Devices, methods, and apparatus for measuring blood pressure based on the pressor process

Abstract

This application provides a device, method, and apparatus for measuring blood pressure based on the inflation process. The device includes: a cuff with an inflatable bladder, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that fits against the arm, and a processor. During blood pressure measurement, the pressure signal acquisition unit acquires pressure signals within the cuff during the bladder inflation phase. During bladder inflation, the piezoelectric signal acquisition unit simultaneously acquires piezoelectric signals generated by the brachial artery pulsation. The processor processes the piezoelectric signals to obtain the signal location of valid Korotkoff sounds and, based on the signal amplitude of the pressure signal at the signal location, obtains the blood pressure measurement result. This device allows for slow and uniform inflation of the bladder during blood pressure measurement, and immediately stops inflating when the target pressure is reached, improving user comfort while accurately detecting Korotkoff sounds and enhancing the accuracy of blood pressure measurement.

Description

Technical Field

[0001] This application relates to the field of detection technology, and in particular to a device for measuring blood pressure based on a pressurization process, a method for measuring blood pressure based on a pressurization process, an apparatus for measuring blood pressure based on a pressurization process, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Korotkoff sound method is a non-invasive method for indirectly measuring blood pressure by capturing changes in the sound of arterial blood flow using a stethoscope. In existing technologies, when measuring blood pressure using cuff-type electronic blood pressure monitors, the depressurization method is typically employed. This involves first inflating the cuff to rapidly increase the pressure on the brachial artery to a target pressure (usually about 20 to 30 mmHg higher than the estimated systolic pressure) to ensure the brachial artery closes under cuff pressure. Afterward, the gas in the cuff is released, and Korotkoff sounds are collected during this process to determine the signal locations corresponding to systolic and diastolic pressures. However, in the depressurization method, excessively rapid inflatation and excessively high pressure can easily cause discomfort to the user. Therefore, the inventors began researching Korotkoff sound signal extraction during the cuff inflation process. They discovered that the Korotkoff sound signal during the inflation process suffers from weak signal strength and significant noise interference. If the Korotkoff sound signal is detected during the inflation phase for blood pressure measurement, the accuracy of the measurement is greatly compromised, and the measurement stability is limited.

[0003] It is evident that existing methods for measuring blood pressure still require improvement. Summary of the Invention

[0004] This application provides a method for measuring blood pressure based on the pressurization process and a device for measuring blood pressure based on the pressurization process, which can improve the accuracy of blood pressure measurement while ensuring the comfort of using the device during the blood pressure measurement process.

[0005] Accordingly, embodiments of this application also provide an electronic device, a storage medium, and a computer program product to ensure the implementation and application of the above-described blood pressure measurement method.

[0006] To solve the above-mentioned technical problems, this application is implemented as follows: In a first aspect, embodiments of this application provide a device for measuring blood pressure based on a pressurization process, comprising: a cuff with an inflatable air bladder, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that fits against the arm, and a processor, wherein... The pressure signal acquisition unit is used to acquire the pressure signal inside the cuff during the inflation phase of the airbag during the blood pressure measurement process using the device. The piezoelectric signal acquisition unit is used to synchronously acquire the piezoelectric signal generated by the brachial artery pulsation during the airbag inflation phase. The processor is used to process the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal; The processor is also used to obtain blood pressure measurement results based on the signal amplitude of the pressure signal at the signal location.

[0007] Secondly, embodiments of this application provide a blood pressure measurement method applied to a device for measuring blood pressure based on a pressor process. The device includes: a cuff with an inflatable bladder, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that conforms to the arm, and a processor. The method includes: During the blood pressure measurement process using the device, the pressure signal inside the cuff is acquired by the pressure signal acquisition unit, and the piezoelectric signal generated by the brachial artery pulsation is acquired synchronously by the piezoelectric signal acquisition unit. The piezoelectric signal is processed to obtain the signal position of the effective Korotkoff sound signal; The blood pressure measurement result is obtained based on the signal amplitude of the pressure signal at the signal location.

[0008] Thirdly, embodiments of this application provide a blood pressure measuring device applied to an apparatus for measuring blood pressure based on a pressor process. The device includes: a cuff with an inflatable bladder, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that conforms to the arm, and a processor. The device further includes: The signal acquisition module is used to acquire the pressure signal inside the cuff acquired by the pressure signal acquisition unit during the airbag inflation phase of blood pressure measurement using the device, and the piezoelectric signal generated by the brachial artery pulsation acquired synchronously by the piezoelectric signal acquisition unit. The signal processing module is used to process the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal; The blood pressure measurement result acquisition module is used to acquire the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal location.

[0009] Fourthly, embodiments of this application provide an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the method described in the second aspect.

[0010] Fifthly, embodiments of this application provide a computer-readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the second aspect.

[0011] Compared with the prior art, the embodiments of this application have the following advantages: By acquiring the pressure signal within the cuff collected by the pressure signal acquisition unit during the cuff inflation phase of blood pressure measurement using the device, and the piezoelectric signal generated by the brachial artery pulsation simultaneously collected by the piezoelectric signal acquisition unit; processing the piezoelectric signals to obtain the signal location of the effective Korotkoff sound signal; and finally, obtaining the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal location, this method achieves the detection of effective Korotkoff sound signals during the cuff inflation phase to determine systolic and diastolic blood pressure. This reduces the need for excessively rapid inflation and excessively high pressure to reach the target pressure, improving the comfort of device use, while accurately detecting Korotkoff sounds and improving the accuracy of blood pressure measurement. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the structure of the blood pressure measuring device disclosed in the embodiments of this application; Figure 2 This is a flowchart of the steps of the method for measuring blood pressure disclosed in the embodiments of this application; Figure 3 This is a schematic diagram of the pressure signal collected by the blood pressure measuring device disclosed in the embodiments of this application; Figure 4 This is a schematic diagram of the piezoelectric signal collected by the blood pressure measuring device disclosed in the embodiments of this application; Figure 5 yes Figure 4 The example shown is a spectrum diagram obtained after frequency domain analysis of the piezoelectric signal. Figure 6 yes Figure 5 A schematic diagram of the target frequency identified in the average spectrum of the spectrum diagram shown; Figure 7 This is a schematic diagram of the blood pressure measuring device disclosed in the embodiments of this application; Figure 8 A block diagram schematically illustrates an electronic device for performing the method according to this application; and Figure 9 A storage unit for holding or carrying program code implementing the method according to this application is illustrated schematically. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] To implement the method for measuring blood pressure based on the pressurization process disclosed in the embodiments of this application, the embodiments of this application also disclose a device for measuring blood pressure based on the pressurization process, such as... Figure 1 As shown, the method is applied to the device for measuring blood pressure based on the pressurization process. The device includes: a cuff 102 with an inflatable air bladder 1021, a pressure signal acquisition unit 104, a piezoelectric signal acquisition unit 106 disposed on the side of the cuff 102 that fits against the arm, and a processor 108. The pressure signal acquisition unit 104 is used to acquire the pressure signal within the cuff during the inflation phase of the air bladder 1021 when measuring blood pressure using the device. The piezoelectric signal acquisition unit 106 is used to simultaneously acquire the piezoelectric signal generated by the brachial artery pulsation during the inflation phase of the air bladder 1021. The processor 108 is used to perform band-stop filtering and band-pass filtering signal processing based on the spectral analysis results of the piezoelectric signal to obtain the signal position of the effective Korotkoff sound signal. The center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter are dynamically determined based on the spectral analysis results. The processor 108 is also used to obtain the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal position.

[0015] The arrangement of the inflatable airbag 1021 and the cuff 102, and the arrangement of the cuff 102 and the piezoelectric signal acquisition unit 106 are described in the prior art and will not be repeated in this embodiment.

[0016] Optionally, the pressure signal acquisition unit 104 can be disposed within the inflatable airbag 1021. In some other optional embodiments, the pressure signal acquisition unit 104 can be disposed in the main unit of the blood pressure measuring device and connected to the inflatable airbag 1021 through a sealed air passage. The pressure signal acquisition unit 104 can be a pressure sensor. Specific implementations of the pressure signal acquisition unit 104 are described in the prior art and will not be repeated in the embodiments of this application.

[0017] Optionally, the piezoelectric signal acquisition unit 106 may employ a sensor sensitive to high-frequency instantaneous vibration, including but not limited to any one or more of the following: accelerometer, piezoelectric film sensor, laser vibrometer, and miniature microphone.

[0018] During the inflation phase of the airbag 1021, under the compression of the cuff 102, the nonlinear pulse wave generated by the pulsation of the brachial artery induces shear vibrations as it propagates through the surrounding tissues. The piezoelectric signal acquisition unit 106 acquires the low-frequency shear waves and tissue vibration waves transmitted by the nonlinear vibration of the arterial wall during the pulsation of the brachial artery, and converts them into electrical signals as piezoelectric signals.

[0019] During the inflation phase of the airbag 1021, the mechanical vibration of the blood pressure measuring device itself, electrode movement, and the vibration of the pump and motor all generate high-frequency interference. This causes the piezoelectric signal acquired by the piezoelectric signal acquisition unit 106 to include interference signals. Compared to acquiring Korotkoff sound signals during the airbag deflation phase (i.e., the depressurization process), the Korotkoff sound signal acquisition during the inflation phase (i.e., the pressurization process) involves more interference factors and greater noise. Therefore, to improve both the comfort and accuracy of blood pressure measurement, the piezoelectric signal needs to be processed to obtain a Korotkoff sound signal with a high signal-to-noise ratio when acquiring Korotkoff sound signals during the pressurization process.

[0020] Optionally, the step of processing the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal includes: denoising the piezoelectric signal based on the spectral analysis results, combined with band-stop filtering and band-pass filtering, to obtain a piezoelectric signal with filtered-out device-specific disturbances, which serves as the Korotkoff tone signal; performing peak detection on the Korotkoff tone signal to obtain its peak position; segmenting the Korotkoff tone signal based on the peak position to determine a sequence of Korotkoff tone signal segments; extracting frequency domain features and time domain features from each Korotkoff tone signal segment in the sequence to obtain a sequence of frequency domain features and a sequence of time domain features; using the sequence of frequency domain features and the sequence of time domain features as input features to a pre-trained neural network model, and performing classification prediction based on the input features by the neural network model to obtain a prediction result for the Korotkoff tone signal; and determining the signal position of the effective Korotkoff tone signal based on the prediction result.

[0021] For details on the specific implementation of each step of the processor 108 in processing the piezoelectric signal, please refer to the relevant descriptions in the embodiments below, which will not be repeated here.

[0022] Optionally, the processor is further configured to analyze and process the pressure signal to obtain the pressure signal amplitude at each sampling point; the processor 108 is further configured to perform signal processing on the piezoelectric signal when the pressure signal amplitude is greater than or equal to a preset pressure threshold, to obtain the signal position of the effective Korotkoff sound signal, and to obtain the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal position.

[0023] The specific implementation method for analyzing and processing the pressure signal to obtain the pressure signal amplitude at each sampling point is described in the prior art and will not be repeated in the embodiments of this application.

[0024] Given that during the inflation phase of the airbag 1021, when the pressure inside the airbag is between 0 and 20 mmHg, it is the pre-pressurization phase of the airbag. At this time, the vibration signal generated by the brachial artery pulsation is weak, and the vibration of the pump body of the blood pressure measuring device causes large fluctuations in the vibration signal. The identifiable pulse vibration component in the piezoelectric signal is minimal or non-existent. Therefore, signal analysis and processing are only performed on the piezoelectric signals synchronously acquired when the pressure is greater than 20 mmHg. Optionally, the preset pressure threshold can be set to 20 mmHg.

[0025] The following is combined Figure 2 The flowchart of the blood pressure measurement method shown illustrates the specific implementation of each step of the processor 108 performing signal processing on the piezoelectric signal.

[0026] like Figure 2 As shown, the blood pressure measurement method includes steps 202 to 206.

[0027] Figure 2 The blood pressure measurement method shown is applied to, for example Figure 1 When the device shown measures blood pressure based on the pressurization process, the pressure signal acquisition unit 104 acquires the pressure signal inside the cuff in real time during the airbag inflation phase; the piezoelectric signal acquisition unit 106 simultaneously acquires the piezoelectric signal generated by the brachial artery pulsation during the airbag inflation phase.

[0028] Processor 108 executes steps 202 to 206, performs signal analysis and processing, and measures blood pressure based on the signal processing results. The specific implementation methods for each step executed by the processor are described below.

[0029] Step 202: Acquire the pressure signal inside the cuff acquired by the pressure signal acquisition unit during the cuff inflation phase of blood pressure measurement using the device, and the piezoelectric signal generated by the brachial artery pulsation acquired synchronously by the piezoelectric signal acquisition unit.

[0030] The methods for acquiring pressure signals and piezoelectric signals are described in the previous embodiments and will not be repeated here.

[0031] During the airbag inflation phase, the collected pressure signal waveform is as follows: Figure 3 As shown, the waveform of the acquired piezoelectric signal is as follows: Figure 4 As shown.

[0032] Step 204: Based on the spectral analysis results of the piezoelectric signal, perform band-stop filtering and band-pass filtering signal processing to obtain the signal position of the effective Korotkoff tone signal.

[0033] The center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter are dynamically determined based on the spectral analysis results.

[0034] Optionally, the step of performing band-stop and band-pass filtering signal processing based on the spectral analysis results of the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal includes: sub-steps S1 to S6.

[0035] Sub-step S1: Based on the spectral analysis results of the piezoelectric signal, the piezoelectric signal is denoised by combining band-stop filtering and band-pass filtering to obtain a piezoelectric signal with filtered out device-specific disturbances, which is then used as a Korotkoff tone signal.

[0036] The band-stop filter is used to filter out disturbance signals generated by the operation of the device itself in the piezoelectric signal, and the band-pass filter is used to filter out other high-frequency noise.

[0037] Optionally, the step of denoising the piezoelectric signal based on the spectral analysis results of the piezoelectric signal, combined with band-stop filtering and band-pass filtering, to obtain a piezoelectric signal filtered to remove the device's own operational disturbances, as a Korotkoff tone signal, includes: performing segmented frequency domain analysis on the piezoelectric signal based on a specified length of time window to obtain the spectrum of the piezoelectric signal within each time window, the spectrum being used to describe the frequency distribution and amplitude distribution of the corresponding frequency signal within the corresponding time window; averaging the spectrum of each time window along the time axis to obtain an average spectrum; performing signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum; filtering the piezoelectric signal using a band-stop filter with a preset frequency bandwidth centered at the target frequency to obtain a band-stop filtered signal; and filtering the band-stop filtered signal using a band-pass filter with a preset frequency range to obtain a Korotkoff tone signal. That is, the center frequency of the band-stop filter is dynamically determined based on the spectral analysis results.

[0038] Optionally, the step of performing segmented frequency domain analysis on the piezoelectric signal based on a specified length time window to obtain the spectrum of the piezoelectric signal within each time window includes: performing partial repeated sliding sampling on the piezoelectric signal using a specified length time window to obtain several piezoelectric signal segments; and performing frequency domain analysis on each of the piezoelectric signal segments to obtain the spectrum of the piezoelectric signal within each time window.

[0039] The frequency domain analysis includes: frequency distribution analysis and amplitude distribution analysis at a specified frequency.

[0040] Optionally, the frequency domain analysis of the piezoelectric signal can be performed by performing a Fourier transform on the piezoelectric signal to obtain its spectrum. In practice, the length N of the frequency domain analysis time window and the step size H between adjacent windows can be determined based on the sampling rate of the piezoelectric signal. For example, with a sampling rate of 640 Hz, the specified length can be set to 500 milliseconds, allowing each time window to cover 320 sampling points. The step size is smaller than the length of the time window to ensure temporal overlap between adjacent time windows, guaranteeing that the instantaneous Korotkoff tone signal is fully presented within at least one time window. For example, the step size can be set to 160 sampling points, resulting in a 50% overlap between adjacent time windows.

[0041] Then, the piezoelectric signal x[n] is segmented by sliding according to the time window with a step size H to obtain multiple piezoelectric signal segments x of length N. m [n], where m is the time window index, which is also the index of the piezoelectric signal segment.

[0042] Optionally, the following formula can be used to perform a Fourier transform on each piezoelectric signal segment to obtain the frequency domain information of the piezoelectric signal within the corresponding time window: ; Where, x m [n] represents the piezoelectric signal within the m-th time window; These are complex exponential basis functions used to extract the corresponding frequency components; Indicates the length of the time window (i.e., the number of sampling points within the time window); This represents the amplitude of the k-th frequency component within the m-th time window; This represents the frequency index; j represents the imaginary unit j. 2 =-1.

[0043] In some alternative embodiments, to reduce the boundary effects of time windows, which are also the effects of discontinuities at the boundaries of adjacent piezoelectric signal segments on the spectral results, the piezoelectric signals within each time window can be weighted before performing a Fourier transform on the piezoelectric signals to obtain weighted piezoelectric signals. For example, a preset window function w[n] (such as a Hanning window) can be applied to the piezoelectric signals for weighting. ; in, This represents the weight sequence, and N represents the number of time windows. This represents a weighted piezoelectric signal.

[0044] In some alternative embodiments, other weight sequences may also be used, such as weighting based on instantaneous amplitude or slope.

[0045] To eliminate the scaling effect of weighting the piezoelectric signal within the time window on the signal amplitude, the Fourier transform result of each time window is divided by the sum of the weight sequences to obtain the normalized piezoelectric signal, ensuring that the spectral amplitude is consistent with the original piezoelectric signal amplitude.

[0046] To enhance the recognizability of Korotkoff tone features, the frequency domain information obtained by Fourier transform, or the normalized frequency domain information, is then nonlinearly mapped to obtain the spectrum of each piezoelectric signal segment.

[0047] For example, using formula Frequency domain information of piezoelectric signals within time window m The nonlinear mapping is performed to obtain the spectrum of the piezoelectric signal within each time window.

[0048] For example, using power functions for frequency domain information Compression is performed to obtain the spectrum of the piezoelectric signal within each time window. The formula is as follows: = a , 0<α<1.

[0049] Finally, the spectra of each time window, i.e., the spectra of each piezoelectric signal segment, are arranged sequentially according to the sampling time of the piezoelectric signal to obtain the spectrum of the piezoelectric signal, for example, represented as: A=[ , , ..., ], where M is the number of effective time windows, i.e., the number of piezoelectric signal segments. , , These represent the spectra of the 1st, 2nd, and Mth piezoelectric signal segments, respectively.

[0050] In the spectrum of the same piezoelectric signal, each frequency component has a corresponding spectral amplitude, which represents the intensity of that frequency within that time window.

[0051] Using time as the x-axis and frequency as the y-axis, the spectrum of the above M piezoelectric signal segments can be obtained as follows: Figure 5 The spectrum diagram shown has time on the horizontal axis and frequency on the vertical axis, with color intensity indicating amplitude. Analysis of the spectrum reveals that frequency components that persist across multiple consecutive time windows and whose frequency positions remain largely unchanged over time exhibit stable energy distributions, representing steady-state noise introduced by the motor operation or structural vibration of the blood pressure measuring device. In contrast, frequency components that appear intermittently across different time windows, with a more dispersed frequency distribution and significant temporal locality, are Korotkoff tone-related signals.

[0052] Within a time window containing Korotkoff tones, the spectral amplitude of some frequency components is greater than that within a time window without Korotkoff tones. For example, in... Figure 5 The frequency range of 0 to 50 Hz is brighter in the 20 to 30 second time window of the spectrum diagram shown than at 15 seconds, indicating that the main frequency of the Korotkoff tone signal is 0 to 50 Hz, and the Korotkoff tone signal mainly appears at 20 to 30 seconds.

[0053] Next, based on the time-frequency characteristics of the piezoelectric signal, noise identification is performed on the piezoelectric signal using its spectrum for filtering.

[0054] First, the spectrum of each time window is averaged along the time axis to obtain the average spectrum.

[0055] Optionally, the spectrum of each time window is averaged along the time axis to obtain an average spectrum, including: averaging the amplitude of each frequency in each time window to obtain a spectrum representing the frequency and average amplitude distribution relationship of the piezoelectric signal, which is used as the average spectrum.

[0056] like Figure 5 As shown, the spectrum has dimensions (t, f, A), where t represents time, f represents frequency (component), and A represents signal amplitude. The average spectrum with dimensions (f, A) is obtained by averaging the amplitudes of each frequency within M time windows. The averaging operation can be performed by summing the amplitudes of piezoelectric signals of the same frequency collected at different sampling time points. After averaging, the long-term stable frequency components introduced by the motor operation or structural vibration of the blood pressure measuring device exhibit higher average amplitudes in the average spectrum.

[0057] Next, signal analysis processing is performed on the average spectrum to identify characteristic frequency points whose average amplitude is significantly higher than that of adjacent frequency components, and the interval between any two characteristic frequency points is not less than a preset frequency difference, which are then used as target frequencies.

[0058] Optionally, the step of performing signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum includes: for multiple adjacent specified frequency intervals in the average spectrum, obtaining the frequency corresponding to the maximum amplitude in each specified frequency interval as a candidate frequency corresponding to the respective specified frequency interval; and selecting the candidate frequencies whose interval with adjacent candidate frequencies is greater than or equal to a preset interval threshold as the target frequency. The length of the specified frequency interval can be 10 Hz, and the specified frequency interval can be divided into multiple frequency intervals based on the frequency range in the average spectrum, using the length of the specified frequency interval as a step size. For example, the frequency band between f-5 and f+5 can be considered as one frequency interval.

[0059] For a specific example, if for a certain frequency f, A(f-5) strictly increases from A(f) and strictly decreases from A(f) to A(f+5), that is, A(f-5)... <A(f-4)<…<A(f)> If A(f+2) > … > A(f+5), then A(f) can be considered a candidate frequency. The specific identification process is as follows: For a frequency f, determine whether the average spectrum A(f) of f is greater than the average spectrum of the preceding and following 5 Hz frequencies. If so, then frequency f is a possible characteristic frequency. Following this method, multiple target frequencies can be identified, such as… Figure 6 The frequency of the x-axis corresponding to the position of the midpoint. Figure 6 In the figure, the vertical axis represents the average spectrum A(f).

[0060] Optionally, the preset interval threshold is determined based on the sampling frequency of the piezoelectric signal and the noise reduction accuracy. Taking a frequency distribution range of ±5 Hz to filter out high-frequency interference generated by the mechanical vibration of the blood pressure measuring device itself, electrode movement, and pump and motor vibration as an example, i.e., the bandwidth of the subsequent band-stop filter is 10 Hz, the preset interval threshold can be set to 10 Hz to avoid interval overlap.

[0061] After determining the characteristic frequency (i.e., the target frequency), all characteristic frequencies are compared. If the frequency difference between two characteristic frequencies is less than 10 Hz, the characteristic frequency with the larger average spectrum is retained. The retained characteristic frequency is then used as the target frequency.

[0062] Next, the original piezoelectric signal is filtered by a band-stop filter with a preset frequency bandwidth centered at the target frequency to obtain a band-stop filtered signal. The preset frequency bandwidth can be set to 10 Hz. The band-stop filter can be any of the following: Butterworth band-stop filter, Chebyshev filter, or elliptic filter. For example, a Butterworth band-stop filter with a bandwidth of ±5 Hz is applied to the average spectrum near the target frequency to reduce the interference of motor and structural vibration noise from the blood pressure measuring device on the Korotkoff tone signal.

[0063] Subsequently, the piezoelectric signal after band-stop filtering is further band-pass filtered to remove low-frequency interference and high-frequency noise components caused by slow displacement and pressure changes.

[0064] Optionally, based on the energy-dense frequency band of the Korotkoff tone signal, a bandpass filter is set to perform bandpass filtering on the piezoelectric signal after band-stop filtering, thereby filtering out noise other than the Korotkoff tone signal. The low-frequency cutoff frequency of the bandpass filter can be set to 1 to 2 Hz, where the low-frequency cutoff frequency is used to remove noise caused by slow displacement in the piezoelectric signal. The high-frequency cutoff frequency can be set based on a preset quantile of the average spectral amplitude. For example, using the 60th quantile of the average spectral amplitude as a threshold, the first frequency with an amplitude below this threshold can be selected as the high-frequency cutoff frequency. That is, the high-frequency cutoff frequency of the bandpass filter is dynamically determined based on the spectral analysis results.

[0065] Dynamically determining the center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter can adapt to different motors or pumps, or environments with high background noise such as hospitals.

[0066] In the embodiments of this application, the piezoelectric signal after band-stop filtering and band-pass filtering is used as the Korotkoff tone signal. After the above band-stop filtering and band-pass filtering, the low-frequency baseline drift and high-frequency random noise in the piezoelectric signal are suppressed, and the main energy frequency band of the Korotkoff tone is preserved, thereby improving the signal-to-noise ratio of the signal and obtaining a Korotkoff tone signal with more prominent transient pulse characteristics.

[0067] Sub-step S2 involves performing peak detection on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal.

[0068] Optionally, peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal, including: sliding the Korotkoff sound signal using a scanning window of a preset length to obtain signal positions that meet preset conditions as candidate peak positions, wherein the preset conditions are: the amplitude of the preceding signal at the signal position in the scanning window increases sequentially, and the amplitude of the following signal decreases sequentially; the candidate peak positions whose amplitude of the corresponding Korotkoff sound signal is greater than a preset amplitude threshold are taken as the peak positions of the Korotkoff sound signal.

[0069] For example, the scanning window length can be set to 5 sampling points. If the piezoelectric signal x(n-5) to x(n) strictly increases and x(n) to x(n+5) strictly decreases, then n can be used as a candidate peak position. Further, candidate peak positions are filtered based on the amplitude of the piezoelectric signal at each candidate peak position, retaining those with amplitudes greater than a preset amplitude threshold. Then, the time interval between the retained candidate peak positions is judged. If the time interval between any two candidate peak positions is less than a preset time interval (e.g., 0.3 seconds), the candidate peak position with the larger amplitude is retained as the peak position of the Korotkoff sound signal. The preset time interval can be calculated based on the heart rate; the preset amplitude threshold can be determined based on the sum of the amplitudes of the collected piezoelectric signals. Sub-step S3 involves segmenting the Korotkoff signal based on the peak position to determine the sequence of Korotkoff signal segments.

[0070] The peak position of the Korotkoff sound signal is the instantaneous location where the Korotkoff sound appears. For each detected peak position, a Korotkoff sound signal sequence is collected from a predetermined number of sampling points before and after the peak position (e.g., 128 before and after the peak position), forming the Korotkoff sound signal segment corresponding to that peak position. All Korotkoff sound signal segments are arranged sequentially according to the corresponding peak positions to obtain a sequence of Korotkoff sound signal segments.

[0071] Sub-step S4 involves extracting frequency domain features and time domain features from each Korotkoff tone signal segment in the sequence to obtain a sequence of frequency domain features and a sequence of time domain features.

[0072] Next, frequency domain feature extraction and time domain feature extraction are performed on each Korotkoff signal segment in the sequence of Korotkoff signal segments to obtain the sequence of frequency domain features and the sequence of time domain features of the Korotkoff signal.

[0073] The specific implementation methods for extracting frequency domain features from the Korotkoff tone signal segment can be found in the previous embodiments for obtaining the spectral information of the piezoelectric signal segment. Existing technologies can also be used, and will not be repeated in this embodiment. The time domain features of the Korotkoff tone signal segment can be extracted based on time domain statistics (peak value, energy, etc.). This embodiment does not limit the specific implementation methods for extracting the time domain features of the Korotkoff tone signal segment.

[0074] Sub-step S5: Using the sequence of frequency domain features and the sequence of time domain features as input features of a pre-trained neural network model, the neural network model performs classification and prediction based on the input features to obtain the prediction result of the Korotkoff tone signal.

[0075] Frequency domain characteristics reflect the vibrational energy distribution of Korotkoff sound signals at different frequencies, including the dominant frequency position, frequency band energy distribution, and spectral broadening. Korotkoff sounds are instantaneous vibrational responses, and their frequency distribution is closely related to vascular elasticity, vascular tension, pressure changes, and the equivalent stiffness of human tissue. Therefore, frequency domain characteristics can characterize the physical vibrational properties of Korotkoff sounds and indirectly reflect blood pressure status.

[0076] Optionally, the neural network model includes: a first network branch, a second network branch, a splicing layer, and a fully connected layer. The first network branch and the second network branch are constructed based on a Long Short-Term Memory (LSTM) network. The first network branch is used to encode the sequence of frequency domain features to obtain a frequency domain hidden vector, and the second network branch is used to encode the sequence of time domain features to obtain a time domain hidden vector. The splicing layer is used to splice the frequency domain hidden vector and the time domain hidden vector to obtain a spliced ​​vector. The fully connected layer is used to perform feature mapping on the spliced ​​vector to obtain a prediction result for the Korotkoff sound signal. The prediction result includes the classification result of each Korotkoff sound signal segment, and the classification result is used to indicate whether the piezoelectric signal at the corresponding peak position (i.e., the Korotkoff sound signal position) is a Korotkoff sound signal.

[0077] The neural network model can be pre-trained in a supervised manner using training samples obtained through signal processing of piezoelectric signals acquired during a pre-collected blood pressure measurement process. The training sample data consists of time-domain and frequency-domain feature sequences obtained after processing the piezoelectric signals acquired during a single blood pressure measurement using the bolus method. The sample labels are the results of artificial Korotkoff sound annotation. The method for obtaining the sample data is detailed above regarding the specific method for signal processing of piezoelectric signals, and will not be repeated here.

[0078] Sub-step S6: Based on the prediction results, determine the signal position of the valid Korotkoff tone signal.

[0079] Optionally, the effective Korotkoff tone signal is the predicted continuous Korotkoff tone signal.

[0080] Taking the segmentation into L Korotkoff sound signal segments as an example, each Korotkoff sound signal segment corresponds to a peak position, and the prediction result includes a sequence of Korotkoff sound signal classification results corresponding to the L peak positions. Taking the segmentation into 7 Korotkoff sound signal segments as an example, the prediction result can be represented as: 0, 1, 1, 1, 1, 1, 1. Then, all Korotkoff sound signal segments between the second and seventh Korotkoff sound signal segments can be considered as real Korotkoff sound signals, that is, the signal positions from the second peak position to the seventh peak position are valid Korotkoff sound signal positions.

[0081] Step 206: Obtain the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal location.

[0082] Then, based on the sampling time correspondence between the pressure signal and the piezoelectric signal, the pressure values ​​corresponding to the peak positions of the second and seventh Korotkoff sounds are found. The corresponding blood pressure values ​​can then be determined based on these two pressure values, thus obtaining the blood pressure measurement result. For example, the pressure signal position corresponding to the peak position of the second Korotkoff sound is taken as the diastolic pressure signal, and the pressure signal position corresponding to the peak position of the seventh Korotkoff sound is taken as the systolic pressure signal. The blood pressure measurement result is further obtained based on the diastolic and systolic pressure signals.

[0083] In summary, the method for measuring blood pressure using a device based on the inflation process disclosed in this application acquires the pressure signal within the cuff collected by the pressure signal acquisition unit during the cuff inflation phase of blood pressure measurement using the device, and the piezoelectric signal generated by the brachial artery pulsation simultaneously acquired by the piezoelectric signal acquisition unit. The piezoelectric signal is processed to obtain the signal location of a valid Korotkoff sound. Finally, based on the signal amplitude of the pressure signal at the signal location, the blood pressure measurement result is obtained, thus achieving blood pressure measurement during the inflation phase. Specifically, during blood pressure measurement, the cuff can be inflated slowly and evenly, and when the pressure is detected to have reached or exceeded the systolic pressure in real time, inflation is immediately stopped and the air is rapidly deflated. A valid Korotkoff sound signal is detected during the inflation process before the pressure is stopped, thereby enabling blood pressure measurement. This method detects a valid Korotkoff sound signal during the cuff inflation phase to determine systolic and diastolic pressure, reducing the need for excessively rapid inflation and excessively high pressure to reach the target pressure, improving device comfort, and simultaneously improving the accuracy of blood pressure measurement by accurately detecting Korotkoff sounds.

[0084] Traditional Korotkoff sound methods for measuring blood pressure generally assume that the Korotkoff sound is produced when the cuff pressure drops below the systolic pressure, causing the brachial artery, which was previously completely blocked, to suddenly open, resulting in a sound as blood flows against the vessel wall. However, during the blood pressure increase phase, the brachial artery is not yet fully closed, and the necessary conditions for Korotkoff sound production are not met. Therefore, existing technologies lack research specifically addressing Korotkoff sounds during the blood pressure increase phase. The method for detecting Korotkoff sound signals during the blood pressure increase phase disclosed in this application, through multiple noise reduction processes on the piezoelectric signal, detects the effective Korotkoff sound signal, representing a significant breakthrough compared to existing technologies.

[0085] Based on the above embodiments, this embodiment also provides a device for measuring blood pressure based on the pressurization process, applicable to, for example... Figure 1The device shown is for measuring blood pressure. The device includes: a cuff with an inflatable air bladder, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that fits against the arm, and a processor.

[0086] like Figure 7 As shown, the device includes: The signal acquisition module 702 is used to acquire the pressure signal inside the cuff acquired by the pressure signal acquisition unit during the airbag inflation phase of blood pressure measurement using the device, and the piezoelectric signal generated by the brachial artery pulsation acquired synchronously by the piezoelectric signal acquisition unit. The signal processing module 704 is used to perform band-stop filtering and band-pass filtering signal processing based on the spectrum analysis results of the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal, wherein the center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter are dynamically determined based on the spectrum analysis results. The blood pressure measurement result acquisition module 706 is used to acquire the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal location.

[0087] Optionally, the step of processing the piezoelectric signal to obtain the signal position of the valid Korotkoff tone signal includes: Based on the spectral analysis results of the piezoelectric signal, the piezoelectric signal is denoised by combining band-stop filtering and band-pass filtering to obtain a piezoelectric signal with filtered out device-specific disturbances, which serves as the Korotkoff tone signal. Peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal; The Korotkoff signal is segmented based on the peak position to determine the sequence of Korotkoff signal segments; Frequency domain feature extraction and time domain feature extraction are performed on each Korotkoff tone signal segment in the sequence to obtain a sequence of frequency domain features and a sequence of time domain features; The sequence of frequency domain features and the sequence of time domain features are used as input features of a pre-trained neural network model. The neural network model performs classification and prediction based on the input features to obtain the prediction result of Korotkoff tone signals. Based on the prediction results, the signal positions of valid Korotkoff tone signals are determined.

[0088] Optionally, the step of denoising the piezoelectric signal based on the spectral analysis results of the piezoelectric signal, combined with band-stop filtering and band-pass filtering, to obtain a piezoelectric signal with filtered-out device-specific disturbances, as a Korotkoff tone signal, includes: The piezoelectric signal is segmented into frequency domains based on a time window of a specified length to obtain the spectrum of the piezoelectric signal within each time window. The spectrum is used to describe the frequency distribution of the piezoelectric signal and the amplitude distribution of the corresponding frequency signal within the corresponding time window. The spectrum of each time window is averaged along the time axis to obtain the average spectrum; Perform signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum; The piezoelectric signal is filtered using a band-stop filter with a preset frequency bandwidth centered at the target frequency to obtain a band-stop filtered signal. The band-stop filtered signal is filtered using a bandpass filter within a preset frequency range to obtain a Korotkoff tone signal.

[0089] Optionally, the step of performing segmented frequency domain analysis on the piezoelectric signal based on a time window of a specified length to obtain the spectrum of the piezoelectric signal within each time window includes: The piezoelectric signal is partially and repeatedly sampled using a time window of a specified length to obtain several piezoelectric signal segments. Frequency domain analysis was performed on each of the piezoelectric signal segments to obtain the spectrum of the piezoelectric signal within each time window.

[0090] Optionally, the step of performing signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum includes: For multiple adjacent specified frequency intervals in the average spectrum, the frequency corresponding to the maximum amplitude value in each specified frequency interval is obtained as the candidate frequency corresponding to the specified frequency interval. Candidate frequencies whose interval with adjacent candidate frequencies is greater than or equal to a preset interval threshold are used as target frequencies.

[0091] Optionally, peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal, including: The Korotkoff sound signal is scanned by a scanning window of a preset length to obtain the signal position that meets the preset condition as the candidate peak position. The preset condition is that the amplitude of the preceding signal at the signal position in the scanning window increases sequentially and the amplitude of the following signal decreases sequentially. The candidate peak position of the Korotkoff tone signal whose amplitude is greater than a preset amplitude threshold is taken as the peak position of the Korotkoff tone signal.

[0092] The blood pressure measuring device disclosed in this application is used to implement the above-mentioned blood pressure measuring method. For the specific implementation of each module of the device, please refer to the specific implementation of the corresponding steps in the foregoing method embodiments, which will not be repeated here.

[0093] In summary, the blood pressure measurement device based on the inflation process disclosed in this application acquires the pressure signal within the cuff collected by the pressure signal acquisition unit during the cuff inflation phase of the blood pressure measurement device, and the piezoelectric signal generated by the brachial artery pulsation simultaneously collected by the piezoelectric signal acquisition unit. The piezoelectric signal is processed to obtain the signal location of a valid Korotkoff sound. Finally, based on the signal amplitude of the pressure signal at the signal location, the blood pressure measurement result is obtained, thus achieving blood pressure measurement during the inflation phase. Specifically, during blood pressure measurement, the cuff can be inflated slowly and evenly, and when the pressure is detected to have reached or exceeded the systolic pressure in real time, the inflation is immediately stopped and the air is rapidly deflated. A valid Korotkoff sound signal is detected during the inflation process before the inflation is stopped, thereby performing blood pressure measurement. This device detects a valid Korotkoff sound signal during the cuff inflation phase to determine systolic and diastolic pressure, reducing the need for excessively rapid inflation speed and excessively high pressure to reach the target pressure, improving the comfort of device use, and simultaneously improving the accuracy of blood pressure measurement by accurately detecting Korotkoff sounds.

[0094] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus embodiments, since they are fundamentally similar to the method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0095] The above provides a detailed description of a device, method, and apparatus for measuring blood pressure based on the pressure-boosting process provided by this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and one of its core ideas. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0096] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0097] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the electronic device according to the embodiments of this application. This application can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0098] For example, Figure 8 An electronic device capable of implementing the methods according to this application is shown. The electronic device may be a PC, mobile terminal, personal digital assistant, tablet computer, etc. The electronic device conventionally includes a processor 810 and a memory 820 communicatively connected to the processor, and program code 830 stored in the memory 820 and executable on the processor 810, wherein the processor 810, when executing the program code 830, implements the methods described in the above embodiments. The memory 820 may be a computer program product or a computer-readable medium. The memory 820 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. The memory 820 has a storage space 8201 for the program code 830 of a computer program for performing any of the method steps described above. For example, the storage space 8201 for the program code 830 may include various computer programs for implementing the various steps in the above methods. The program code 830 is computer-readable code. These computer programs can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact discs (CDs), memory cards, or floppy disks. The computer program includes computer-readable code that, when executed on an electronic device, causes the electronic device to perform the methods according to the embodiments described above.

[0099] This application also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps described in this application.

[0100] Such a computer program product can be a computer-readable storage medium, which can have the same characteristics as... Figure 8The memory 820 in the illustrated electronic device is similarly arranged as storage segments, storage spaces, etc. Program code can be stored, for example, in a compressed form on the computer-readable storage medium. The computer-readable storage medium is typically as shown in the reference. Figure 9 The portable or fixed storage unit is described above. Typically, the storage unit includes computer-readable code 830', which is code read by a processor and, when executed by the processor, implements the various steps of the method described above.

[0101] The terms "an embodiment," "embodiment," or "one or more embodiments" as used herein mean that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of this application. Furthermore, please note that the examples of the phrase "in one embodiment" do not necessarily all refer to the same embodiment.

[0102] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0103] In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A device for measuring blood pressure based on the pressor process, characterized in that, include: The system includes an inflatable cuff, a pressure signal acquisition unit, a piezoelectric signal acquisition unit located on the side of the cuff that fits against the arm, and a processor. The piezoelectric signal acquisition unit is a high-frequency instantaneous vibration sensitive sensor. The pressure signal acquisition unit is used to acquire the pressure signal inside the cuff during the inflation phase of the airbag during the blood pressure measurement process using the device. The piezoelectric signal acquisition unit is used to synchronously acquire the piezoelectric signal generated by the brachial artery pulsation during the airbag inflation phase. The processor is used to perform band-stop filtering and band-pass filtering signal processing based on the spectral analysis results of the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal, wherein the center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter are dynamically determined based on the spectral analysis results. The processor is also configured to obtain blood pressure measurement results based on the signal amplitude of the pressure signal at the signal location; The step of obtaining the signal position of the effective Korotkoff tone signal by performing band-stop and band-pass filtering signal processing based on the spectral analysis results of the piezoelectric signal includes: The piezoelectric signal is segmented into frequency domains based on a time window of a specified length to obtain the spectrum of the piezoelectric signal within each time window. The spectrum is used to describe the frequency distribution of the piezoelectric signal and the amplitude distribution of the corresponding frequency signal within the corresponding time window. The spectrum of each time window is averaged along the time axis to obtain the average spectrum; Perform signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum; The piezoelectric signal is filtered using a band-stop filter with a preset frequency bandwidth centered at the target frequency to obtain a band-stop filtered signal. The band-stop filtered signal is filtered using a bandpass filter within a preset frequency range to obtain Korotkoff tone signals; Peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal; The Korotkoff signal is segmented based on the peak position to determine the sequence of Korotkoff signal segments; Frequency domain feature extraction and time domain feature extraction are performed on each Korotkoff tone signal segment in the sequence to obtain a sequence of frequency domain features and a sequence of time domain features; The sequence of frequency domain features and the sequence of time domain features are used as input features of a pre-trained neural network model. The neural network model performs classification and prediction based on the input features to obtain the prediction result of Korotkoff tone signals. Based on the prediction results, the signal positions of valid Korotkoff tone signals are determined.

2. The device according to claim 1, characterized in that, The segmented frequency domain analysis of the piezoelectric signal based on a specified time window to obtain the spectrum of the piezoelectric signal within each time window includes: The piezoelectric signal is partially and repeatedly sampled using a time window of a specified length to obtain several piezoelectric signal segments. Frequency domain analysis was performed on each of the piezoelectric signal segments to obtain the spectrum of the piezoelectric signal within each time window.

3. The device according to claim 1, characterized in that, The step of performing signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum includes: For multiple adjacent specified frequency intervals in the average spectrum, the frequency corresponding to the maximum amplitude value in each specified frequency interval is obtained as the candidate frequency corresponding to the specified frequency interval. Candidate frequencies whose interval with adjacent candidate frequencies is greater than or equal to a preset interval threshold are used as target frequencies.

4. The device according to claim 1, characterized in that, Peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal, including: The Korotkoff sound signal is scanned by a scanning window of a preset length to obtain the signal position that meets the preset condition as the candidate peak position. The preset condition is that the amplitude of the preceding signal at the signal position in the scanning window increases sequentially and the amplitude of the following signal decreases sequentially. The candidate peak position of the Korotkoff tone signal whose amplitude is greater than a preset amplitude threshold is taken as the peak position of the Korotkoff tone signal.

5. A method for measuring blood pressure based on the pressurization process, applied to a device for measuring blood pressure based on the pressurization process, characterized in that, The device includes: a cuff with an inflatable airbag, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that fits against the arm, and a processor, wherein the piezoelectric signal acquisition unit is a high-frequency transient vibration sensitive sensor, and the method includes: During the blood pressure measurement process using the device, the pressure signal inside the cuff is acquired by the pressure signal acquisition unit, and the piezoelectric signal generated by the brachial artery pulsation is acquired synchronously by the piezoelectric signal acquisition unit. Based on the spectral analysis results of the piezoelectric signal, band-stop filtering and band-pass filtering are performed to obtain the signal position of the effective Korotkoff tone signal. The center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter are dynamically determined based on the spectral analysis results. Based on the signal amplitude of the pressure signal at the signal location, the blood pressure measurement result is obtained; The step of obtaining the signal position of the effective Korotkoff tone signal by performing band-stop and band-pass filtering signal processing based on the spectral analysis results of the piezoelectric signal includes: The piezoelectric signal is segmented into frequency domains based on a time window of a specified length to obtain the spectrum of the piezoelectric signal within each time window. The spectrum is used to describe the frequency distribution of the piezoelectric signal and the amplitude distribution of the corresponding frequency signal within the corresponding time window. The spectrum of each time window is averaged along the time axis to obtain the average spectrum; Perform signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum; The piezoelectric signal is filtered using a band-stop filter with a preset frequency bandwidth centered at the target frequency to obtain a band-stop filtered signal. The band-stop filtered signal is filtered using a bandpass filter within a preset frequency range to obtain Korotkoff tone signals; Peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal; The Korotkoff signal is segmented based on the peak position to determine the sequence of Korotkoff signal segments; Frequency domain feature extraction and time domain feature extraction are performed on each Korotkoff tone signal segment in the sequence to obtain a sequence of frequency domain features and a sequence of time domain features; The sequence of frequency domain features and the sequence of time domain features are used as input features of a pre-trained neural network model. The neural network model performs classification and prediction based on the input features to obtain the prediction result of Korotkoff tone signals. Based on the prediction results, the signal positions of valid Korotkoff tone signals are determined.

6. A device for measuring blood pressure based on the pressurization process, characterized in that, The device is applied to a device for measuring blood pressure based on a pressor process. The device includes: a cuff with an inflatable air bladder, a pressure signal acquisition unit, a piezoelectric signal acquisition unit disposed on the side of the cuff that fits against the arm, and a processor. The piezoelectric signal acquisition unit is a high-frequency transient vibration-sensitive sensor. The device includes: The signal acquisition module is used to acquire the pressure signal inside the cuff acquired by the pressure signal acquisition unit during the airbag inflation phase of blood pressure measurement using the device, and the piezoelectric signal generated by the brachial artery pulsation acquired synchronously by the piezoelectric signal acquisition unit. The signal processing module is used to perform band-stop filtering and band-pass filtering signal processing based on the spectrum analysis results of the piezoelectric signal to obtain the signal position of the effective Korotkoff tone signal, wherein the center frequency of the band-stop filter and the high-frequency cutoff frequency of the band-pass filter are dynamically determined based on the spectrum analysis results. The blood pressure measurement result acquisition module is used to acquire the blood pressure measurement result based on the signal amplitude of the pressure signal at the signal location; The step of obtaining the signal position of the effective Korotkoff tone signal by performing band-stop and band-pass filtering signal processing based on the spectral analysis results of the piezoelectric signal includes: The piezoelectric signal is segmented into frequency domains based on a time window of a specified length to obtain the spectrum of the piezoelectric signal within each time window. The spectrum is used to describe the frequency distribution of the piezoelectric signal and the amplitude distribution of the corresponding frequency signal within the corresponding time window. The spectrum of each time window is averaged along the time axis to obtain the average spectrum; Perform signal analysis processing on the average spectrum to obtain the target frequency in the average spectrum; The piezoelectric signal is filtered using a band-stop filter with a preset frequency bandwidth centered at the target frequency to obtain a band-stop filtered signal. The band-stop filtered signal is filtered using a bandpass filter within a preset frequency range to obtain Korotkoff tone signals; Peak detection is performed on the Korotkoff sound signal to obtain the peak position of the Korotkoff sound signal; The Korotkoff signal is segmented based on the peak position to determine the sequence of Korotkoff signal segments; Frequency domain feature extraction and time domain feature extraction are performed on each Korotkoff tone signal segment in the sequence to obtain a sequence of frequency domain features and a sequence of time domain features; The sequence of frequency domain features and the sequence of time domain features are used as input features of a pre-trained neural network model. The neural network model performs classification and prediction based on the input features to obtain the prediction result of Korotkoff tone signals. Based on the prediction results, the signal positions of valid Korotkoff tone signals are determined.

7. An electronic device, comprising a memory, a processor, and program code stored in the memory and executable on the processor, characterized in that, When the processor executes the program code, it implements the method of claim 5.

8. A computer-readable storage medium having program code stored thereon, characterized in that, When the program code is executed by the processor, it implements the steps of the method described in claim 5.

Images