1000 (unit: seconds). The time correlation parameter of the local maximum value Δt=t up -t down By calculating the time correlation parameters of each local maximum, a set of data reflecting the temporal characteristics of the local fluctuation of the pressure signal is obtained.

[0084] Step 208: Determine the first interference signal strength based on the average of the time intervals between adjacent local maximum values, the average of the peak-to-peak values of adjacent local maximum values, and the average of the time correlation parameter of each local maximum value.

[0085] Furthermore, the mean of the time intervals between adjacent local maxima is calculated Reflects the average period of pressure signal fluctuation; then calculates the mean of the peak-to-peak values of adjacent local maxima (i.e., the difference between adjacent local maxima and local minima) It reflects the average amplitude of the pressure signal fluctuation; then calculate the mean of the time correlation parameter of each local maximum Furthermore, the above three parameters are combined to obtain the first interference signal strength S1. The embodiment of the present invention adopts the formula: Wherein, α, β, and γ are coefficients adjusted according to actual conditions, and α+β+γ=1, for example, α=0.3, β=0.5, and γ=0.2.

[0086] Step 209 : Determine the sub-signals in the first target time window and the sub-signals in the second target time window whose interference signal strength is greater than or equal to a first preset strength threshold as motion interference signals.

[0087] Furthermore, a first preset strength threshold is set, and the first interference signal strength calculated for each first target time window is compared with the threshold. If the first interference signal strength is greater than or equal to the first preset strength threshold, it is considered that the sub-signal within the first target time window contains a strong motion interference signal; at the same time, the second target time window is also taken into consideration due to the particularity of its pressure signal mean. Ultimately, the sub-signals within the first target time window and the sub-signals within the second target time window that meet the conditions are determined to be motion interference signals, and the motion interference component is accurately extracted from the pressure signal.

[0088] In one embodiment, the first preset intensity threshold is 1.2. For the first target time window, where the first interference signal intensity S1 is calculated to be 1.57, since 1.57 > 1.2, the sub-signal within the first target time window is determined to be a motion interference signal. Simultaneously, all sub-signals within the second target time window are also determined to be motion interference signals. These determined sub-signals collectively constitute the motion interference signal in the pressure signal in the running mode.

[0089] The embodiment of the present invention comprehensively considers the mean, fluctuation period, amplitude and time characteristics of the pressure signal in the running mode, and can more accurately and comprehensively identify the motion interference signal in the pressure signal in the running mode, effectively remove the influence of motion interference on the blood pressure signal, and obtain the blood pressure oscillation wave signal in the pressure signal in the running mode, effectively avoiding the influence of motion interference signals on the measurement results, improving the accuracy and reliability of blood pressure measurement in the running mode, and meeting the user's needs for accurate blood pressure monitoring during exercise.

[0090] In one embodiment, steps 210 to 213 are described as follows:

[0091] In step 210, if the current motion mode is the cycling mode, a fast Fourier transform is performed on the pressure signal to obtain a frequency domain representation of the pressure signal, and a first target frequency interval and a second target frequency interval are determined based on a preset energy proportion and an energy distribution of the frequency interval in the frequency domain representation.

[0092] Optionally, when it is determined that the current motion mode is the cycling mode, in order to identify the motion interference signal in the pressure signal, the pressure signal will first be subjected to a fast Fourier transform (FFT), wherein the FFT can convert the pressure signal from the time domain to the frequency domain, so that the signal is displayed in the form of frequency components.

[0093] Furthermore, based on the energy distribution of each frequency interval in the frequency domain representation and the preset energy ratio, the first target frequency interval and the second target frequency interval are determined. The preset energy ratio is a pre-set standard used to measure which frequency intervals contain more energy. Usually, motion interference signals are concentrated in certain frequency intervals with a higher energy ratio. Therefore, it can be understood that the energy ratios of the first target frequency interval and the second target frequency interval are both greater than or equal to the preset energy ratio, and the frequency of the second target frequency interval is higher than the frequency of the first target frequency interval.

[0094] In one embodiment, the collected pressure signal has a duration of 10 seconds and a sampling frequency of 1000 Hz. The pressure signal of these 10,000 sampling points is subjected to FFT transformation to obtain a frequency domain representation. The frequency domain range is divided into multiple frequency intervals, such as 0-10Hz, 10-20Hz, 20-30Hz, etc. The energy proportion in each frequency interval is calculated, and the preset energy proportion is set to 10%. After calculation, it is found that the energy proportion of the 20-30Hz frequency interval is 15%, and the energy proportion of the 50-60Hz frequency interval is 12%, then 20-30Hz is determined as the first target frequency interval, and 50-60Hz is determined as the second target frequency interval.

[0095] Step 211: determine a frequency screening range based on the riding speed collected by the speed sensor, and screen the signal frequencies in the first target frequency interval based on the frequency screening range to obtain a first target signal frequency and a second target signal frequency in the first target frequency interval.

[0096] Furthermore, a speed sensor is used to collect the riding speed. There is a certain correlation between the riding speed and the frequency of the motion interference signal. Therefore, a frequency screening range is determined according to the collected riding speed. The frequency screening range can screen the signal frequencies in the first target frequency interval and exclude frequency components that are irrelevant to the current riding speed, so that the screened frequencies are more likely to be the frequencies of the motion interference signal. After screening, the first target signal frequency and the second target signal frequency in the first target frequency interval are obtained.

[0097] In one embodiment, the riding speed detected by the speed sensor is 20 km / h. Experimental data or an empirical formula are used to determine the frequency screening range at this speed to be 22-28 Hz. For the previously determined first target frequency range of 20-30 Hz, signal frequencies within this range are screened as the first target signal frequency. Signal frequencies outside this range but still within 20-30 Hz are screened as the second target signal frequency. For example, 23 Hz, 25 Hz, and 27 Hz are determined as the first target signal frequency, and 21 Hz and 29 Hz are determined as the second target signal frequency.

[0098] Step 212: For each first target signal frequency in the first target frequency interval, determine the second interference signal strength based on the average of the difference between each signal frequency peak and the center signal frequency, and the average of the difference between each signal frequency peak and the two signal frequency valleys before and after it.

[0099] Furthermore, for the first target signal frequency in each first target frequency interval, the mean of the difference between each signal frequency peak and the center signal frequency is calculated, where the center signal frequency is the center value of the frequency interval, and the mean of the difference between each signal frequency peak and the two signal frequency valleys before and after it.

[0100] Furthermore, the above two parameters are weighted and summed to obtain the second interference signal strength, wherein the second interference signal strength reflects to what extent the frequency component is a motion interference signal.

[0101] In one embodiment, taking the first target signal frequency of 23 Hz as an example, for example, the center signal frequency of the frequency interval is 25 Hz. The peak values of the signals of this frequency collected within a period of time are P1, P2, and P3, and the corresponding two signal frequency valleys before and after are V 11 ,V 12 ,V 21 ,V 22 ,V 31 ,V 32 The difference between the peak signal frequency and the center signal frequency is calculated as |P1-25|, |P2-25|, |P3-25|, and the average is The difference between the peak value of the signal frequency and the two valley values of the signal frequency before and after is calculated as P1-V 11 ,P1-V 12 ,P2-V 21 ,P2-V 22 ,P3-V 31 ,P3-V 32 ,, its mean is The calculation formula for the second interference signal strength is: Wherein w1 and w2 are weight coefficients, and w1+w2=1, for example, w1=0.4, w2=0.6. Substituting the calculated mean into the formula, the second interference signal strength of the first target signal frequency can be obtained.

[0102] Step 213 : Determine as motion interference signals the signals in the first target frequency interval, the signals corresponding to the second target frequency interval, and the signals corresponding to the second target signal frequency, whose second interference signal strength is greater than or equal to the second preset strength threshold.

[0103] Furthermore, a second preset intensity threshold is set, and the second interference signal intensity calculated from the first target signal frequency in each first target frequency interval is compared with the threshold. If the second interference signal intensity is greater than or equal to the second preset intensity threshold, the corresponding signal in the first target frequency interval is considered to be a motion interference signal. At the same time, the signal corresponding to the second target frequency interval and the signal corresponding to the second target signal frequency are also identified as motion interference signals due to their own characteristics. Ultimately, these signals that meet the conditions are determined to be motion interference signals in the pressure signal under riding mode. In one embodiment, the second preset intensity threshold is 0.8. For the first target signal frequency of 23Hz, whose second interference signal intensity is S2=1.2 calculated above, since 1.2>0.8, the signal corresponding to this frequency is determined to be a motion interference signal. At the same time, the signal corresponding to the second target frequency interval of 50-60Hz and the signals corresponding to the second target signal frequencies of 21Hz and 29Hz in the first target frequency interval are also determined to be motion interference signals.

[0104] The embodiment of the present invention comprehensively considers the relationship between the signal frequency peak and the center frequency, as well as the signal frequency valley, and can more comprehensively and accurately identify the motion interference signal in the pressure signal under the riding mode, effectively avoiding the influence of the motion interference signal on the measurement results, and obtaining the blood pressure oscillation wave signal in the pressure signal under the riding mode, effectively avoiding the influence of the motion interference signal on the measurement results, improving the accuracy and reliability of blood pressure measurement under the riding mode, and meeting the user's needs for accurate blood pressure monitoring during exercise.

[0105] In one embodiment, steps 214 to 217 are described as follows:

[0106] Step 214 : If the current motion mode is the swimming mode, the motion cycle is determined based on the posture information collected by the gyroscope, and the pressure signal is split based on the motion cycle to obtain multiple signal subsequences.

[0107] Optionally, when it is determined that the current motion mode is swimming mode, the built-in gyroscope is used to collect the user's posture information during swimming. The human body's paddling, turning and other movements during swimming will cause the posture to change periodically. Therefore, the motion cycle can be determined by analyzing the posture data collected by the gyroscope (such as angular velocity, angle change, etc.).

[0108] Furthermore, the pressure signal is split using the motion cycle as a time interval, and the continuous pressure signal is divided into multiple signal subsequences with the same time length, each subsequence corresponding to the pressure change within one motion cycle.

[0109] In one embodiment, a built-in gyroscope collects posture information at a frequency of 100 Hz. Analysis of the angular velocity data collected by the gyroscope during swimming reveals that the angular velocity exhibits a complete cyclical variation with each completed stroke, with a calculated motion cycle of T = 0.8 seconds. The pressure signal is sampled at a frequency of 500 Hz and segmented at intervals of T = 0.8 seconds. Since the number of pressure signal sampling points in 0.8 seconds is 0.8 * 500 = 400, the pressure signal is segmented into multiple signal subsequences Q1, Q2, ..., each containing 400 sampling points.

[0110] Step 215 : Filter the signal subsequences based on the preset period threshold and the preset amplitude threshold combined with the fluctuation period and amplitude change of the signal in each signal subsequence to obtain a first target signal subsequence and a second target signal subsequence containing water wave interference.

[0111] Furthermore, preset period thresholds and preset amplitude thresholds are set to screen signal subsequences. The preset period threshold is used to determine whether the signal fluctuation period in a signal subsequence conforms to the normal period range for swimming, and the preset amplitude threshold is used to determine whether the signal amplitude variation in a signal subsequence is within a reasonable range. Therefore, signal subsequences whose signal fluctuation period conforms to the preset period threshold and whose amplitude variation conforms to the preset amplitude threshold are identified as the first target signal subsequence; such signal subsequences are likely to contain normal exercise-related pressure variations. Signal subsequences whose fluctuation period and / or amplitude variation are abnormal, potentially affected by factors such as water wave interference, are identified as the second target signal subsequence.

[0112] In one embodiment, the preset period threshold is 0.7-0.9 seconds, and the preset amplitude threshold is 3-8 mV. For signal subsequence Q1, the calculated signal fluctuation period is 0.82 seconds, and the signal amplitude in the subsequence is between 3.5-7 mV, meeting both the preset period threshold and the preset amplitude threshold. Therefore, Q1 is identified as the first target signal subsequence. However, signal subsequence Q2, while having a signal fluctuation period of 0.8 seconds, has an amplitude between 2-10 mV, exceeding the preset amplitude threshold. Therefore, Q2 is identified as the second target signal subsequence containing water wave interference.

[0113] Step 216: For each first target signal subsequence, determine the third interference signal strength based on the mean of the difference between the signal peak value and the amplitude value of each signal, the mean of the difference between the signal valley value and the amplitude value of each signal, and the mean of the amplitude standard deviation of each signal.

[0114] Furthermore, for each first target signal subsequence, the mean of the difference between the signal peak value and the amplitude value of each signal is calculated to reflect the degree to which the signal peak value deviates from the average amplitude, and the mean of the difference between the signal valley value and the amplitude value of each signal is calculated to reflect the deviation of the signal valley value from the average amplitude, and the mean of the amplitude standard deviation of each signal is calculated to measure the degree of dispersion of the signal amplitude.

[0115] Furthermore, the three means are weighted and summed to obtain a third interference signal strength, wherein the third interference signal strength quantifies the strength of the motion interference signal in each first target signal subsequence. The specific formula can be referred to the calculation formula in the running mode and will not be repeated here.

[0116] Step 217: Determine the signals corresponding to the first target signal subsequence and the signals corresponding to the second target signal subsequence, whose third interference signal strength is greater than or equal to a third preset strength threshold, as motion interference signals.

[0117] Furthermore, the third interference signal strength calculated for each first target signal subsequence is compared with a set third preset strength threshold. If the third interference signal strength is greater than or equal to the third preset strength threshold, the signal corresponding to the first target signal subsequence is considered to contain a strong motion interference signal. At the same time, the second target signal subsequence, because it has been determined to contain abnormal conditions such as water wave interference, is also identified as a motion interference signal. Ultimately, the signals corresponding to the first target signal subsequence and the signals corresponding to the second target signal subsequence that meet the conditions are determined to be motion interference signals in the pressure signal in the swimming mode.

[0118] In one embodiment, the third preset strength threshold is 0.6. For the first target signal subsequence R1, the calculated third interference signal strength is 0.7. Because 0.7 > 0.6, the signal corresponding to the first target signal subsequence R is determined to be a motion interference signal. The second target signal subsequence R2, due to its inherent characteristics, is directly determined to be a motion interference signal, resulting in a motion interference signal in the pressure signal under the swimming mode.

[0119] The embodiment of the present invention fully considers the periodicity and water wave interference of swimming motion, and can more comprehensively and accurately identify the motion interference signal in the pressure signal in the swimming mode, effectively avoiding the influence of the motion interference signal on the measurement results, and obtaining the blood pressure oscillation wave signal in the pressure signal in the swimming mode, effectively avoiding the influence of the motion interference signal on the measurement results, improving the accuracy and reliability of blood pressure measurement in the swimming mode, and meeting the user's demand for accurate blood pressure monitoring during exercise.

[0120] Optional, see Figure 2 , Figure 2 : is a flow chart of the blood pressure monitoring method provided by the present invention. In the embodiment of the present invention, the execution subject of the blood pressure monitoring method is a blood pressure monitoring device, and the blood pressure monitoring method includes:

[0121] Step 10: collecting pressure signals of the user in motion based on the pressure sensor, and collecting acceleration signals of the user in motion based on the acceleration sensor; the acceleration signals include motion frequency signals and motion amplitude signals.

[0122] Optionally, the blood pressure monitoring device collects pressure signals of the user in motion through a pressure sensor, and collects acceleration signals of the user in motion through an acceleration sensor, wherein the acceleration signals include motion frequency signals and motion amplitude signals.

[0123] Step 20 : performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal.

[0124] Furthermore, the blood pressure monitoring device performs pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern, and performs signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal under the current motion pattern.

[0125] Step 30: Perform signal filtering on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal.

[0126] Furthermore, the blood pressure monitoring device performs signal filtering on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal. Common filtering methods include low-pass filtering, band-pass filtering, and adaptive filtering. Low-pass filtering can remove high-frequency motion interference signals from the pressure signal while retaining low-frequency blood pressure oscillation wave signals, as the frequency of blood pressure oscillation wave signals is generally around 0.5-15Hz; band-pass filtering can only allow signals within a specific frequency range to pass based on the frequency range of the blood pressure oscillation wave signal, further enhancing the extraction effect of the blood pressure oscillation wave signal; adaptive filtering can automatically adjust the filtering parameters according to changes in the signal, more effectively removing motion interference signals in complex motion environments.

[0127] In practical applications, multiple filtering methods are often combined for processing. First, a low-pass filter is used to preliminarily filter the pressure signal to remove most high-frequency motion interference signals. The preliminarily filtered signal is then further processed using a bandpass filter to retain the signal within the frequency range of the blood pressure oscillation wave signal. Finally, an adaptive filter is used to fine-tune the signal, further optimizing the filtering effect based on real-time signal changes, resulting in a relatively pure blood pressure oscillation wave signal.

[0128] Step 40: Determine the blood pressure parameters of the user in the exercise state based on the blood pressure oscillation wave signal.

[0129] Furthermore, the blood pressure oscillation signal contains a wealth of information related to blood pressure. Determining a user's blood pressure parameters during exercise based on the blood pressure oscillation signal primarily involves calculating systolic, diastolic, and mean arterial pressure. Common calculation methods are based on the oscillometric principle, which determines blood pressure values by analyzing the relationship between the amplitude of the blood pressure oscillation wave and cuff pressure. Specifically, this method extracts features from the blood pressure oscillation signal, such as identifying characteristic parameters such as the peak and valley values of the oscillation wave and the slope of the waveform. Generally speaking, as cuff pressure decreases, the amplitude of the blood pressure oscillation wave gradually increases. When the cuff pressure equals the mean arterial pressure, the amplitude of the oscillation wave reaches its maximum. When the cuff pressure approaches the systolic pressure, the amplitude of the oscillation wave begins to increase significantly. When the cuff pressure approaches the diastolic pressure, the amplitude of the oscillation wave begins to decrease. By establishing a mathematical model between the characteristic parameters of the blood pressure oscillation wave and the blood pressure value, the characteristic parameters are used to calculate the systolic, diastolic, and mean arterial pressure values. For example, empirical formulas or regression models trained through machine learning can be used, taking the extracted characteristic parameters as input and outputting the corresponding blood pressure parameter values.

[0130] By collecting acceleration signals and accurately identifying the user's current motion mode during exercise, the embodiments of the present invention can specifically obtain motion interference signals under different motion modes. The pressure signal is then filtered based on the motion interference signals to accurately identify and remove the motion interference signals, thereby accurately filtering out the blood pressure oscillation wave signal from the pressure signal. Furthermore, the blood pressure parameters of the user during exercise are determined based on the blood pressure oscillation wave signal, effectively avoiding the influence of motion interference signals on the measurement results, improving the accuracy and reliability of blood pressure measurement during exercise, and meeting the user's need for accurate blood pressure monitoring during exercise.

[0131] See also Figure 3 , Figure 3 This is a diagram of an embodiment of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented:

[0132] The pressure sensor collects the pressure signal of the user in motion, and the acceleration sensor collects the acceleration signal of the user in motion; the acceleration signal includes a motion frequency signal and a motion amplitude signal;

[0133] Performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal;

[0134] Performing signal filtering on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal;

[0135] The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

[0136] See also Figure 4 , Figure 4 Detailed description of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, the following steps are implemented:

[0137] The pressure sensor collects the pressure signal of the user in motion, and the acceleration sensor collects the acceleration signal of the user in motion; the acceleration signal includes a motion frequency signal and a motion amplitude signal;

[0138] Performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal;

[0139] Performing signal filtering on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal;

[0140] The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

[0141] In another aspect, the present invention further provides a computer program product, which includes a computer program. The computer program may be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the blood pressure monitoring method provided by the above methods. The blood pressure monitoring method includes:

[0142] The pressure sensor collects the pressure signal of the user in motion, and the acceleration sensor collects the acceleration signal of the user in motion; the acceleration signal includes a motion frequency signal and a motion amplitude signal;

[0143] Performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal;

[0144] Performing signal filtering on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal;

[0145] The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

[0146] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0147] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A wearable blood pressure monitoring device, characterized in that: It includes a pressure sensor, an acceleration sensor and a central processing unit; the central processing unit is connected to the pressure sensor and the acceleration sensor respectively; The pressure sensor is used to: collect the pressure signal of the user in motion; The acceleration sensor is used to collect acceleration signals of the user in motion; the acceleration signals include motion frequency signals and motion amplitude signals; The CPU is used to: performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain a current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal; performing signal filtering processing on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal; The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

2. The wearable blood pressure monitoring device according to claim 1, wherein The performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain the current motion pattern includes: determining a frequency characteristic index based on the number of motion frequencies in the motion frequency signal and a ratio between the square of each motion frequency and the mean of the motion frequencies; Determining a frequency interval identifier for each frequency characteristic indicator based on the frequency interval in which each frequency characteristic indicator is located; determining an amplitude change index based on a sum of absolute values of amplitude differences at adjacent time points in the motion amplitude signal and a maximum amplitude value; Pattern recognition is performed based on the frequency interval identifier and the amplitude change index to obtain the current motion pattern.

3. The wearable blood pressure monitoring device according to claim 2, wherein: The performing pattern recognition based on the frequency interval identifier and the amplitude change index to obtain the current motion mode includes: If the frequency interval identifier is the first interval identifier and the amplitude change index is less than a first preset amplitude change threshold, determining that the current motion mode is the cycling mode; If the frequency interval identifier is the second interval identifier, and the amplitude change index is greater than or equal to the first preset amplitude change threshold and less than the second preset amplitude change threshold, determining that the current exercise mode is the running mode; If the frequency interval identifier is the third interval identifier, and the amplitude change index is greater than or equal to a second preset amplitude change threshold, determining that the current exercise mode is the swimming mode; The first interval identifier, the second interval identifier, and the third interval identifier are different identifiers.

4. The wearable blood pressure monitoring device according to claim 2, wherein: The acceleration signal also includes a running direction signal; the central processing unit is further configured to: Comparing the running direction angle difference between adjacent time points in the running direction signal with a preset direction change rate threshold; If the running direction angle difference is greater than the preset direction change rate threshold, it is determined that there is a sudden change in direction at the current time point, and the number of sudden changes is recorded to obtain the number of sudden changes in direction; If the current motion mode is the cycling mode, and the number of sudden direction changes is greater than the first upper limit of the number of sudden direction changes in the cycling mode, then the current motion mode is changed from the cycling mode to the running mode; If the current motion mode is the swimming mode, and the number of sudden direction changes is greater than a second upper limit of the number of sudden direction changes in the swimming mode, the current motion mode is corrected from the swimming mode to the running mode.

5. The wearable blood pressure monitoring device according to any one of claims 1 to 4, characterized in that: The performing signal recognition on the pressure signal based on the current motion mode to obtain a motion interference signal in the pressure signal includes: If the current motion mode is running mode, dividing the pressure signal into time windows of a preset size, and determining a first target time window and a second target time window based on a preset mean threshold and the pressure signal value of each sampling point in the sub-signal within each time window; For each first target time window, determining a local maximum and a local minimum, and determining a time interval between adjacent local maxima; For each local maximum, determining a time correlation parameter for each local maximum based on a difference between a rise time from a preceding local minimum to each local maximum and a fall time from each local maximum to a subsequent local minimum; Determining a first interference signal strength based on a mean of time intervals between adjacent local maxima, a mean of peak-to-peak values between adjacent local maxima, and a mean of a time correlation parameter of each local maximum; The sub-signal in the first target time window and the sub-signal in the second target time window whose interference signal strength is greater than or equal to a first preset strength threshold are determined as the motion interference signal.

6. The wearable blood pressure monitoring device according to any one of claims 1 to 4, characterized in that: The performing signal recognition on the pressure signal based on the current motion mode to obtain a motion interference signal in the pressure signal includes: If the current motion mode is the cycling mode, performing a fast Fourier transform on the pressure signal to obtain a frequency domain representation of the pressure signal, and determining a first target frequency interval and a second target frequency interval based on a preset energy proportion and an energy distribution of the frequency intervals in the frequency domain representation; Determining a frequency screening range based on the riding speed collected by the speed sensor, and screening the signal frequencies in the first target frequency interval based on the frequency screening range to obtain a first target signal frequency and a second target signal frequency in the first target frequency interval; For the first target signal frequency in each first target frequency interval, determine the second interference signal strength based on the average of the difference between each signal frequency peak and the center signal frequency, and the average of the difference between each signal frequency peak and the two signal frequency valleys before and after it; The signals within the first target frequency interval whose second interference signal strength is greater than or equal to a second preset strength threshold, the signals corresponding to the second target frequency interval, and the signals corresponding to the second target signal frequency are determined as the motion interference signals.

7. The wearable blood pressure monitoring device according to any one of claims 1 to 4, characterized in that: The performing signal recognition on the pressure signal based on the current motion mode to obtain a motion interference signal in the pressure signal includes: If the current motion mode is a swimming mode, determining a motion cycle based on posture information collected by a gyroscope, and splitting the pressure signal based on the motion cycle to obtain a plurality of signal subsequences; The signal subsequences are screened based on a preset period threshold and a preset amplitude threshold in combination with fluctuation period and amplitude change of the signal in each signal subsequence to obtain a first target signal subsequence and a second target signal subsequence containing water wave interference; For each first target signal subsequence, determining a third interference signal strength based on a mean of differences between a signal peak value and an amplitude value of each signal, a mean of differences between a signal valley value and an amplitude value of each signal, and a mean of amplitude standard deviations of each signal; The signal corresponding to the first target signal subsequence and the signal corresponding to the second target signal subsequence whose third interference signal strength is greater than or equal to a third preset strength threshold are determined as the motion interference signal.

8. A blood pressure monitoring method, implemented based on the wearable blood pressure monitoring device according to any one of claims 1 to 7, characterized in that: The blood pressure monitoring method comprises: The pressure sensor collects the pressure signal of the user in motion, and the acceleration sensor collects the acceleration signal of the user in motion; the acceleration signal includes a motion frequency signal and a motion amplitude signal; performing pattern recognition based on the motion frequency signal and the motion amplitude signal to obtain a current motion pattern, and performing signal recognition on the pressure signal based on the current motion pattern to obtain a motion interference signal in the pressure signal; performing signal filtering processing on the pressure signal based on the motion interference signal to obtain a blood pressure oscillation wave signal in the pressure signal; The blood pressure parameters of the user in the exercise state are determined based on the blood pressure oscillation wave signal.

9. An electronic device comprising: The memory and processor are characterized in that a computer software program is stored in the memory, and when the processor reads and executes the computer software program, the blood pressure monitoring method as claimed in claim 8 is implemented.

10. A non-transitory computer-readable storage medium, characterized in that The storage medium stores a computer software program, which, when executed by a processor, implements the blood pressure monitoring method according to claim 8.

Images