Blood pressure measurement device, and kussmaul sound signal detection method and apparatus

Abstract

The application provides a blood pressure measuring device, a Korotkoff signal detection method and device. The blood pressure measuring device comprises a processor, a pressure signal detection unit, a cuff and a pneumatic oscillation unit. The cuff is provided with an air bag, and the pneumatic oscillation unit is arranged on the air bag. The pressure signal detection unit detects the pressure signal formed by the change of the gas pressure in the air bag with time at the stage of air bag deflation during the blood pressure measurement. The pneumatic oscillation unit vibrates in response to the transient pressure change in the air bag to generate a transient high-frequency oscillation component. The processor processes the pressure signal to obtain a pulse wave signal superimposed with the transient high-frequency oscillation component, and detects the Korotkoff signal based on the pulse wave signal. The device uses the gas in the air bag as a force transmission medium, converts the transient pressure change in the air bag into a global pressure transient response in the air bag through the pneumatic oscillation unit, strengthens the pulse wave signal peak, and improves the pulse wave detection accuracy.

Description

Technical Field

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

[0002] In existing technologies, cuff-type blood pressure measurement devices typically include a cuff with an inflatable bladder and an acoustic sensor (such as a microphone or piezoelectric ceramic plate) integrated inside or below the bladder. During blood pressure measurement, when the bladder deflates, the acoustic sensor directly acquires the audio signal (Korotkoff sounds) generated by the brachial artery pulsation. After amplification and filtering, the processor identifies the appearance (first tone) and disappearance (fifth tone) of the Korotkoff sounds, corresponding to systolic and diastolic blood pressure, respectively. Because the acoustic sensor is extremely sensitive to vibrations transmitted through the air, conversations, friction noises, or mechanical vibrations from the device itself can enter the sensing path, causing interference signals to overlap with the Korotkoff sound characteristics, leading to misjudgments. To reduce misjudgments, some solutions incorporate a pressure sensor within the cuff to assist the acoustic sensor in detecting Korotkoff sound signals. However, this requires embedding additional electronic components and wiring within the standard bladder structure, which can compromise the cuff's flexibility. Furthermore, frequent inflation and deflation can easily degrade sensor performance or cause wire breakage, reducing the durability of the blood pressure measurement device.

[0003] It is evident that existing methods for detecting Korotkoff sounds still require improvement. Summary of the Invention

[0004] This application provides a method for detecting Korotkoff sound signals and a blood pressure measuring device, which can reduce the structural complexity of the blood pressure measuring device, reduce the cost of the device, and effectively improve the accuracy of Korotkoff sound signal detection.

[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-mentioned Korotkoff signal detection method.

[0006] To solve the above-mentioned technical problems, this application is implemented as follows:

[0007] In a first aspect, embodiments of this application provide a blood pressure measuring device, comprising a cuff, a pneumatic oscillation unit, a pressure signal detection unit, and a processor. An air bladder is disposed on the cuff, and the pneumatic oscillation unit is disposed on the air bladder.

[0008] The pressure signal detection unit is used to detect the pressure signal formed by the change of gas pressure in the air bladder over time during the deflation phase of the air bladder in the process of measuring blood pressure using the blood pressure measuring device. The pressure signal includes a pulse wave component caused by the pulsation of the brachial artery.

[0009] The pneumatic oscillation unit is used to vibrate in response to the pulsation of the brachial artery, and the vibration in response to the transient pressure change in the air bladder caused by the pulsation of the brachial artery generates a transient high-frequency oscillation component of the pressure signal.

[0010] The processor is used to process the pressure signal to obtain a pulse wave signal superimposed with the transient high-frequency oscillation component;

[0011] The processor is further configured to perform peak detection on the pulse wave signal, obtain the peak position of the pulse wave, and extract features from the pulse wave at the peak position to obtain the pulse wave features corresponding to each peak position.

[0012] The processor is also used to detect Korotkoff sound signals generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the wave peak positions.

[0013] Secondly, embodiments of this application provide a method for detecting Korotkoff sound signals, applied to a blood pressure measuring device. The blood pressure measuring device includes: a cuff, a pressure signal detection unit, a pneumatic oscillation unit, and a processor. An air bladder is disposed on the cuff, and the pneumatic oscillation unit is disposed on the air bladder. The method includes:

[0014] The pressure signal collected by the pressure signal detection unit is processed to obtain a pulse wave signal superimposed with a transient high-frequency oscillation component; wherein, the pressure signal is formed by the change in gas pressure inside the air bladder over time during the deflation phase of the air bladder when measuring blood pressure using the blood pressure measuring device, and the pressure signal includes a pulse wave component caused by the pulsation of the brachial artery; the transient high-frequency oscillation component is generated by the high-frequency vibration of the pneumatic oscillation unit in response to the transient pressure change inside the air bladder caused by the pulsation of the brachial artery;

[0015] Peak detection is performed on the pulse wave signal to obtain the peak position of the pulse wave;

[0016] Feature extraction is performed on the pulse waves at the peak positions to obtain the pulse wave features corresponding to each peak position;

[0017] Based on the pulse wave characteristics corresponding to each of the aforementioned peak positions, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

[0018] Thirdly, embodiments of this application provide a Korotkoff sound signal detection device, which is applied to a blood pressure measuring device. The blood pressure measuring device includes: a cuff, a pressure signal detection unit, a pneumatic oscillation unit, and a processor. An air bladder is disposed on the cuff, and the pneumatic oscillation unit is disposed on the air bladder. The device includes:

[0019] The signal acquisition module is used to process the pressure signal collected by the pressure signal detection unit to obtain a pulse wave signal superimposed with a transient high-frequency oscillation component; wherein, the pressure signal is formed by the change in gas pressure inside the air bladder over time during the deflation phase of the air bladder when measuring blood pressure using the blood pressure measuring device, and the pressure signal includes a pulse wave component caused by the pulsation of the brachial artery; the transient high-frequency oscillation component is generated by the high-frequency vibration of the pneumatic oscillation unit in response to the transient pressure change inside the air bladder caused by the pulsation of the brachial artery;

[0020] The peak location module is used to detect the peak value of the pulse wave signal and obtain the peak position of the pulse wave.

[0021] The pulse wave feature extraction module is used to extract features from the pulse waves at the peak positions to obtain the pulse wave features corresponding to each peak position.

[0022] The Korotkoff sound signal detection module is used to detect the Korotkoff sound signal generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the said peak positions.

[0023] 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.

[0024] 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.

[0025] Compared with the prior art, the embodiments of this application have the following advantages:

[0026] A blood pressure measuring device incorporates a cuff, a pressure signal detection unit, and a pneumatic oscillation unit. A single air bladder is housed within the cuff, and the pneumatic oscillation unit is mounted on the air bladder. A sealed air passage connects the pressure signal detection unit to the air bladder, forming a sealed air passage structure. During blood pressure measurement, the pressure signal collected by the pressure signal detection unit is acquired. This pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery. The pneumatic oscillation unit vibrates in response to the transient pressure changes within the air bladder caused by the brachial artery pulsation, thereby generating… The transient high-frequency oscillation component of the pressure signal is obtained; the processor processes the pressure signal to obtain a pulse wave signal, and the pulse wave signal is superimposed with the transient high-frequency oscillation component of the pressure signal; then, peak detection is performed on the pulse wave signal superimposed with the transient high-frequency oscillation component to obtain the peak position of the pulse wave, and feature extraction is performed on the pulse wave at the peak position to obtain the pulse wave feature corresponding to each peak position. The pulse wave feature includes the high-dimensional feature of the transient high-frequency oscillation component. Finally, based on the pulse wave feature corresponding to each peak position, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

[0027] Because the pneumatic oscillation unit is located within a sealed air passage structure, it can use the gas inside the airbag as a force transmission medium. When the blood vessel wall is impacted, causing local boundary perturbations within the airbag, the pneumatic oscillation unit responds to these perturbations by vibrating and converting them into a global pressure transient response within the airbag. This enhances the peak value of the pulse wave signal and improves the accuracy of pulse wave detection. Furthermore, the cuff does not require additional electronic components, resulting in a simpler structure and maintaining its flexibility, which helps improve the durability of the blood pressure measurement device. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the blood pressure measuring device disclosed in the embodiments of this application;

[0029] Figure 2 This is a schematic diagram of the pressure detection principle inside the air bladder of the blood pressure measuring device disclosed in the embodiments of this application;

[0030] Figure 3 This is a schematic diagram of a pulse wave collected by a blood pressure measurement device in the existing technology;

[0031] Figure 4 This is a schematic diagram of the pulse wave collected by the blood pressure measuring device disclosed in the embodiments of this application;

[0032] Figure 5 This is a schematic flowchart of the Korotkoff sound signal detection method disclosed in the embodiments of this application;

[0033] Figure 6This is a schematic diagram of the method for determining the position of the pulse wave peak disclosed in an embodiment of this application;

[0034] Figure 7 This is a schematic diagram of the Korotkoff sound signal detection device disclosed in the embodiments of this application;

[0035] Figure 8 A block diagram schematically illustrates an electronic device for performing the method according to this application; and

[0036] Figure 9 A storage unit for holding or carrying program code implementing the method according to this application is illustrated schematically. Detailed Implementation

[0037] 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.

[0038] To implement the Korotkoff sound signal detection method disclosed in the embodiments of this application, the embodiments of this application also disclose a blood pressure measuring device, such as... Figure 1 and Figure 2 As shown, the method is applied to a blood pressure measuring device, which includes a cuff 102, a pneumatic oscillation unit 1022, a pressure signal detection unit 1023, and a processor 104. An air bladder 1021 is disposed within the cuff 102, and the pneumatic oscillation unit 1022 is disposed on the air bladder 1021. The pressure signal detection unit 1023 is disposed at the main unit of the blood pressure measuring device, and is connected to the air bladder 1021 by a sealed air passage 1024, forming a sealed air passage structure. The pneumatic oscillation unit 1022 is disposed on the air bladder 1021, and can be located on the inner or outer surface of the air bladder 1021. The air bladder 1021 is an inflatable air bladder.

[0039] The pressure signal detection unit 1023 is used to detect the pressure signal formed by the change of gas pressure in the airbag over time during the deflation phase of the airbag 1021 during the blood pressure measurement process using the blood pressure measuring device. The pressure signal includes a pulse wave component caused by the pulsation of the brachial artery.

[0040] The pneumatic oscillation unit 1022 is used to vibrate in response to the transient pressure change in the air bladder 1021 caused by the brachial artery pulsation, thereby generating a transient high-frequency oscillation component of the pressure signal;

[0041] The processor 104 is used to process the pressure signal to obtain a pulse wave signal superimposed with the transient high-frequency oscillation component;

[0042] The processor 104 is further configured to perform peak detection on the pulse wave signal, obtain the peak position of the pulse wave, and extract features from the pulse wave at the peak position to obtain the pulse wave features corresponding to each peak position.

[0043] The processor 104 is also used to detect Korotkoff sound signals generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the wave peak positions.

[0044] For specific implementation methods of the above operations performed by the processor 104, please refer to the steps of the Korotkoff tone signal detection method described below, which will not be repeated here.

[0045] Optionally, the pressure signal detection unit 1023 is disposed on the host end of the blood pressure measuring device. Specifically, the pressure signal detection unit 1023 is disposed on the circuit board of the host end.

[0046] The way the airbag 1021 is set in the cuff 102 is the same as the way the airbag is set in the prior art electronic blood pressure monitor, and will not be repeated in the embodiments of this application.

[0047] In some optional embodiments, the pneumatic oscillation unit 1022 is partially fixed to the inner surface of the airbag 1021. The pneumatic oscillation unit 1022 is used to passively respond to the transient changes in gas pressure inside the airbag 1021 when the brachial artery pulsation causes a transient change in pressure inside the airbag 1021, generating high-frequency vibrations to amplify the effect of the brachial artery pulsation on the pressure inside the airbag 1021.

[0048] The pneumatic oscillation unit 1022 is a passive vibration structure. Optionally, the pneumatic oscillation unit 1022 can be, but is not limited to, any of the following shapes: sheet-like or small-mass block-like. The material of the pneumatic oscillation unit 1022 includes, but is not limited to, any of the following: metal or flexible composite material. For example, the pneumatic oscillation unit 1022 can be any of the following: a thin metal sheet or a thin sheet of flexible material with a certain mass. The pneumatic oscillation unit 1022 does not contain any active devices, and its specific form can be flexibly implemented according to the airbag structure and manufacturing process, which helps to reduce the implementation complexity and cost of blood pressure measurement equipment.

[0049] The following further combines Figure 2 The detection principle of the pressure signal is explained.

[0050] During blood pressure measurement using the aforementioned blood pressure measuring device, with the user wearing a cuff 102 on their upper arm, air is inflated into the cuff 1021. When the increased pressure in the cuff 1021 causes the cuff to compress and close the brachial artery, inflation of the cuff 1021 is stopped, and the air in the cuff 1021 is released, gradually decreasing the air pressure. This reduces the pressure of the cuff 102 on the user's brachial artery, allowing blood flow to break through the closed artery. The change in air pressure in the cuff 1021 over time forms a pressure signal. During the release of air from the cuff 1021, the brachial artery is in a periodic opening and closing state under the pressure of the cuff 102, and the pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery. Furthermore, the pulsation of the brachial artery causes minute volume changes or transient boundary changes within the cuff 1021 at the interface between the vessel wall and the cuff 102, resulting in transient changes in the gas volume within the cuff 1021. The minute volume changes within the airbag, the transient changes in gas volume, and the transient boundary changes are equivalent.

[0051] According to the gas law, when the gas volume inside the airbag 1021 undergoes a transient change, a transient pressure wave will be generated within the sealed space of the airbag 1021. This transient pressure response is a quasi-static pressure change caused by the overall gas volume disturbance within the airbag 1021 due to changes in the contact interface between the blood vessel wall and the cuff. When the brachial artery pulsation causes a slight change in the volume within the airbag 1021, the pneumatic oscillation unit 1022 responds to the equivalent volume disturbance within the airbag 1021 caused by the transient boundary change and the corresponding rapid change in gas pressure, manifesting as high-frequency vibration. This makes the originally small-amplitude and short-duration pressure disturbance signal more significant in terms of duration or amplitude, thereby amplifying the influence of the brachial artery pulsation on the pressure within the airbag 1021. The greater the instantaneous change in gas pressure within the airbag 1021, the greater the vibration amplitude of the pneumatic oscillation unit 1022, and the greater the amplitude of the generated transient high-frequency oscillation.

[0052] During the deflation phase of airbag 1021, the collected pressure signals include: a baseline pressure signal component characterizing the overall trend of pressure change, a pulse wave component characterizing the periodic changes of pressure, and a transient high-frequency oscillation component characterizing the transient fluctuations and morphological changes of pressure. Figure 3 The image in the middle is a schematic diagram of the pulse wave waveform decomposed from the pressure signal detected by the pressure signal detection unit 1023 when the cuff is not equipped with the pneumatic oscillation unit 1022. Figure 4 The diagram shows the pulse wave waveform decomposed from the pressure signal detected by the pressure signal detection unit 1023 when a pneumatic oscillation unit 1022 is installed in the cuff. Figure 4As shown in the waveform within the rectangular frame, after the pneumatic oscillation unit 1022 is installed in the cuff, the pressure signal is superimposed with a transient high-frequency oscillation component exhibiting transient fluctuations and morphological changes at the peak position of the pulse cycle. Furthermore, the oscillation amplitude of this transient high-frequency oscillation component varies with the pressure change within the airbag, exhibiting temporal characteristics, such as gradually decreasing as the measurement time increases. Figure 3 and Figure 4 A comparison of the pressure signal waveforms shows that by setting the pneumatic oscillation unit 1022 in the cuff, more obvious transient fluctuations and morphological changes can be superimposed on the pressure signal detected by the pressure signal detection unit 1023, thereby improving the blood pressure detection device's ability to detect Korotkoff sounds.

[0053] The pressure signal detection unit 1023 is used to detect the gas pressure signal inside the airbag 1021. Optionally, the pressure signal detection unit 1023 includes, but is not limited to, a pressure sensor. The pressure signal detection unit 1023 is connected to the distal airbag 1021 via a sealed air passage 1024. The airbag 1021, the sealed air passage 1024, and the pressure signal detection unit 1023 form a sealed air passage structure. In a sealed air passage environment, the pressure signal detection unit 1023 converts the mechanical signal inside the airbag into a continuous electrical signal. The pressure signal detection unit 1023 samples the continuous electrical signal converted from the mechanical signal at at least a preset frequency to obtain a pressure signal superimposed with transient high-frequency oscillation components generated by the periodic pulsation of the brachial artery.

[0054] The pressure signal detection unit 1023 is electrically connected to the processor 104.

[0055] After detecting that the airbag 1021 has entered the deflation stage, the processor 104 acquires the pressure signal inside the airbag 1021 collected by the pressure signal detection unit 1023, and performs Korotkoff sound signal detection based on the pressure signal.

[0056] The blood pressure measuring device disclosed in this application does not require an acoustic sensor in the cuff, resulting in a simpler structure. Since it eliminates the need to embed additional electronic components and their wiring within a standard airbag structure, it reduces the risk of damage to the cuff's flexibility and the risk of sensor performance degradation or wire breakage due to frequent inflation and deflation, thus improving the durability of the blood pressure measuring device.

[0057] The following is combined with Figure 1 The specific implementation method of the processor 104 for detecting Korotkoff sound signals based on the pressure signal is described.

[0058] The Korotkoff tone signal detection method disclosed in this application is used for, for example... Figure 1 and Figure 2 The blood pressure measuring device shown. (Refer to...)Figure 5 The method includes steps 502 to 508.

[0059] Step 502: Process the pressure signal collected by the pressure signal detection unit to obtain a pulse wave signal superimposed with transient high-frequency oscillation components.

[0060] The pressure signal is: during the deflation phase of the air bladder in the process of measuring blood pressure using the blood pressure measuring device, the pressure signal is formed by the change of gas pressure in the air bladder over time, and the pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery; the transient high-frequency oscillation component is: generated by the high-frequency vibration of the pneumatic oscillation unit in response to the transient pressure change in the air bladder caused by the pulsation of the brachial artery.

[0061] During the deflation phase of the cuff when measuring blood pressure using the aforementioned blood pressure measuring device, the device acquires a pressure signal generated by the brachial artery pulsation. This pressure signal includes: a baseline pressure signal component representing the downward trend of pressure, a pulse wave component representing the periodic changes in pressure, and a transient high-frequency oscillation component representing the transient fluctuations and morphological changes in pressure. The method for acquiring this pressure signal is described above and will not be repeated here. As mentioned above, the pressure signal detection unit 1023 acquires a pressure signal with the transient high-frequency oscillation component superimposed on the pulse wave. After decomposing the pressure signal and removing the baseline pressure signal representing the downward trend, the result is as follows: Figure 4 The pulse wave shown is superimposed with transient high-frequency oscillation components. Specific implementation methods for extracting the pulse wave superimposed with the transient high-frequency oscillation components from the pressure signal are described in existing technologies (such as high-pass filtering techniques), and will not be repeated in the embodiments of this application.

[0062] Step 504: Perform peak detection on the pulse wave signal to obtain the peak position of the pulse wave.

[0063] Optionally, the step of peak detection of the pulse wave signal to obtain the peak position of the pulse wave includes: dynamically obtaining a sampling scale based on the sampling frequency of the pulse wave; and taking a target sampling point among the sampling points obtained by sampling the pulse wave signal within the sampling scale according to a preset sampling step size as the peak position of the pulse wave, wherein the target sampling point is the sampling point where the amplitude of the pulse wave signal is the largest and the signal amplitude is greater than a preset amplitude threshold. The preset sampling step size is determined according to the sampling frequency and measurement accuracy; for example, the preset sampling step size can be set to 8 sampling points.

[0064] Optionally, before the target sampling point in the sampling points obtained by sampling the pulse wave signal, a high-pass filter with a cutoff frequency of 1 Hz can be used to perform high-pass filtering on the pressure signal to filter out low-frequency baseline drift and DC components, and a band-stop filter can be used to filter the pressure signal to suppress power frequency interference, thereby obtaining the pulse wave signal.

[0065] Optionally, the minimum sampling scale can be... Set to 1, and set the maximum sampling scale according to the sampling rate of the pulse wave. The sampling rate of the pulse wave is expressed as... For example, the maximum sampling scale can be set to 0.3 times the pulse wave sampling rate. Taking a pulse wave sampling rate of 640 as an example, the maximum sampling scale... The corresponding number of sampling points is .

[0066] During the sampling of the peak position, the pulse wave signal sequence can be traversed, targeting the current sampling point in the pulse wave. Perform the following judgment: if the sampling point amplitude Greater than the preset amplitude threshold, and the sampling scale arrive Within the range, multiple sampling points are obtained by sampling points with a step size of 8 sampling points. amplitude If the maximum value is reached, then the sampling point will be... As an optional peak position.

[0067] For specific examples, such as Figure 6 As shown, for the sampling points First, determine the sampling point. Whether the amplitude is greater than the preset amplitude threshold, at the sampling point If the amplitude is greater than the preset amplitude threshold, sampling points are taken according to the sampling scale. Sampling points Pz1 and Py1 at the left and right positions, and determine the sampling points. Is the amplitude greater than the amplitudes of sampling points Pz1 and Py1? If it is greater than or equal to, then the sampling scale is expanded, and the next sampling step is performed at the position where sampling point Pz1 is moved 8 sampling points to the left and sampling point Py1 is moved 8 sampling points to the right; if it is less than, then the current sampling point is considered... This is not the peak position; continue to determine the next sampling point in the pulse wave. If the amplitudes of the sampling point obtained by moving 8 sampling points to the left of sampling point Pz1 and the sampling point obtained by moving 8 sampling points to the right of sampling point Py1 are both less than the amplitude of the current sampling point... The amplitude is then further increased by a step size of 8 sampling points until the sampling scale reaches the maximum sampling scale, thus determining the current sampling point. The signal position is considered the peak position; otherwise, the current sampling point is considered to be... If this is not the peak position, continue to determine the next sampling point in the pulse wave.

[0068] This method detects the peak position of the pulse wave by using a skip-scale approach, which significantly reduces computational complexity and memory usage.

[0069] Step 506: Extract features from the pulse waves at the peak positions to obtain pulse wave features corresponding to each peak position.

[0070] In a pulse wave, each peak corresponds to a pulse cycle. During blood pressure measurement, the amplitude, waveform, and other characteristics of the pulse wave in each pulse cycle during the airbag deflation phase characterize the blood pressure of different users. After acquiring the user's pulse wave, the pulse wave features of each pulse cycle are first extracted for subsequent Korotkoff sound detection.

[0071] Optionally, the step of extracting features from the pulse waves at the peak positions to obtain pulse wave features corresponding to each peak position includes: determining pulse wave segments corresponding to each peak position based on the peak position; extracting time-domain feature signals from each pulse wave segment and performing feature embedding processing on the time-domain feature signals to obtain a time-domain feature vector representation corresponding to each peak position, wherein the time-domain feature vector representation contains information about the transient high-frequency oscillation component; performing a fast Fourier transform on each pulse wave segment to obtain a frequency-domain representation, and performing feature embedding processing on the frequency-domain representation to obtain a frequency-domain feature vector representation corresponding to each peak position, wherein the frequency-domain feature vector representation contains information about the transient high-frequency oscillation component; and concatenating the time-domain feature vector representation and the frequency-domain feature vector representation for each peak position to obtain the pulse wave features corresponding to the corresponding peak position. Specific implementation methods for extracting time-domain features from each pulse wave segment are found in the prior art and will not be repeated in this embodiment.

[0072] Optionally, determining the pulse wave segments corresponding to each peak position based on the peak position includes: determining sampling windows of a specified length centered on each peak position, wherein the distribution position of the transient high-frequency oscillation component is superimposed on the peak position; and taking the pulse wave in each sampling window as the pulse wave segment corresponding to the corresponding peak position. The specified length is greater than or equal to the number of pulse wave sampling points within one pulse wave cycle. Preferably, the specified length is greater than or equal to the number of pulse wave sampling points within one pulse wave cycle, and the specified length is a power of 2. Taking the sampling rate of the pulse wave by the pressure signal detection unit 1023 as 640 as an example, the specified length can be 512 sampling points, that is, selecting the peak position as the center, taking 256 sampling points to the left and 256 sampling points to the right, resulting in a sampling window including 512 sampling points. Using this method, multiple sampling windows with a length of 512 sampling points centered on each peak position can be determined. The 512 pulse wave signals in each sampling window constitute a pulse wave segment.

[0073] Next, for each pulse wave segment, the signal amplitude of each sampling point in the segment can be extracted and arranged sequentially according to the sampling time to obtain the corresponding signal amplitude sequence, which serves as the time-domain feature signal of the pulse wave segment, i.e., the time-domain feature signal corresponding to the peak position, used to characterize the local variation characteristics of the pulse wave in the time domain near the peak position. The time-domain feature signals of the N pulse wave segments constituting the pulse wave are then matrix-encapsulated according to the chronological order of the pulse wave segments, thereby transforming the discrete pulse wave waveform into a two-dimensional time-domain signal sequence with temporal characteristics.

[0074] Since the pulse wave at the peak position is superimposed with a transient high-frequency oscillation component, and the amplitude of the superimposed transient high-frequency oscillation component is different in different pulse cycles, the transient high-frequency oscillation component can enhance the temporal variation characteristics of the time domain features of each pulse wave segment, which helps to highlight the temporal differences between pulse waveforms at different times, thereby improving the accuracy of Korotkoff tone detection based on temporal modeling of time domain features.

[0075] Furthermore, a fast Fourier transform is performed on each pulse wave segment to extract the frequency domain energy distribution features of the pulse wave segment. The extracted frequency domain energy distribution features are then converted into decibel values ​​to construct the frequency domain feature signal of the pulse wave segment, thereby obtaining the frequency domain feature signal corresponding to the peak position.

[0076] Subsequently, a pre-trained convolutional neural network can be used to perform feature embedding processing on the time-domain feature signal and the frequency-domain feature signal, respectively, to obtain vector representations of the time-domain feature signal and the frequency-domain feature signal at each peak position. Specifically, the convolutional neural network automatically learns vector representations of the time-domain and frequency-domain feature signals containing transient high-frequency oscillation components based on the time-domain and frequency-domain feature signals of the pulse wave. These vector representations comprehensively reflect the local waveform structure and nonlinear relationships of the input signal in the time or frequency domain within each pulse cycle, providing a discriminative feature basis for subsequent time-series modeling and helping to improve the accuracy of Korotkoff sound detection.

[0077] Finally, the time-domain feature vector representation and the frequency-domain feature vector representation of each peak position are concatenated to obtain the pulse wave feature corresponding to the peak position. All detected pulse wave features corresponding to the peak positions are arranged in chronological order according to the occurrence time of the peak positions to obtain the pulse wave feature sequence.

[0078] With the sampling window length as Taking the detection of N peak positions from the pulse wave as an example, a value of N× can be extracted. A set of time-domain characteristic signals and a size of N× A set of frequency domain characteristic signals.

[0079] Furthermore, by splicing together a set of time-domain features and frequency-domain features corresponding to the same peak position, a set of pulse wave features corresponding to that peak position is obtained, which is also a set of pulse wave features of the pulse wave segment corresponding to that peak position.

[0080] Step 508: Based on the pulse wave characteristics corresponding to each of the stated peak positions, detect the Korotkoff sound signal generated by the brachial artery pulsation.

[0081] Next, the step of detecting the Korotkoff sound signal generated by the brachial artery pulsation based on the pulse wave features corresponding to each of the peak positions includes: inputting the pulse wave features into a preset neural network model in the form of a feature sequence according to the time sequence of the peak positions; generating a confidence sequence corresponding to the peak positions through the preset neural network model; and detecting the Korotkoff sound signal generated by the brachial artery pulsation based on the confidence sequence and a preset confidence threshold.

[0082] The confidence level indicates the probability that the corresponding peak position is a Korotkoff sound signal. A higher confidence level indicates a greater probability that the pulse wave signal at that peak position is a Korotkoff sound signal.

[0083] The value of the preset confidence threshold can be determined by statistical analysis of the Korotkoff sound signal detection results.

[0084] Optionally, the preset neural network model can be built based on a bidirectional long short-term memory network. For example, the preset neural network model includes, in sequence, a bidirectional long short-term memory network, a fully connected network, and an activation function. The bidirectional long short-term memory network is used to fuse and encode the pulse wave features at each peak position based on the contextual association in the feature vector sequence composed of time-domain and frequency-domain feature vectors, obtaining a hidden layer encoding vector; the fully connected network is used to classify and map the hidden layer encoding vector, obtaining the classification mapping result of the pulse wave features at each peak position; the activation function is used to map the classification mapping result to a preset output space, obtaining the confidence level of whether the pulse wave signal at each peak position is a Korotkoff tone signal.

[0085] The confidence value represents the degree of matching between the transient response characteristics of the airbag under the current pressure state and the effective range of Korotkoff sounds. Its value is between 0 and 1. The closer it is to 0, the lower the degree of matching between the peak position of the pulse wave (i.e., the pulse cycle) and the Korotkoff sound signal. The closer it is to 1, the higher the degree of matching.

[0086] In specific implementation, the preset neural network model can be trained using pre-constructed training data. This allows the preset neural network model to learn the characteristics and contextual relationships of Korotkoff sound signals, thereby improving its ability to recognize Korotkoff sound signals based on pulse wave feature sequences. The training data consists of pulse wave features collected from the tested user's pulse waves using the aforementioned method. The sample labels are manually labeled sequences indicating whether the peak positions in the corresponding pulse waves represent the true values ​​of Korotkoff sound signals. The training process of the preset neural network model follows the supervised training process of neural network models in existing technologies, and will not be elaborated upon in this embodiment.

[0087] The above is merely one embodiment of the preset neural network model built on the bidirectional long short-term memory network. In specific implementations, other preset neural network models with different structures can also be built on the bidirectional long short-term memory network. The specific structure of the preset neural network model built on the bidirectional long short-term memory network is not limited in this application embodiment.

[0088] The preset neural network model is built based on a bidirectional long short-term memory network. The multi-scale time-frequency features obtained by pulse wave peak segmentation during the deflation process are jointly modeled in context. The output is a confidence sequence reflecting whether the current pressure state is within the effective range of Korotkoff sounds, which can effectively improve the recognition accuracy of Korotkoff sound signals.

[0089] Subsequently, the feature sequence of the pulse wave is used as the input of the bidirectional long short-term memory network. The signal features of each peak position in the feature sequence are feature-encoded by the bidirectional long short-term memory network, and classified and mapped by the fully connected network to obtain the confidence level of the corresponding peak position as a Korotkoff tone signal.

[0090] Optionally, detecting the Korotkoff sound signal generated by the brachial artery pulsation based on the confidence sequence and a preset confidence threshold includes: determining the peak position corresponding to the first confidence level greater than or equal to the preset confidence threshold in the confidence sequence, as the signal position corresponding to the first Korotkoff sound generated by the brachial artery pulsation; and determining the peak position corresponding to the last confidence level greater than or equal to the preset confidence threshold in the confidence sequence, as the signal position of the fifth Korotkoff sound generated by the brachial artery pulsation. The pulse wave signal at the corresponding signal position is the corresponding Korotkoff sound signal.

[0091] Taking the prediction result of a pulse wave feature sequence including 10 peak positions as an example, the prediction result output by the preset neural network model can be represented as a confidence sequence in the following form: 0.0, 0.1, 0.6, 0.3, 0.8, 0.9, 0.9, 0.2, 0.0, 0.0. Taking a preset confidence threshold of 0.5 as an example, the pulse wave signal corresponding to the first peak position with a confidence level greater than or equal to the preset confidence threshold, i.e., the peak position with a confidence level of 0.6, is selected as the first Korotkoff sound (i.e., systolic blood pressure). The pulse wave signal corresponding to the last peak position with a confidence level greater than or equal to the preset confidence threshold, i.e., the peak position with a confidence level of 0.9, is selected as the fifth Korotkoff sound (i.e., diastolic blood pressure).

[0092] In some alternative embodiments, detecting the Korotkoff sound signal generated by the brachial artery pulsation based on the confidence sequence and a preset confidence threshold includes: determining the peak position corresponding to the first confidence level in the first target confidence subsequence of the confidence sequence as the signal position corresponding to the first Korotkoff sound generated by the brachial artery pulsation; and determining the peak position corresponding to the last confidence level in the last target confidence subsequence of the confidence sequence as the signal position of the fifth Korotkoff sound generated by the brachial artery pulsation, wherein the confidence level in the target confidence subsequence is: at least two consecutive confidence levels in the sequence that are greater than or equal to the preset confidence threshold.

[0093] Using the prediction results output by the preset neural network model as the confidence sequence: 0.0, 0.1, 0.6, 0.3, 0.8, 0.9, 0.9, 0.2, 0.0, 0.0, with a preset confidence threshold of 0.5 as an example, the pulse wave signal at the starting peak position (the 5th peak position) where two consecutive confidence levels (i.e., the confidence levels at the 5th and 6th peak positions) are greater than or equal to the preset confidence threshold is selected as the first Korotkoff sound (i.e., systolic blood pressure). The pulse wave signal at the ending peak position (the 7th peak position) where two consecutive confidence levels (i.e., the confidence levels at the 6th and 7th peak positions) are greater than or equal to the preset confidence threshold is selected as the fifth Korotkoff sound (i.e., diastolic blood pressure).

[0094] In some alternative embodiments, the selection of which judgment method to use based on the confidence sequence and a preset confidence threshold to detect the Korotkoff sound signal generated by the brachial artery pulsation can be determined according to the settings of the blood pressure measuring device.

[0095] In summary, the Korotkoff sound signal detection method disclosed in this application involves setting a cuff, a pressure signal detection unit, and a pneumatic oscillation unit in a blood pressure measuring device. A single air bladder is placed inside the cuff, and a pneumatic vibration unit is mounted on the air bladder. A closed air passage connects the pressure signal detection unit to the air bladder, forming a closed air passage structure with the pressure signal detection unit and the air bladder. During blood pressure measurement using the blood pressure measuring device, the pressure signal collected by the pressure signal detection unit is acquired. This pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery. The pneumatic vibration unit responds to the brachial artery pulsation. The transient pressure change within the airbag causes vibration, thereby generating a transient high-frequency oscillation component of the pressure signal. The processor processes the pressure signal to obtain a pulse wave signal, which is superimposed with the transient high-frequency oscillation component of the pressure signal. Then, peak detection is performed on the pulse wave signal superimposed with the transient high-frequency oscillation component to obtain the peak position of the pulse wave. Features are extracted from the pulse waves at the peak positions to obtain the pulse wave features corresponding to each peak position. Finally, based on the pulse wave features corresponding to each peak position, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

[0096] Because the pneumatic oscillation unit is located within a sealed air passage structure, it can use the gas inside the airbag as a force transmission medium. When the blood vessel wall is impacted, causing local boundary perturbations within the airbag, the pneumatic oscillation unit responds by vibrating and converting these perturbations into a global pressure transient response within the airbag. This enhances the peak value of the pulse wave signal and improves the accuracy of pulse wave detection. Furthermore, since the transient high-frequency oscillation component superimposed on the pulse wave varies with pressure, the pulse wave features extracted based on this superimposed transient high-frequency oscillation component enhance the trend of pulse wave feature changes over measurement time, which helps improve the accuracy of Korotkoff sound sequence detection.

[0097] Furthermore, the blood pressure measurement device implementing this design does not require an acoustic sensor within the cuff, resulting in a simpler structure. Since it eliminates the need for additional electronic components and wiring embedded in a standard cuff structure, it reduces the risk of damage to the cuff's flexibility and the risk of sensor performance degradation or wire breakage due to frequent inflation and deflation, thus improving the durability of the blood pressure measurement device.

[0098] Based on the above embodiments, this embodiment also provides a Korotkoff tone signal detection device, which is applied to, for example... Figure 1 The blood pressure measuring device shown includes: a cuff, a pressure signal detection unit, a pneumatic oscillation unit, and a processor. An air bladder is mounted on the cuff, and the pneumatic oscillation unit is mounted on the air bladder. Figure 7 As shown, the device includes:

[0099] The signal acquisition module 702 is used to process the pressure signal acquired by the pressure signal detection unit to obtain a pulse wave signal superimposed with a transient high-frequency oscillation component; wherein, the pressure signal is formed by the change in gas pressure inside the air bladder over time during the deflation phase of the air bladder when measuring blood pressure using the blood pressure measuring device, and the pressure signal includes a pulse wave component caused by the pulsation of the brachial artery; the transient high-frequency oscillation component is generated by the pneumatic oscillation unit vibrating at high frequency in response to the transient pressure change inside the air bladder caused by the pulsation of the brachial artery;

[0100] The peak location module 704 is used to perform peak detection on the pulse wave signal and obtain the peak position of the pulse wave.

[0101] The pulse wave feature extraction module 706 is used to extract features from the pulse waves at the peak positions to obtain pulse wave features corresponding to each peak position.

[0102] The Korotkoff sound signal detection module 708 is used to detect the Korotkoff sound signal generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the said peak positions.

[0103] Optionally, the Korotkoff tone signal detection module 708 is further used for:

[0104] The pulse wave features are input into a preset neural network model in the form of a feature sequence according to the time sequence of the peak positions, and the preset neural network model generates a sequence of confidence scores corresponding to the peak positions.

[0105] Based on the confidence level sequence and a preset confidence threshold, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

[0106] Optionally, detecting the Korotkoff sound signal generated by the brachial artery pulsation based on the sequence with the confidence level and a preset confidence threshold includes:

[0107] The position of the peak corresponding to the first confidence level greater than or equal to the preset confidence threshold in the confidence level sequence is determined as the signal position corresponding to the first Korotkoff sound generated by the brachial artery pulsation; and the position of the peak corresponding to the last confidence level greater than or equal to the preset confidence threshold in the confidence level sequence is determined as the signal position of the fifth Korotkoff sound generated by the brachial artery pulsation; or...

[0108] The peak position corresponding to the first confidence level in the first target confidence subsequence of the confidence level sequence is determined as the signal position corresponding to the first Korotkoff sound generated by the brachial artery pulsation; and the peak position corresponding to the last confidence level in the last target confidence subsequence of the confidence level sequence is determined as the signal position of the fifth Korotkoff sound generated by the brachial artery pulsation, wherein the confidence level in the target confidence subsequence is: at least two consecutive confidence levels in the sequence that are greater than or equal to the preset confidence threshold.

[0109] Optionally, the pulse wave feature extraction module 706 is further used for:

[0110] Based on the peak positions, determine the pulse wave segments in the pulse wave corresponding to each peak position;

[0111] Time-domain feature signals are extracted from each of the pulse wave segments, and feature embedding processing is performed on the time-domain feature signals to obtain time-domain feature vector representations corresponding to each of the wave peak positions. The time-domain feature vector representations contain information about the transient high-frequency oscillation components.

[0112] A fast Fourier transform is performed on each of the pulse wave segments to obtain a frequency domain representation, and feature embedding processing is performed on the frequency domain representation to obtain a frequency domain feature vector representation corresponding to each of the wave peak positions. The frequency domain feature vector representation contains information about the transient high-frequency oscillation component.

[0113] The time-domain feature vector representation and the frequency-domain feature vector representation of each peak position are concatenated to obtain the pulse wave feature corresponding to the peak position.

[0114] Optionally, determining the pulse wave segment corresponding to each of the peak positions based on the peak positions includes:

[0115] A sampling window of a specified length is determined with the peak position as the center, wherein the distribution position of the transient high-frequency oscillation component is superimposed on the peak position;

[0116] The pulse wave in each of the sampling windows is taken as the pulse wave segment corresponding to the corresponding peak position.

[0117] Optionally, the peak positioning module 704 is further used for:

[0118] The sampling scale is dynamically obtained based on the sampling frequency of the pulse wave;

[0119] The target sampling point among the sampling points obtained by sampling the pulse wave signal within the sampling scale according to the preset sampling step size is taken as the peak position of the pulse wave. The target sampling point is the sampling point where the amplitude of the pulse wave signal is the largest and the amplitude is greater than the preset amplitude threshold.

[0120] The pneumatic oscillation unit is disposed on the inner surface of the airbag or the outer surface of the airbag.

[0121] The Korotkoff sound signal detection device disclosed in this application is used to implement the above-mentioned Korotkoff sound signal detection 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.

[0122] In summary, the Korotkoff sound signal detection device disclosed in this application comprises a cuff, a pressure signal detection unit, and a pneumatic oscillation unit within a blood pressure measuring device. A single air bladder is disposed within the cuff, and a pneumatic vibration unit is mounted on the air bladder. A sealed air passage connects the pressure signal detection unit to the air bladder, forming a sealed air passage structure with the pressure signal detection unit and the air bladder. During blood pressure measurement using the blood pressure measuring device, the pressure signal collected by the pressure signal detection unit is acquired. This pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery. The pneumatic vibration unit responds to the pulsation of the brachial artery. The air bladder vibrates due to transient pressure changes, thereby generating a transient high-frequency oscillation component of the pressure signal. The processor processes the pressure signal to obtain a pulse wave signal, and the transient high-frequency oscillation component of the pressure signal is generated by superimposing the pulse wave. Then, peak detection is performed on the pulse wave signal with superimposed transient high-frequency oscillation component to obtain the peak position of the pulse wave, and feature extraction is performed on the pulse wave at the peak position to obtain the pulse wave feature corresponding to each peak position. Finally, based on the pulse wave feature corresponding to each peak position, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

[0123] Because the pneumatic oscillation unit is located within a sealed air passage structure, it can use the gas inside the airbag as a force transmission medium. When the blood vessel wall is impacted, causing local boundary perturbations within the airbag, the pneumatic oscillation unit responds by vibrating and converting these perturbations into a global pressure transient response within the airbag. This enhances the peak value of the pulse wave signal and improves the accuracy of pulse wave detection. Furthermore, since the transient high-frequency oscillation component superimposed on the pulse wave varies with pressure, the pulse wave features extracted based on this superimposed transient high-frequency oscillation component enhance the trend of pulse wave feature changes over measurement time, which helps improve the accuracy of Korotkoff sound sequence detection.

[0124] Furthermore, the blood pressure measurement device implementing this design does not require an acoustic sensor within the cuff, resulting in a simpler structure. Since it eliminates the need for additional electronic components and wiring embedded in a standard cuff structure, it reduces the risk of damage to the cuff's flexibility and sensor performance degradation or wire breakage caused by frequent inflation and deflation, thus improving the durability of the blood pressure measurement device.

[0125] 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.

[0126] The above provides a detailed description of a blood pressure measuring device, a Korotkoff sound signal detection method and apparatus provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method of this application and its core idea. 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 idea of ​​this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0127] 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.

[0128] 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.

[0129] For example, Figure 8An 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.

[0130] 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.

[0131] Such a computer program product can be a computer-readable storage medium, which can have the same characteristics as... Figure 8 The 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 in the method described above.

[0132] 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.

[0133] 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.

[0134] 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.

[0135] 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 blood pressure measuring device, characterized in that, include: The system includes a cuff, a pneumatic oscillation unit, a pressure signal detection unit, and a processor. An airbag is mounted on the cuff, and the pneumatic oscillation unit is mounted on the airbag. The pressure signal detection unit is used to detect the pressure signal formed by the change of gas pressure in the air bladder over time during the deflation phase of the air bladder in the process of measuring blood pressure using the blood pressure measuring device. The pressure signal includes a pulse wave component caused by the pulsation of the brachial artery. The pneumatic oscillation unit is used to vibrate in response to the transient pressure change in the air bladder caused by the pulsation of the brachial artery, thereby generating a transient high-frequency oscillation component of the pressure signal; The processor is used to process the pressure signal to obtain a pulse wave signal superimposed with the transient high-frequency oscillation component; The processor is further configured to perform peak detection on the pulse wave signal, obtain the peak position of the pulse wave, and extract features from the pulse wave at the peak position to obtain the pulse wave features corresponding to each peak position. The processor is also used to detect Korotkoff sound signals generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the wave peak positions.

2. The blood pressure measuring device according to claim 1, characterized in that, The detection of Korotkoff sound signals generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the aforementioned peak positions includes: The pulse wave features are input into a preset neural network model in the form of a feature sequence according to the time sequence of the peak positions, and the preset neural network model generates a sequence of confidence scores corresponding to the peak positions. Based on the confidence level sequence and a preset confidence threshold, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

3. The blood pressure measuring device according to claim 2, characterized in that, The detection of Korotkoff sound signals generated by the brachial artery pulsation based on the sequence with the confidence level and a preset confidence threshold includes: The position of the peak corresponding to the first confidence level greater than or equal to the preset confidence threshold in the confidence level sequence is determined as the signal position corresponding to the first Korotkoff sound generated by the brachial artery pulsation; and the position of the peak corresponding to the last confidence level greater than or equal to the preset confidence threshold in the confidence level sequence is determined as the signal position of the fifth Korotkoff sound generated by the brachial artery pulsation; or... The peak position corresponding to the first confidence level in the first target confidence subsequence of the confidence level sequence is determined as the signal position corresponding to the first Korotkoff sound generated by the brachial artery pulsation; and the peak position corresponding to the last confidence level in the last target confidence subsequence of the confidence level sequence is determined as the signal position of the fifth Korotkoff sound generated by the brachial artery pulsation, wherein the confidence level in the target confidence subsequence is: at least two consecutive confidence levels in the sequence that are greater than or equal to the preset confidence threshold.

4. The blood pressure measuring device according to claim 1, characterized in that, The step of extracting features from the pulse waves at the peak positions to obtain pulse wave features corresponding to each peak position includes: Based on the peak positions, determine the pulse wave segments in the pulse wave corresponding to each peak position; Time-domain feature signals are extracted from each of the pulse wave segments, and feature embedding processing is performed on the time-domain feature signals to obtain time-domain feature vector representations corresponding to each of the wave peak positions. The time-domain feature vector representations contain information about the transient high-frequency oscillation components. A fast Fourier transform is performed on each of the pulse wave segments to obtain a frequency domain representation, and feature embedding processing is performed on the frequency domain representation to obtain a frequency domain feature vector representation corresponding to each of the wave peak positions. The frequency domain feature vector representation contains information about the transient high-frequency oscillation component. The time-domain feature vector representation and the frequency-domain feature vector representation of each peak position are concatenated to obtain the pulse wave feature corresponding to the peak position.

5. The blood pressure measuring device according to claim 4, characterized in that, The step of determining the pulse wave segments corresponding to each of the stated peak positions based on the peak positions includes: A sampling window of a specified length is determined with the peak position as the center, wherein the distribution position of the transient high-frequency oscillation component is superimposed on the peak position; The pulse wave in each of the sampling windows is taken as the pulse wave segment corresponding to the corresponding peak position.

6. The blood pressure measuring device according to claim 1, characterized in that, The step of performing peak detection on the pulse wave signal to obtain the peak position of the pulse wave includes: The sampling scale is dynamically obtained based on the sampling frequency of the pulse wave; The target sampling point among the sampling points obtained by sampling the pulse wave signal within the sampling scale according to the preset sampling step size is taken as the peak position of the pulse wave. The target sampling point is the sampling point where the amplitude of the pulse wave signal is the largest and the amplitude is greater than the preset amplitude threshold.

7. The blood pressure measuring device according to any one of claims 1-6, characterized in that, The pneumatic oscillation unit is disposed on the inner surface of the airbag or the outer surface of the airbag.

8. A method for detecting Korotkoff tone signals, characterized in that, The method is applied to a blood pressure measuring device, which includes: a cuff, a pressure signal detection unit, a pneumatic oscillation unit, and a processor. An air bladder is disposed on the cuff, and the pneumatic oscillation unit is disposed on the air bladder. The method includes: The pressure signal collected by the pressure signal detection unit is processed to obtain a pulse wave signal superimposed with a transient high-frequency oscillation component; wherein, the pressure signal is formed by the change in gas pressure inside the air bladder over time during the deflation phase of the air bladder when measuring blood pressure using the blood pressure measuring device, and the pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery; the transient high-frequency oscillation component is generated by the high-frequency vibration of the pneumatic oscillation unit in response to the transient pressure change inside the air bladder caused by the pulsation of the brachial artery; Peak detection is performed on the pulse wave signal to obtain the peak position of the pulse wave; Feature extraction is performed on the pulse waves at the peak positions to obtain the pulse wave features corresponding to each peak position; Based on the pulse wave characteristics corresponding to each of the aforementioned peak positions, the Korotkoff sound signal generated by the brachial artery pulsation is detected.

9. A Korotkoff tone signal detection device, characterized in that, The device is applied to a blood pressure measuring equipment, which includes: a cuff, a pneumatic oscillation unit, a pressure signal detection unit, and a processor. An air bladder is disposed on the cuff, and the pneumatic oscillation unit is disposed on the air bladder. The device includes: The signal acquisition module is used to process the pressure signal collected by the pressure signal detection unit to obtain a pulse wave signal superimposed with a transient high-frequency oscillation component; wherein, the pressure signal is formed by the change in gas pressure inside the air bladder over time during the deflation phase of the air bladder when measuring blood pressure using the blood pressure measuring device, and the pressure signal includes a pulse wave component caused by the periodic pulsation of the brachial artery; the transient high-frequency oscillation component is generated by the high-frequency vibration of the pneumatic oscillation unit in response to the transient pressure change inside the air bladder caused by the pulsation of the brachial artery; The peak location module is used to detect the peak value of the pulse wave signal and obtain the peak position of the pulse wave. The pulse wave feature extraction module is used to extract features from the pulse waves at the peak positions to obtain the pulse wave features corresponding to each peak position. The Korotkoff sound signal detection module is used to detect the Korotkoff sound signal generated by the brachial artery pulsation based on the pulse wave characteristics corresponding to each of the said peak positions.

10. 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 8.

11. 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 of claim 8.

Images